-------
              Emission Factors

              Emissions data based on boiler tests conducted over the past several years were
obtained.  Tables 4-11 and 4-12 contain draft CDD/CDF emission factors for coal-fired units
and oil-fired units, respectively.38  Detectable levels of CDD/CDF from gas-fired boilers were
not identified.  It is important to note that these data are preliminary and have not been
approved and  finalized by the EPA.  In addition, the emission factors are for a composite of
various furnace configurations and control devices.  Thus, no SCCs or ratings were assigned
to these data.

              Source Locations

              There are approximately 700 known utility boilers located throughout the
United States.  Because of this large number of coal-fired sources, providing site-specific
locations in this report is not practical.

              Information  on precise utility plant locations can be obtained by contacting
utility trade associations such as the Electric Power Research Institute in  Palo Alto, California
(415-855-2000); the Edison Electric Institute in Washington, D.C. (202-828-7400); or the
U.S. Department of Energy (DOE) in Washington, D.C.  Publications by EPA and the DOE
on the utility industry also  would be useful in determining specific facility locations, sizes,
and fuel use.

4.2.2          Industrial Sector

              Industrial boilers are widely used in manufacturing, processing, mining,  and
refining, primarily to generate process steam and provide space heating.  Some boilers are
also used for electricity generation. Industrial boilers can fire fossil and non-fossil fuels.
Wood is the only non-fossil fuel discussed here, since wood-fired industrial boilers are more
likely sources  of CDD/CDF emissions due to the presence of CDD/CDF  precursors in wood.
                                          4-58

-------
       TABLE 4-11.  DRAFT SUMMARY OF CDD/CDF EMISSIONS FROM
                 COMPOSITE COAL-FIRED UTILITY BOILERS&

Isomer
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
Total HpCDD
Total HxCDD
Total OCDD
Total PeCDD
Total TCDD
Total CDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7, 8-PeCDF
1,2,3,4,7, 8-HxCDF
1, 2,3,6,7, 8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
Total HxCDF
Total OCDF
Total PeCDF
Total TCDF
Total CDF
Median
Ib/trillion Btu
1.6xl(r6
4.3xlO'6
9.7X10'6
5.8xlO'6
7.3xlO-6
5.7xlO-6
l.lxlO'4
2.4x1 0-5
5.8xlO-5
9.8xlO-6
7.1x10-"
2.1xlO'4
3.9xlO-6
2.4xlO-6
l.OxlO'5
1.3xlO'5
4.0x10-°
8.5xlO-6
1.6xlO'5
2.0x1 0'5
1.7xlO-4
2.4xlO'5
1.9xlO-5
1.7xlO"s
l.SxlO'5
1.2xlO'5
9.0x1 0-5
Emission Factor
g/MJ
6.9xlO'13
l.SxlO-'2
4.2xlQ-12
2.5xlO'12
3.1xlO'12
2.5xlO'12
4.7x10-"
1.0x10-"
2.5x10-"
4.2xlO-12
3.1xlO'12
8.9x10-"
1.7xlO'12
l.OxlO-12
4.3xlO'12
5.6xl012
1.7xlO-'2
3.7xlO-12
6.9xlO'12
8.6xlO-'2
7.3x10-"
1.0x10-"
8.2xlO-12
7.3xlO'12
7.7xlO'12
5.2xlO-12
3.8x10-"
Source:  Reference 38.

a The emission factors presented here represent a composite of various combustor configurations and
  control devices and are considered draft values, pending EPA approval.
                                     4-59

-------
       TABLE 4-12.  DRAFT SUMMARY OF CDD/CDF EMISSIONS FROM
                  COMPOSITE OIL-FIRED UTILITY BOILERS3

Isomer
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
Total HpCDD
Total HxCDD
Total OCDD
Total PeCDD
Total TCDD
Total CDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
Total HxCDF
Total OCDF
Total PeCDF
Total TCDF
Total CDF
Median
Ib/trillion Btu
6.5xlO'6
5.8xlO'6
1.2xlO-5
5.4xlO-6
8.3xlO'6
2.0xlO-5
2.0x1 0-5
S.lxlO'6
2.3xlO-5
5.8xlO-6
5.7xl(T6
6.3xlO'5
4.6x1 0-6
4.3xlO-6
4.8xlO-6
6.1xlO-6
3.8xlQ-6
5.8xlO-6
4.8x10-°
9.4x10-°
l.OxlO-5
l.SxlQ-6
9.6x1 0'6
l.OxlO'5
7.3xlQ-6
5.0x10-"
3.3xlQ-5
Emission Factor
g/MJ
2.8xlO-12
2.5xlO-12
5.2xlO-12
2.3xlO-'2
3.6xlO-12
8.6xlO-12
8.6xlO-12
3.5xlO-'2
9.9xlO'12
2.5xlO-'2
2.5xlO-12
2.7x10-"
2.0xlQ-12
l.SxlO'12
2.1xlO-12
2.6xlO-12
1.6xlQ-12
2.5xlO'12
2.1xlO'12
4.0xlO'12
4.3xlO-12
6.5xlO-13
4.1X10'12
4.3xlO-12
3.1xlO-12
2.2xlO-12
1.4x10'"
Source:  Reference 38.

a The emission factors presented here represent a composite of various combustor configurations and
  control devices and are considered draft values, pending EPA approval.
                                      4-60

-------
              Process Description

              Industrial boilers burning fossil fuels are identified by their heat transfer
method. These include watertube, firetube, and cast iron.  Watertube boilers are designed to
.pass water through the inside of heat transfer tubes while the outside of the tubes is heated by
direct contact with the hot combustion gases. Industrial watertube boilers can burn coal, oil,
or natural gas.  Coal-fired industrial boilers are generally of the watertube design.  Firing
mechanisms include pulverized coal and stoker (spreader, underfeed, and overfeed stoker).
The most common of these are pulverized coal boilers, especially for larger  coal-fired boilers.
In firetube boilers, the hot gas flows through the tubes and the water being heated circulates
outside of the tubes.  Most installed firetube boilers burn oil or gas.  In cast iron boilers, the
hot gas is contained inside the tubes and the water being heated circulates outside the tubes.

              The burning of wood waste in boilers is mostly confined to those industries
where  it is available as a byproduct.  Currently, the bulk of wood residue or bark burning in
industrial boilers is carried out in  forest products industrial boilers.39 It is burned both to
obtain  heat  energy and to alleviate solid waste  disposal problems.  The bulk  of wood
combusted is from debarking of logs or byproducts from wood products operations where the
original wood is not tainted with inorganic chlorides such  as would be  the case with  logs
stored  or transported over sea water.  CDD/CDF emissions from facilities burning  salt-laden
wood residue may be considerably higher than from those burning salt-free wood residues,
and are not considered here.   Wood waste may include large pieces, such as slabs, logs, and
bark strips,  as well as cuttings, shavings, pellets,  and sawdust.40

              Various boiler firing configurations are used in burning wood waste. One
common type in smaller operations is the dutch oven or extension type of furnace  with a flat
grate (see Figures 4-12 and 4-13).  This unit is widely used because it  can burn fuels with
very high moisture. Fuel is  fed into the oven through apertures in a firebox and is fired  in a
cone-shaped pile on a flat grate.  The burning is  done in two stages: (1) drying and
gasification, and (2) combustion of gaseous products.  The first stage takes place in a ceil
                                           4-61

-------
 To Cirvder
Collectors,
Air Heater,
 and Stack
Auxiliary Fuel
   Burner
  (if used)
                                                                                                       Underfire
                                                                                                        Air In
                              Figure 4-12.  One-Cell Dutch Oven-Type Boiler
      Source: Reference 40.
                                                     4-62

-------
     Wood Fuel
  Combustion Air
                                                                                    To Stack
                                                                          Bag house Outlet
                                     Boiler Exhaust Gas 80%            Baghouse Inlet
                                                             Multicyclone
                                                      20%
Multicyclone Ash
Reinjected in Boiler
                              Wood Fired Boiler
                                                           ->• Bottom Ash
                                                                                   Baghouse Dust
                  Figure 4-13. Schematic Process Flow for Dutch Oven Boiler
Source: Reference 40.
                                             4-63

-------
separated from the boiler section by a bridge wall. The combustion stage takes place in the
main boiler section.39'40

              In another type of boiler, the fuel-cell oven, fuel is dropped onto suspended
fixed grates and is fired in a pile. The fuel cell further uses combustion air preheating and
repositioning of the secondary and tertiary air injection ports to improve boiler efficiency.40

              In many large operations, more conventional boilers have been modified to
burn wood waste. These units may  include spreader stokers with traveling grates or vibrating
grate stokers, as well as tangentially fired or cyclone-fired boilers. The most widely used of
these configurations is the spreader stoker, which can burn dry or wet wood.  Fuel is dropped
in front of an air jet,  which casts the fuel over a moving grate.  The burning is done in three
stages:  (1) drying; (2) distillation and burning of volatile matter; and  (3) burning of residual
carbon.39

              Sander dust is often burned in various boiler types at plywood, particle board,
and furniture plants.  Sander dust contains fine wood particles with low moisture  content (less
than 20 percent by weight).  In some boilers, it is fired through a flaming horizontal torch,
usually with natural gas  as an ignition aid or supplementary fuel.40

              A recent development in wood-firing is FBC boilers. Refer to Section 4.2.1
Utility Sector for a description of this boiler-type.  Because of the large thermal mass
represented by  the hot inert bed particles, FBCs can handle fuels with high moisture content
(up to 70 percent, total basis).  Fluidized  beds can also handle  dirty fuels (up to 30 percent
inert material).  Wood is pyrolized faster  in a fluidized bed than  on a  grate due to its
immediate contact with hot  bed material.  As a result,  combustion is rapid and results in
nearly complete combustion of organic matter, thereby minimizing emission of unburned
organic compounds.40

              Emissions of CDD/CDF from wood-fired boilers are dependent on  several
variables: (1) wood waste composition and variability; (2) fossil fuel  type and quantity, if
                                          4-64

-------
any, co-fired with the wood waste; (3) combustor type and performance; and (4) air pollution
control systems.

              The composition of wood waste has an impact on CDD/CDF emissions.  The
composition of wood waste depends largely on the industry from which it originates. Pulping
operations, for example, produce great quantities of bark that may contain more than
70 percent by weight moisture, along with sand and other  noncombustibles. Because of this,
bark boilers in pulp mills may emit considerable amounts of organic compounds to the
atmosphere unless they are well controlled. On the other hand, some operations, such as
furniture manufacturing, produce a clean, dry wood waste, 5 to 50 percent by weight
moisture, with relatively low particulate emissions when properly burned.  Still other
operations, such as sawmills, burn a varying mixture of bark and viood waste that results in
particulate  emissions somewhere between those of pulp mills  and furniture manufacturing.
Additionally, when fossil fuels are co-fired with wood waste, the combustion efficiency is
typically improved; therefore, organic emissions may decrease.

              Combustor performance, especially the ability to provide ample air and fuel
mixing and to maintain adequate temperatures for hydrocarbon destruction, are critical to
minimizing emissions of CDD/CDF and precursor compounds.  Key combustor design and
operating parameters are ample time and temperature for drying high moisture content
materials, and adequate supply and proper placement of undergrate and overfire combustion
air.  If the  requirements are satisfied, the potential for emission of CDD/CDF  is significantly
reduced.

              Emission Control Techniques

              Emissions controls for fossil  fuel-fired industrial boilers are similar to those
previously  described for coal-fired utility boilers.  Traditionally, control devices on wood-fired
boilers were intended primarily for particulate control.  Mechanical collectors  such as
multicyclones were most often used, especially on stoker-fired industrial boilers, to capture
large, partially burned material and reinject it to the boiler. These devices, however, do not
                                         4-65

-------
meet the current New Source Performance Standard for wood-fired boilers. Thus, for
wood-fired utility boilers, these multicyclones are often used in conjunction with secondary
dust collectors, including ESPs, wet scrubbers, or fabric  filters.39'40  Scrubbers are most
commonly venturi, although, impingement wet scrubbers are also used.

              On the West Coast, fabric filters are primarily used to collect chloride fumes in
boilers combusting salt-laden bark.39  Some gravel-bed filters have been used in lieu of fabric
filters to eliminate fabric fire hazards.  Only limited numbers of wood-fired boilers use ESPs
for control because they are less effective at collecting high-carbon ash.  Some boilers may
require additional controls for nitrogen oxides and acid gas. Acid gas removal techniques,
such as  using limestone scrubbers, are often needed for boilers burning demolition debris and
other chloride-containing fuels.39

              Several recent studies indicate that  CDD/CDF emissions may increase across
fabric filters, wet scrubbers, and ESPs.  This may be caused by low-temperature de novo
synthesis, or a transformation reaction of CDD/CDF in the air pollution control equipment.

              Emission Factors

              The U.S. EPA reports emission factors  (in AP-42) for wood waste combustion
for total CDD/CDF developed from several test reports.  One test was conducted on a
Wellons Quad Cell wood-fired boiler used for generating electricity.  This boiler was tested
under normal steady-state operating conditions of 60,000 Ib (27,216 kg) of steam per hour and
5 MW electricity.  The fuel consisted of coarse wood  waste and coarse sawdust from non-
industrial  logging operations. The exhaust gas stream from the boiler passed through a
multicyclone before entering the stack where CDD/CDF was sampled.  In another study,
speciated  CDD/CDF data were reported from a test conducted on a wood-fired boiler exhaust
in 1992.  The boiler, firing wood/bark waste, was equipped with a scrubber and produced less
than 50,000 Ib (22,680 kg) steam. These data are presented in Table 4-13.40 Because the data
are from only two tests and one  of the boilers tested was a utility boiler, it is not possible to
conclude that the emission factors presented in Table 4-13  are representative of the category.
                                          4-66

-------
              TABLE 4-13.  TOTAL CDD/CDF EMISSION FACTORS FROM
                             WOOD WASTE COMBUSTION
             SCC 1-01-009-01/02/03, 1-02-009-01 through -07, 1-03-009-01/02/03
                   (Utility, Industrial, and Commercial/Institutional Boilers)
Average Emission Factor
Isomer
Total CDD
Total CDF
2,3,7,8-TCDD
Ib/tOn wood waste
burned
1.2xlO'8
2.9xlO'8
3.6x10-"
kg/Mg wood Factor Quality
waste burned Rating
6.0x1 0-9
l.SxlO'8
1.8x10-"
C
c
D
     Source: Reference 40.

             The National Council of the Paper Industry for Air and Stream Improvement
Inc.  (NCASI), recently summarized CDD/CDF levels measured from four industrial sources
burning wood residue  and/or bark.41  All TEFs reported are based on the I-TEF/89 scheme,
which has been adopted by EPA for assessing the risks associated with CDD/CDF exposure.

             The four wood-fired boilers tested ranged in size from 30xl03  to 209x1O3 Ib/hr
(1.4xl04 to 9.5xl04 kg/hr)  steam production capacity. One boiler was operated at a steady
rate of 60x1O3 Ib (2.7xl04 kg) steam/hr during the tests and burned wood waste and sawdust
in a  Wellons  Quad Cell.  This boiler is the same unit as described above and included in the
emission factors presented  in AP-42 for total CDD/CDF from wood residue combustion.  The
exhaust gases passed through a multicyclone before exiting through the stack.  The second
boiler tested was also a Wellons Quad Cell operating between 30xl03 to 60xl03 Ib (1.4xl04 to
2.7xl04 kg) stream/hr and burning wood chips and bark.  The flue gases passed through a
multicyclone  and an ESP before entering the stack.

             The third boiler tested was a fluidized bed combustor.  It operated at
209xl03 Ib (9.5xl04 kg) steam/hr and burned wood  and agricultural waste.  Air pollution
control techniques on this unit were vaporized ammonia injection for NOX control, a

                                        4-67

-------
multicyclone, and an ESP.  The fourth boiler tested had two parallel spreader stokers.  At test
conditions, each boiler burned wood waste and produced HOxlO3 Ib (S.OxlO4 kg) steam/hr.
The exhaust gases from each boiler passed through its own dedicated ESP, and then through a
common stack.41

             NCASI also reported data from another study that tested five pulp and paper
industry boilers firing bark, wood residue, or a combination of these two fuels.  During
testing, the boilers operated between 320xl03 to 600xl03 Ib (1.4xl05 to 2.7xl05 kg) steam/hr.

             Table 4-14 provides a summary of the average CDD/CDF emissions (in TEQ
units) from the  nine boilers tested and described above.  It should be noted that accurate
estimates of the amount of bark or wood residue fired were not measured during the tests.
However, NCASI used an average F factor and an average heat content value for wood
combustion to convert concentrations to emission factors.  An F  factor is the ratio of the gas
volume of the products of combustion to the heat content of the  fuel.  Using an F factor of
1,850 standard cubic feet (scf) CO2/MMBtu and a heat value of 9,000 Btu/lb dry wood
residue (or bark), an average industrial wood combustion emission factor of 1.2xlO"3 fag
TEQ/kg dry fuel was obtained.41  The typical moisture content of bark/wood residue is about
50 percent.  Thus, an average emission factor on an as-fired basis of 6.2xlO"4 |j,g TEQ/kg as-
fired wood residue is obtained.41

             Bleached Kraft Mill Sludge Burning in Wood-Fired Boilers

             Primary and secondary sludges from pulp mills are increasingly dewatered and
burned in industrial  bark boilers.41  The sludge  from mills with bleaching operations can
contain fairly significant levels of chloride.  The concern exists that CDD/CDF emissions will
increase significantly from the addition of bleached kraft mill (BKM) sludge to the bark or
wood residue fuel.  NCASI reports emissions data from three tests of a spreader stoker boiler
equipped with an ESP, burning BKM sludge with  the wood residue.  Results from a second
test are also reported where bark and coal were burned in a spreader stoker with small
amounts of BKM sludge. The results of these two tests suggest that the burning of
                                         4-68

-------







D
0
O
^

,_j
^-4
<<-
^<«
2
H
00
*— )
O
Z

2
O
c^
M-H
U.
00
Z r.
Ooo
~ oi
00 W
00 J
11
u, Q
Q PJ
0 &
~2? *— '
o ^r
8^
w D
J Q
< £
jr a
O oL
t™~
u.
0
>•
Od
<•
"^
^
i^2
W H

^—l
D
oo

Tf
1
TABLE 4



c
*5
*£

X
ra
2

CJ
oo
n
bi
OJ
>
<

^
•o
^
34;
u.
CQ
U

•o
W>
%
CQ
U-
U

DO
1
CQ
U<
£

GO
.0
w
03
U.
£


Dl;
JZ

^•

o
u
CO
u.
>
^

-c
CO
CQ
H.
fe
f^
£.
<
CQ
U-
^

Q
'S
D
b o
"x ><
OO "^**
fsj 3,
•* ^~^
b 2
"x x
00 °°
— • d
^ J^~^
"^ 2
^
00 ^

^-v

2 2
"x *
3 -0
Os C-

*> rC^
b 2
">< i
•* °9
 ~
-^ fN

c>
u 'p-
^^ 5
^ " •£
w ^ 5
H ^ op
[i — •£
- -9^
b o
"x ^
oo o\
<^i C-
o^ ^
b b
x x
•-,
U
t3
XJ
~
3
£
0
O
O^
oC
a
CQ

«i • §
c " s
O c/^ ^
1 ^ P E
1 -s * • !i &
> — >- a. 3 c.
™ 3 Jo w ^ • • "
_ E ~ w = ^^ O
!s,!^s!£
ojo^urau-gu
« '>.£-'> T: "5 u
5oj-cojc5u<-:'
^-oo-aS^h10
X OO
•o-Tr-OT^Ta-aC1- -
oSo£oo0-
>C|C||OM
S-O?O:S:S*~".S
II ^ II ^ II II II £
" 2"cij S'oj'^'oj ^
3 .5 3 .5 3 3 3 ^
U-U-U,U-U.U.U-<
cox> UTS i>ti- oox:
4-69

-------
BKM sludge has little impact on CDD/CDF emissions from wood combustion.41  Table 4-15
presents the average emissions data from these tests.
 TABLE 4-15.  TOTAL CDD/CDF EMISSIONS FROM A WOOD-FIRED BOILER WHILE
                    BURNING BLEACHED KRAFT MILL SLUDGE3
                                            Average Emissions at 12% CO-,
                                        Ib/dscf                     ng/dscm
 I-TEF TEQ                           8.7xlO-18                      1.4X10"4
 I-TEF TEQb                           1.3xlQ-'5
Source:  Reference 41.
a Spreader stoker equipped with an ESP; Test date, November 1993.
b Fired with coal, wood residue, and BKM sludge.
             Source Locations

             Most of the coal-fired industrial boiler sources are located in the Midwest,
Appalachian, and Southeast  regions.  Industrial wood-fired boilers tend to be located almost
exclusively at pulp and paper, lumber products, and  furniture industry facilities.  These
industries are concentrated in the Southeast, Gulf Coast, Appalachian, and Pacific Northwest
regions. The Pacific Northwest contains many of the boilers firing salt-laden wood bark.  As
of 1980, there were approximately 1,600 wood-fired boilers  operating in the United States,
with a total capacity of over 30,000 megawatts (MW).40

             Trade associations such as the American Boiler Manufacturers Association in
Arlington, Virginia (703-522-7350) and the Council  of Industrial Boiler Owners in Fairfax
Station, Virginia (703-250-9042) can provide  information on industrial boiler locations and
trends.
                                         4-70

-------
4.2.3         Residential Sector

              The residential sector includes furnaces, stoves, and fireplaces burning coal, oil,
gas, and wood to produce heat for individual homes. Residential coal-fired furnaces are
usually underfeed hand-stoked units, while oil-fired furnaces are designed with
-varying burner configurations. Emission factors are presented in this section for both of these
sources. Gas-fired furnaces, which are unlikely sources of CDD/CDF,  are not included here.

              Residential wood combustion devices include furnaces, fireplaces, and
woodstoves.  Furnaces firing wood are similar in design and operation  to those burning coal.
Fireplaces are used primarily  for supplemental heating and for aesthetic effects.  Energy
efficiencies of prefabricated fireplaces are slightly higher than those of masonry fireplaces.42

              The combustion of fossil fuels or wood in residential units (woodstoves,
furnaces, fireplaces) is a relatively slow and low-temperature process.   Because combustion in
the residential sector tends to  be  less efficient than in other sectors, the potential to form
CDD/CDF may be greater.  Also, inadequate maintenance of these units may increase
potential for  CDD/CDF formation on particulate matter.  Furthermore,  residential combustion
units are generally not equipped with gaseous or particulate control devices.

              Process Description

              In the residential sector, coal is usually combusted in underfeed or hand-stoked
furnaces.  Stoker fed units are the most common design for warm-air furnaces and for boilers
used for steam or hot water production.  These units are typically controlled with an
automatic thermostat and designed for a specific type of coal.  Other coal-fired heating units
include hand-fed room heaters, metal stoves, and metal and masonry fireplaces.  Most  of the
goal combusted in all of these units is either bituminous or anthracite.14'43  These units  operate
at low temperatures and  do not efficiently combust fuel.  Generally, coal  contains small
quantities of chlorine and CDD/CDF precursors. Therefore,  the potential for CDD/CDF
formation exists.
                                          4-71

-------
              Oil-fired residential furnaces are designed with varying burner configurations,
each attempting to optimize fuel combustion efficiency.  Emissions from fuel oil combustion
depend on the grade and composition of the oil, the design of the furnace, and the level of
equipment maintenance.

              Important  fuel oil combustion properties include pumpability, heating value,
and ash content.  Lighter grade oils are more easily atomized and generally exhibit better
combustion properties than heavier grade oils.44 Both furnace design and equipment
maintenance influence combustion efficiency.  Particulate matter emissions depend most on
the ash content and grade of oil fired, with lighter grade oils exhibiting lower emissions.  Oil
contains only small amounts of chlorine and CDD/CDF precursors.

              Woodstoves  are used commonly in residences as space heaters to  supplement
conventional heating systems.  Woodstoves transfer heat by radiation from the hot stove walls
to the room.  Circulating stoves convert radiant energy to warm convection air.  Combustion
efficiencies for woodstoves are dependent on stove design and operating characteristics.
Consequently, combustion efficiency and emissions vary greatly among woodstoves.  For
purposes of estimating emissions, woodstoves are classified  into four categories: conventional
woodstoves, noncatalytic woodstoves, pellet stoves, and catalytic woodstoves.  These
categories are based on fuel type and emission reduction features.14'42 Woodstoves have a
greater potential to emit CDD/CDF than fossil  fuel-fired units due to the presence of
CDD/CDF precursors present in wood.  Figure 4-14 depicts a typical noncatalytic woodstove.

              Emission  Control Techniques

              As mentioned previously, residential combustion sources do not generally use
air pollution control devices.  The effect of controls that are used on CDD/CDF emissions has
not been studied.

              Coal-fired residential  combustion sources are  generally not equipped with PM
or gaseous pollutant control devices.  Changes in stove design and operating practices,
                                         4-72

-------
                           Probe
76"
      20"
                                             30"
                                                             To Continuous Gas Analyzers

                                                          • to FID
                                                                   Thermocouple
                                                                   Wires T Deg C
                                                                     T  ._   V.

                                                                              n
                                                                                      Asbestos
                                                                                      Aluminum
                                                                           -Scale
              Figure 4-14.  Simplified Diagram of a Freestanding Noncatalytic Woodstove


    Source: Reference 42.
                                                 4-73

-------
however, have been made to effect lower PM, hydrocarbon, and CO emissions. Changes
include modified combustion air flow control, better thermal control and heat storage, and the
use of combustion catalysts.14

             Residential oil- and wood-fired furnaces are not equipped with pollution control
equipment.  Residential fireplaces do not typically employ control devices.

             Wood stove emissions reduction features include baffles, secondary combustion
chambers, and catalytic combustors.  Catalytic combustors or converters are similar to those
used in automobiles.  Wood stove control devices may lose efficiency over time.  Control
degradation for any stoves, including noncatalytic woodstoves, may occur as a result  of
deteriorated seals and gaskets, misaligned baffles and bypass mechanisms, broken  refractories,
or other damaged functional components.42  In addition, combustion efficiencies may be
affected by differences in the sealing of the chamber and  control of the intake and exhaust
systems.14'43

             Emission Factors

             Emission factors  for coal-fired  residential furnaces are presented in
Table 4-16.43 These emission factors are based on average particulate CDD/CDF
concentrations from chimney soot samples collected from 7 coal stoves, and particulate
emission factors obtained from AP-42.42-44  These emission factors represent the
maximum emission rates from these sources,  as chimney soot may not be representative of the
particulate actually emitted to the atmosphere.

             Emission factors  for oil-fired residential furnaces are presented in Table 4-17.
These emission factors are based on average CDD/CDF concentrations in soot measured from
21 furnaces used in central heating, and particulate emission rates obtained from AP-42.43'45

             Emission factors  for a residential wood  stove, fireplace, and furnace are
presented in Table 4-18. These emission factors are based on average CDD/CDF
                                          4-74

-------
         TABLE 4-16.  CDD/CDF EMISSION FACTORS FOR COAL-FIRED
                           RESIDENTIAL FURNACES
                         FACTOR QUALITY RATING: U
 Isomer
                                                 Coal Furnaces"
                                      Ib/ton coal burned (mg/Mg coal burned)
        Anthracite
(AMS Code 21-04-001-000)
        Bituminous
(AMS Code 21-04-002-000)
 DIOXINS
    2,3,7,8-TCDD
    Total Other TCDD
    Total PeCDD
    Total HxCDD
    Total HpCDD
    Total OCDD
    Total CDD
    3.2xlO-9 (1.6xlO-3)
    1.2xlO-7 (6.0x1 Q-2)
    6.2xlO-8 (S.lxlO'2)
    1.2x10'7 (6.0x1O'2)
    I.lxl0-7(5.7xl0-2)
    1.5xlO-7(7.7xlO'2)
    5.6xlO'7 (2.9x10'')
    4.8x10-" (2.4xlO'3)
    l.SxlO-7 (9.0xlO'2)
    9.2xlO-8 (4.6xlO'2)
    1.8xlO-7(9.0xlO'2)
    l.VxlO'7 (8.6xlO-2)
    2.4xlO'7 (1.2x10'')
    8.7xlO-7 (4.3x10'')
 FURANS
    2,3,7,8-TCDF
    Total Other TCDF
    Total PeCDF
    Total HxCDF
    Total HpCDF
    Total OCDF
    Total CDF
    8.4xlQ-8 (4.2xlO':)
    7.4xlO-7 (3.7x10'')
    6.8xlO-7 (3.4x10'')
    2.6xlO'7 (1.3x10-')
    6.4xlQ-8 (3.2xlQ-2)
    8.4x10-9 (4.2x10'3)
    LSxlO'6 (9.2x10-')
     1.3xlO'7 (6.3xlO'2)
     l.lxlO'0 (5.5x10'')
     l.lxlO'" (5.5x10-')
     3.8xlO'7 (1.9x10'')
     9.4xlO's (4.7xlQ-2)
     1.3xlO-8 (6.3xlO-3)
      2.8xlO-(l (1.41)
Source:  Reference 43.
a Based on CDD/CDF particulate concentrations and paniculate emission factors from AP-42 as follows:
             Fuel
             anthracite
             bituminous
                 Emission Factor
           11.0 Ib/ton (5.5xl06 mg/Mg)
           16.5 Ib/ton (8.2xl06 mg/Mg)
                                      4-75

-------
TABLE 4-17. CDD/CDF EMISSION FACTORS FOR OIL-FIRED
              RESIDENTIAL FURNACES

           AMS 21-04-004-000, 21-04-005-000
            FACTOR QUALITY RATING: U
Isomer
DIOXINS
2,3,7,8-TCDD
Total Other TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
FURANS
2,3,7,8-TCDF
Total Other TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
Source: Reference 43.
a Based on participate CDD/CDF
Fuel
oil
Oil
Ib/gal

4.7xlO-13
6.9xlO'13
6.8xlO'13
5.5xlO-13
S.SxlO'13
S.SxlO'13
3.5xlO-12

4.4xlO'13
S.lxlO-12
3.5xlO-12
1.4xlQ-12
6.1xlO-13
2.5xlO-13
1.1x10-"

Central Heating"
mg/L

5.6xlO-8
8.3xlO-8
8.2xlO-8
6.6xlO-8
6.3xlO-8
6.6xlO'8
4.2x1 0-7

5.3xlO-8
G.lxlO-7
4.2xlO-7
1.7xlO-7
7.3xlO-8
3.0xlO-8
1.4X10'6

concentrations and paniculate emission factors from AP-42 as follows:
Emission Factor
2.5xlO-3 Ib/gal (3.0xlO: mg/L)

                       4-76

-------
   TABLE 4-18. AVERAGE CDD/CDF EMISSION FACTORS FOR WOOD-FIRED
                       RESIDENTIAL COMBUSTORS

                     FACTOR QUALITY RATING: U
 Isomer
 Wood Stove"
 (AMS Code
21-04-008-010)
Ib/ton (mg/Mg)
 wood burned
  Fireplace3
 (AMS Code
21-04-008-001)
Ib/ton (mg/Mg)
 wood burned
Wood Furnace3
 (AMS Code
21-04-008-010)
Ib/tori (mg/Mg)
 wood burned
DIOXINS
2,3,7,8-TCDD
Total Other TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
FURANS
2,3,7,8-TCDF
Total Other TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
Source: Reference 43.

9.2xlO-9 (4.6xlO-3)
1.7xlO-7(8.6xlO-2)
9.8xlO'7 (4.9x10-')
4.4xlO-7 (2.2x10-')
3.4xlQ-7 (1.7x10"')
3.0xlO-7 (1.5x10-')
2.2xlO-6 (1.1)

4.0xlO'7 (2.0x10'')
2.8xlO-6 (1.4)
3.4x10-° (1.7)
2.2xlO-6(l.l)
S.OxlO-7 (2.5x10"')
2.8xlO'7 (1.4x10'')
9.6x1 Q-6 (4.8)


2.8xlO'8 (1.4xlO-2)
NR
6.2xlO-7 (3.1x10-')
4.8xlO'8 (2.4xlO'2)
1.4xlQ-8 (7.0xlQ-3)
l.lxlO-8 (5.6xlO'3)
7.2xlQ-7 (3.6x10-')

NA
8.4xlQ-9 (4.2x10°)
4.0xlQ-8 (2.0xlQ-2)
4.8xlO-7 (2.4x10-')
l.lxlO-8 (5.6xlO-3)
2.8xlO-9 (1.4xlO'3)
5.4xlO'7 (2.7x10-')

a Based on paniculate CDD/CDF concentrations and emission factors from AP-42 as
Device Emission Factor
wood stove 46.2
fireplace 30.8
wood furnace 46.2
Ib/ton (2.3x10' mg/Mg)
Ib/ton (l.SxlO7 mg/Mg)
Ib/ton (2.3x1 07 mg/Mg)
(assumed to be identical

5.4xlO'9 (2.7xlO'3)
4.8xlO'7 (2.4x10-')
8.8xlO-7 (4.4x10-')
5.2xlO'7 (2.6x10-')
7.2xlO'7 (3.6x10'')
9.0xlO'7 (4.5x10"')
3. 5x10-" (1.8)

4.8xlO'7 (2.4x10'')
6.8x10'" (3.4)
l.lxlO'5 (5.5)
3.2xlO'(l (1.6)
5.8xlO'7 (2.9x10-')
1.2xlO'7 (5.9xlO-2)
2.2xlO'5 (11.1)

follows:
to wood stove)
NA = Not available.
NR = Not reported.
                                 4-77

-------
concentrations in soot measured at 18 combustion unit chimneys, and particulate emission
rates obtained from AP-42.43 These emission factors represent the maximum emission rates
from these sources, as chimney soot may not be representative of the particulate actually
emitted to the atmosphere.

              In one study, an Atlanta Stove Works freestanding noncatalytic woodstove,
depicted in Figure 4-14, was sampled.  The stove combusted oak and pine aged for one year.
The stove was operated at low burn rates and low operating temperatures for maximum wood-
use efficiency, which is representative of normal  residential use.  Burn rates for individual test
runs ranged from 2.9 to 7.7 Ib/hr (1.3 to 3.5 kg/hr).47  Sampling for CDD/CDF emissions was
performed at  the outlet exhaust stack in each of a series of three test runs. However, no valid
flue gas CDD/CDF emissions data were obtained because of the large amounts of
hydrocarbons present.46

              Woodstove ash and flue wipe samples showed minimal CDD/CDF content.
OCDD was the only homologue detected  in the ash samples analyzed.  The maximum OCDD
content of the ash samples was 0.09 ppb.   Small  quantities of OCDD were found in each of
the two flue wipe samples analyzed, with HpCDD also being detected in one of the two
samples.  The maximum OCDD content of the flue wipe samples was 0.6 parts per
billion (ppb),  and the measured HpCDD content was 0.04 ppb.46

              CDD/CDF  precursor analyses were performed on samples of the wood fed to
the stove.  The specific CDD/CDF precursors analyzed for were chlorophenols,
chlorobenzenes, PCB, and total chlorides.  Chlorobenzenes, chlorophenols, and PCB were not
detected in the oak and pine samples analyzed. The total chloride contents of the oak and
pine samples  were 125 parts per million (ppm) and 49 ppm, respectively. In addition,
continuous emissions monitoring was performed at the stove exhaust location for  O2.  The
average O2 content of the flue gas was 17.0 percent volume.46

              Three additional  studies provided information on CDD concentrations in the  ash
collected from 24 woodstoves.  The woodstoves tested were located in rural areas in three
                                        4-78

-------
different regions of the country.  Presumably, the wood being burned was untreated, that is, it
had not been exposed to fungicides, herbicides, or wood preservatives.  No analysis was done
for the PeCDD homologue.  For the 24 woodstoves tested, CDD concentrations in ash
samples ranged from 0.007-210 ppb, with a mean concentration of 23.4 ppb.  Seventeen
samples were analyzed for the 2,3,7,8-TCDD isomer.  2,3,7,8-TCDD was not detected in two
samples. The detection limits of the samples ranged from 0.0009 to 0.0014 ppb.  The other
15 samples had concentrations of 2,3,7,8-TCDD varying from 0.001 to 0.20 ppb,  with an
average concentration of 0.05 ppb.  The authors of one of the studies, in which 18
woodstoves were tested, attributed some of the variability in the results to differences in
woodstove design and sampling points.  They also suggested that some of the variability could
potentially be attributed to fuel contamination, although feed samples were not analyzed for
CDD content.9

             In another study, ash  samples  from the chimneys of two fireplaces were
analyzed for CDD.  One fireplace was 12 years old and one was 25 years old. The latter had
total CDD concentrations of 44.7 ppb, including 1 ppb of 2,3,7,8-TCDD.  Ash samples from
the 12-year-old fireplace contained 1.79  ppb  CDD.  No TCDD isomers were detected at a
detection limit of 0.04 ppb.  The PeCDD homologue was not analyzed for in  either of these
samples. Ash samples scraped from the flue pipe of a residential heater combusting wood
found CDD levels of 0.97 ppb.9

             Source Locations

             Locations of residential  combustion sources are tied directly to  population
trends.47  Coal consumption for residential combustion purposes occurs mainly in the
Northeast, Appalachian, and  Midwest regions.  Residential oil consumption is greatest in the
Northeast and Mid-Atlantic regions.  Wood-fired residential units are generally concentrated
in heavily forested areas of the United States, which reflects fuel selection based on
availability and price.14
                                         4-79

-------
4.2.4         Waste Tire Incineration

              Waste tires are incinerated for energy recovery and disposal purposes.  Tires
are combusted at tire-to-energy facilities, cement kilns, tire manufacturing facilities, and as
supplemental fuel in boilers, especially in the pulp and paper industry.  The U.S. EPA
estimates that about 0.5 million metric tons (500 million kg) of tires are incinerated annually
in the United States.48  In 1990, 25.9  million (about 11 percent) of the 242 million tires
discarded in the United States were converted to energy.48  One report indicates that seven
cement kilns utilized about 23 percent of the scrap tires in 1990,  and that one tire-to-energy
facility utilized about  19 percent.  The Scrap Tire Management Council reports that about 46
percent of discarded tires were utilized by eight different pulp and paper facilities.49

              Process Description

              The combustion processes and procedures for burning discarded tires are the
same as described previously within this section of this report.

              Emission Control Techniques

              Available information from one tire-to-energy facility indicates the use  of a
spray dryer combined  with a fabric  filter for an  air pollution control device.50  These devices
are capable of greater  than 95 percent reduction and control of CDD/CDF compounds.
However, operational and control  device information for other tire incineration facilities in the
United States is not known.

              Emission Factors

              Emissions data and test reports from tire incineration facilities available at the
time this report was prepared were limited.  One test report available was from a study
conducted at a tire-to-energy facility in California.50  The facility consists of two excess-air
incinerators equipped with steam boilers for energy recovery.  Whole tires were fed at a rate
                                           4-80

-------
of 1364 Ib/hr (3000 kg/hr). The facility uses a spray dryer and flue gas desulfurization
followed by a fabric filter to control emissions.50

             Emission factors for total CDD/CDF and TEQ in units of mg/kg of tires
combusted were developed from emissions test results at this one facility.50  From these data,
average CDD/CDF emission factors were estimated and are presented in Table 4-19.  Extreme
caution should be used in applying these emission factors to any other incinerator.  If another
facility was not equipped with the same devices, then the uncontrolled emissions of
CDD/CDF could be much greater.
         TABLE 4-19.  CDD/CDF EMISSION FACTORS FROM WASTE TIRE
                                  INCINERATION
                                  SCC 5-03-001-08
                          FACTOR QUALITY RATING:  E
Isomer
2,3,7,8-TCDD
2,3,7,8-TCDF
2,3,7,8-TCDD TEQ
Total CDD
Total CDF
Ib/ton tires burned
2.16x10-"
5.42x10-"
1. OSxlO'9
6.50x1 0-9
2.14xlO-8
mg/kg tires bumed
1.08xlO's
2.71xlO's
5.40x10'"
3.25x10-°
1.07xlO'5
Source:  Reference 50.

             Source Locations

             Because  the burning of tires as waste occurs nationwide for various types of
industries and combustors, no attempt was made to list specific sources or sites.
                                        4-81

-------
4.3           CREMATORIES

4.3.1          Process Description

              Propane-fired Eclipse Burners (afterburner and ignition) are used at cemeteries
for human body cremation. Eclipse Burners are rated at 2,115,000 Btu per hour capacity.
Newer units installed in the late 1980's are equipped with a modulating ignition burner.
When afterburner temperatures reach about 1800°F (980°C), the ignition burner modulates to
a low-fire mode that will reduce the Btu per hour usage.

              When the crematory reaches an operating temperature of 1,250°F (680°C) the
body container is placed on the combustion  chamber grate and the ignition burner is fired to
attain a target combustion temperature sufficient for the proper reduction of human  remains.
The chamber preheat by the afterburner reaches  1,250°F (680°C) in about 30 to 45  minutes
prior to ash removal.  When the body container is introduced into the combustion chamber,
and the burner is ignited, cremation begins at about 1600 to  1800°F (870 to 980°C). Flame
impingement on the body takes two to three minutes; cremation occurs for about two hours.
The remains are then raked towards the ignition  burner for about two minutes.  Cooldown
follows for 45 minutes to 1.5  hours.  During normal operation, three bodies per day are
cremated in each retort.

4.3.2          Emission Factors

              Evaluation tests on two propane fired crematories at a cemetery in California
were conducted through a cooperative effort with the Sacramento Metropolitan Air Quality
Management  District to determine emissions of toxic substances from a crematory.51  The
units were calibrated to operate at a maximum of 1,450,000 Btu/hour.  Emissions testing was
performed over a two  week period; thirty-six bodies were cremated during the test period.
This equates to two bodies per crematory per day for nine days.  The body and cardboard
weights and wood process rates for each test per crematory were reported.
                                         4-82

-------
              CDD/CDF sampling, recovery, and analysis were performed in accordance with
California Air Resources Board (CARB) Method 428, which is based on the use of EPA
Reference Modified Method 5 sampling train.  Data from stack gas measurements from each
of the nine tests performed during the evaluation program were tabulated and reported.
Emission factors developed from these data are presented in Table 4-20.

4.3.3          Source Location

              In 1991, there were about 400,500 cremations in more than 1,000 crematories
located throughout the United  States.  Table 4-21 lists the number of crematories located in
each state and the estimated number of cremations performed in each state for the year  1990,
where itemized data were available.52

4.4           IRON AND STEEL FOUNDRIES/SCRAP METAL MELTING

4.4.1          Process Description

              Iron and steel foundries can be defined as those that produce gray, white,
ductile, or malleable iron and steel castings.  Both cast irons and steels are solid solutions of
iron, carbon, and various alloying materials.  Iron foundries produce iron castings from  scrap
iron, pig iron, and foundry returns by melting, alloying, and molding.  The major operations
include:  (1) raw material handling and preparation,  (2) metal melting, (3) mold and core
production,  and (4) casting and finishing.  A generic flow diagram for iron and steel foundries
is shown in Figure 4-15. Figure 4-16 depicts the emission points in a typical iron foundry.

              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.  The metal
melting process is  accomplished primarily in cupola (or blast) furnaces, and to a lesser extent
in electric arc furnaces (EAF).  The cupola, which is the major type of furnace used in
industry today, is typically a vertical, cylindrical steel shell with either a refractory lined or
water cooled inner wall.  Refractory linings usually consist of silica brick, or dolomite or
                                          4-83

-------
        TABLE 4-20.  CDD/CDF EMISSION FACTORS FROM A CREMATORY
Isomer
Average Emission Factor in Ib/body
           incinerated
       (kg/body incinerated)
     Range of Data
  Ib/body incinerated
 (kg/body incinerated)
2,3,7,8-TCDD


Total TCDD


1,2,3,7,8-PeCDD


Total PeCDD


1,2,3,4,7,8-HxCDD


1,2,3,6,7,8-HxCDD


1,2,3,7,8,9-HxCDD


Total HxCDD


1,2,3,4,6,7,8-HpCDD


Total HpCDD


Total OCDD


Total CDD


2,3,7,8-TCDF


Total TCDF


1,2,3,7,8-PeCDF


2,3,4,7,8-PeCDF
            4.58xlQ-14
           (2.08x1 0-14)
            8.86xlO-13
           (4.03xlO'13)

            1.44xlO-13
           (6.53xlO-14)
            1.37xlO-12
           (6.21xlO-13)

            1.73xlO-13
           (7.85xlO-14)

            2.50xlO-13
            3.12xlO-'3
           (1.42X10'13)
            3.55xlO-'2
            2.37xlO-12
           (l.OSxlO-12)

            5.09xlO'12
           (2.31xlQ-12)

            3.77xlO'12
            1.47x10-"
           (6.67xlO-12)

            3.31X10'13
           (l.SOxlO-13)

            6.90xlO'12
           (3.13xlO-12)

           <2.01xlO-13
          (< 9.12xlO-14)

           <5.76xlO'13
          (< 2.61xlO'13)
 1.69xlO-14 - 9.54xlO-14
(7.67xlO-15 - 4.33xlO-14)

 1.76xlO-13- 1.91X10'12
(7.67xlO-15 - 4.33xlO-14)

 3.60xlO-14 - 2.79xlO-13
(1.63xlO-14- 1.27xlO-'3)

 4.33xlO'13 - 3.23xlO-12
(1.96xlO-13- 1.47xlO-'2)

 4.85xlO-14 - 3.96xlO-13
(2.20xlO'14 - l.SOxlQ-13)

 5.21xlO-14 - 6.02xlO-13
(2.36xlO-14 - 2.73xlO'13)

 3.96xlO-'4 - S.OSxlO-13
(l.SOxlO'14- 3.67xlO'13)

 9.54x10-13 - 8.08x10-'2
(4.33xlO-13 - 3.67xlO'12)

 3.60xlO'13 - 5.29xlO'12
(1.63xlO'13 - 2.40xlO-'2)

 S.OSxlO'12 - l.lOxlO'"
(3.67xlO-12 - 4.99xlO-12)

 6.53xlQ-13 - 6.46xlO-12
(3.02xlO-13 - 2.93xlO'12)

 3.02xlO'12 - 3.07x10-"
(1.37xlO'12 - 1.39x10-")

 9.03xlO-'4 - 5.07xlO-13
(4.10xlO-14- 2.30xlO-'3)

 2.28xlO'12 - 1.62x10-"
(1.03xlO'12- 7.35xlO-12)

 8.81xlO-14 - 4.26xlO-'3
(3.40xlO'14- 1.93xlO-'3)

 l.lOxlO-13 - l.lOxlO-12
(4.99xlO'14 - 4.99xlO-13)
                                        4-84

-------
 TABLE 4-20.  CDD/CDF EMISSION FACTORS FROM A CREMATORY (CONTINUED)
Average Emission Factor in Ib/body Range of Data
incinerated Ib/body incinerated
Isomer (kg/body incinerated) (kg/body incinerated)
Total PeCDF
1,2,3,4,7,8-HxCDF

1,2,3,6,7,8-HxCDF

1,2,3,7,8,9-HxCDF

2,3,4,6,7,8-HxCDF

Total HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF

Total HpCDF
Total OCDF
Total CDF
4.06xlO'12
(1.84xlO'12)
5.97xlO-13
(2.71xlO-13)
5.38xlO-13
(2.44xlO-13)
l.OSxlO-12
(4.76xlO-13)
2.16xlO-13
(9.80xlO-14)
6.85xlO-12
< 3.08xlO'12
(< 1.40xlO-12)
< 1.89xlO-13
(< 8.57xlO-14)
< 3.62xlO-12
(< 1.64xlO-12)
l.OlxlO-12
(4.58xlQ-13)
< 2.24x10-"
(<1. 02x10-")
2.13xlO-13
(9.66xlO-14
2.06xlO'13
(9.34xl.O-14
1.76xlO-13
(7.98xlQ-14
3.82xlO-13
(1.73xlO-13
8.08xlO'14
(3.67xlO-14
2.20xlO-12
(9.98xlO'13
7.34xlO'13
(3.33xlQ-13
3.30xlO-14
(l.SOxlQ-14
S.OSxlO'13
(3.67xlO-13
3.82xlO-13
(1.73xlO-13
6.89xlO-12
(3.13xlO-12
- 1.03x10-"
- 4.67xlQ-12)
- 1.25xlO'12
- 5.67xlO-13)
- 1.25xlO'12
- 5.67xlQ-13)
- 2.35xlO-12
- 1.07xlO-12)
- 4.70xlO-13
- 2.13xlO-13)
- 1.62x10-"
- 7.35xlO-12)
- 7.27xlO'12
- 3.30xlO-12)
-4.92xlO'13
- 2.23xlO-13)
- 8.8U1Q-'2
- 4.00x1 Q-12)
- 1.69X10'12
- 7.67xlO'13)
- 5.31x10'"
- 2.41x10-")
Source: Reference 51.

a Sampled at the stack.
b Both units equipped with afterburners.
                                     4-85

-------
           TABLE 4-21.  1991 U.S. CREMATORY LOCATIONS BY STATE
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
No. of
Crematories
6
7
26
13
141
28
10
4
1
95
14
10
12
44
21
15
10
5
6
4
17
13
38
18
4
19
No. of
Cremations3
1,138
790
10,189
1,787
86,374
7,432
4,260
1,165
b
46,775
2,684
3,495
1,949
12,083
3,636
2,241
1,559
1,192
1,853
2,656
5,587
8,104
13,431
5,662
450
4,637
State
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming

No. of
Crematories
12
6
11
6
16
9
40
24
1
41
9
34
44
5
10
4
8
36
5
5
25
46
6
29
2

No. of
Cremations3
2,502
1,139
5,009
1,842
14,427
2,134
23,946
4,749
b
12,552
1,372
9,020
12,153
1,842
1,764
b
1,712
9,340
769
1,570
6,097
15,673
582
5,541
b

Source:  Reference 52.

a 1990 data. Data allocated by state for 1991 were not available.
" No information available.
                                        4-86

-------
                                                                                      (L

                                                                                      i
                                                                                      "m
                                                                                       a
                                                                                      O
                                                                                      •o
                                                                                       CD
                                            Q-O
                                            5-0
                                             O c
    o>

   '•5
      O)

      5e

      i
                        o
  a  ca
  «  g-
    CL
§
CO
     1 S
     a a
     w g-
CD
CO
si
2 IS
3 «
ECJ
8*
1 e
                                                                   O
                                                                   O
                                                                f    s
                                                                                   -a

                                                                                    3
                                                                                    O
                                                                                   U.

                                                                                   "o
                                                                                    u
                                                                                   -*—•
                                                                                   C/3
                                                                                   •a

                                                                                    as

                                                                                    o
                                                                                                           CO
                                                                                    C3

                                                                                    Cu
                                                                                    ?*"*

                                                                                    a

                                                                                   £


                                                                                    I
                                                                                    ab

                                                                                   Q
                                                               _w as

                                                               O £"
                                                                  Q.
                                                                                                           e
                                                                                                           £X
00

                                                                   « O 3
                                                                                        O CD
                                                                                        O)
                                                                                                                  m
                                                                                                                  m
                                                                                                                   o
                                                                                                                   o

                                                                                                                   u
                                                                                                u
                                                                                                o:
                                                                                                u
                                                                                                a
                                                                                                i
                                                                                                00
                                                4-87

-------
                                                           Fugitive
                                                         Particulates
   Fugitive
    Dust
                                   Raw Materials,
                                 Unloading, Storage,
                                      Transfer
                                 • Flux
                                 • Metals
                                 • Carbon Sources
                                 • Sand
                                 • Binder
                                      Scrap
                                    Preparation
          Fumes and
         Fugitive Dust

            Fugitive
        -*•   Dust
                            Hydrocarbons,
                           Co, and Smoke
                          Furnace
                           Vent
 Furnace
» Cupola
» Electric Arc
» Induction
» Other
                                                 r	
                                     Tapping,
                                     Treating
                                   Mold Pouring,
                                     Cooling
                           Fugitive
                          Fumes and
                             Dust
                                                               Fugitive
                                                              Fumes and
                                                                 Dust
                                                                             Core Making
                                                       Oven
                                                       Vent
                                          Core Baking
      Sand
                                     Casting
                                    Shakeout
                                     Cooling
                                Cleaning, Finishing
                          Fugitive
                            Dust
                          Fumes and
                            Fugitive
                             Dust
                           Fugitive
                             Dust
                                     Shipping
                Figure 4-16. Emission Points in a Typical Iron and Steel Foundry


Source: Reference 53.
                                               4-S

-------
magnesium brick.  Cupolas are charged with alternate layers of coke, metallics, and fluxes.
Combustion air is introduced into the cupola through tuyeres located at the base. About
70 percent of all iron castings are produced using cupolas, while  steel foundries rely almost
exclusively on EAFs or induction furnaces for melting purposes.

             The heat produced by the burning coke melts the iron, which flows down and
is tapped from the bottom of the cupola.  Fluxes combine with non-metallic impurities in the
charge and form slag, which is removed through tap holes at the  bottom of the cupola.
Cupola capacities range from 1 to 30 tons (1 to 27 Mg) per hour, with a few large units
capable of producing close to 100 tons (90 Mg) per hour.  Larger furnaces are operated
continuously, with periodic inspections and cleanings between burn cycles.53

             In 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 necessary.   The castings
are then inspected and shipped to another industry for machining and/or assembly into a final
product.54

4.4.2         Emission Control Techniques

             Emissions from cupolas can vary widely, depending on blast rate, blast
temperature, melt rate, coke-to-melt ratio and control technologies.  Control technologies
commonly used to control emissions from iron and steel foundry  metal melting  operations
include baghouses, wet scrubbers, and afterburners.  Additionally, emissions due to coke
combustion may be reduced by substitution of gas for heat or the use of graphite as a carbon
source.53

             Scrap preparation with heat will emit smoke, organic compounds, and carbon
monoxide, and scrap preparation with solvent degreasers will emit organics.  Catalytic
incinerators and afterburners can control about 95 percent of organic and carbon monoxide
emissions.  Emissions released from the melting furnaces include particulate matter, carbon
                                          4-89

-------
monoxide, organic compounds, sulfur dioxide, nitrogen oxides, and small quantities of
chloride and fluoride compounds.

4.4.3          Emission Factors

              Scrap metal melting processes have been found to be one source of CDD/CDF.
The use of chlorinated compounds in iron and steel processes and the use of recycled scrap
metal contaminated with cutting oils and plastics containing chlorine provide all conditions
required for the formation of chlorinated aromatic compounds.

              A study funded  by the Swedish Steel Producers Association noted that the
amount of chlorine  loaded (into a furnace) is of importance, but the design of the charging
process seems to be the determining factor for the formation of CDD/CDF. The Swedish
study was carried out  in a pilot plant with a 10-ton electric furnace charging scrap metal
under different operational conditions, either continuously through  the furnace or batchwise
into the open furnace.  CDD/CDF were detected in the range of 0.1 to 1.5 ng TCDD-
equivalents per normal cubic meter (Nm3) dry gas.  The largest emissions were observed
during charging of scrap metal containing PVC plastics."3

              The emission factors presented in Table 4-22 were developed from a  facility
test reporgenerated to  comply with the requirements of California Assembly Bill (AB2588).55
The test program  quantified emissions from a batch-operated cupola furnace charged with pig
iron, scrap iron, steel scrap, coke, and limestone. Emission control devices operating during
the tests were  an oil-fired afterburner and a baghouse.

              Coke combined  with combustion air provided the heat necessary  to melt the
metal, which was continuously tapped from the cupola, converted to ductile iron, and poured
into steel pipe molds.   Combustion gases  from the cupola were vented to a gas-/oil-fired
afterburner followed by a baghouse.
                                         4-90

-------
     TABLE 4-22. CDD/CDF EMISSION FACTORS FROM A CUPOLA FURNACE3

                                SCC 3-04-003-01


                         FACTOR QUALITY RATING:  D

Isomer
2,3,7,8-TCDD
Total TCDD
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,6,7,8-HxCDD
Total HxCDD
1,2,3,4,6,7,8-HpCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
1,2,3, 7,8-PeCDF
2,3,4,7,8-PeCDF
Total PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
2,3,4,6,7,8-HxCDF
Total HxCDF
1,2,3,4,6,7,8-HpCDF
Total HpCDF
Total OCDF
Total CDF
Source: Reference 55.
Average
Ib/ton
6.61x10-"
7.92xlO-9
1.71x10-'°
3.52xlO'9
l.OlxlO-10
LlOxlO-9
1.85x10-'°
3.81x10-'°
NR
1.34xlO-8
1.04xlO'9
5.16xlO-8
6.10x10-'°
6.99x10-'°
1.70xlO-8
3.79x10-'°
3.39x10-'°
2.02x10-'°
3.47xlO-9
3.85x10-'°
4.87x10-'°
1.17x10-'°
7.63xlO-8

Emission Factor
kg/Mgb
3.31x10-"
3.96xlO-9
8.55x10-"
1.76xlO'9
5.05x10-"
5.50x10-'°
9.25x10-"
1.91x10-'°
NR
6.72x1 0-9
5.20x10-'°
2.58xlO-8
3.05x10-'°
3.50x10-'°
8.50xlQ-7
1.90x10-'°
1.70x10-'°
l.OlxlO'10
1.74xlO-';
1.93x10-'°
2.44x10-'°
5.85x10'"
8.8xlQ-7

j* Control device: Afterbumer/baghouse.
" Emission factors are Ib (kg) of pollutant per ton (Mg) of metal charged.

NR = Not reported
                                     4-91

-------
4.4.4         Source Location

              Based on a survey conducted by the EPA in support of the iron and steel
foundry maximum achievable control technology (MACT) standard development, there were
756 iron and steel foundries in the United States in 1992.56  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
obtained from the following trade associations:

              •     American Foundrymen's Society,  Des Plaines, Illinois;
              •     National Foundry Association, Des Plaines, Illinois;
              •     Ductile Iron Society, Mountainside.  New Jersey;
              •     Iron Casting Society, Warrendale. Pennsylvania; and
              •     Steel Founders' Society of America. Des Plaines. Illinois.

4 5           COMBUSTION-AIDED METAL RECOVERY

              This section discusses CDD CDF  emissions from secondary metals recover)
facilities that use combustion to eliminate combustible materials present in scrap raw material.
During combustion, various  solids (e.g , plastics) and liquids (e.g., solvents or oils) are burned
off in initial processing steps, leaving the metals free of combustible contaminants and
suitable for further processing.  The combustion of chlorine-containing plastics or liquids in
these processes can produce  CDD/CDF emissions.

              This section describes the processes used for five types of secondary metaj
recovery:  (1) secondary copper smelters.  (2) secondary aluminum production, (3) secondary
lead production. (4) scrap metal reclamation  furnaces, and (5) drum and barrel reclamation
                                          4-92

-------
furnaces.  CDD/CDF emission factors are given for each process.  Each emission factor
presented for these sources represents a unique furnace.  A factor for one furnace should not
be applied to another similar furnace without considering differences in feedstock and
combustion conditions.

              In addition to the five listed above, secondary ferrous metals recovery processes
could produce CDD/CDF emissions. However, CDD/CDF emissions test data are not
currently available for those processes.

4.5.1          Secondary Copper Smelters

              Secondary copper smelters recover copper from copper-bearing scrap materials,
including  electronic materials scrap, brass, iron-bearing copper scrap, and other copper-bearing
materials.  Some of the  scrap materials contain chlorinated plastics such as polyvinyl  chloride
(PVC).  CDD/CDF are produced as the plastic and other combustible materials are  combusted
in the blast furnace.  Figure 4-17 presents a general process flow diagram of a secondary
copper smelter.

              Process Description

              The feed  material used  in secondary copper recovery' can be pretreated using
several different procedures, either separately or in combination.  Feed scrap is concentrated
by manual or mechanical methods  such as sorting,  stripping, shredding, and magnetic
separation. Feed scrap is sometimes formed into briquettes  in a hydraulic press.
Pyrometallurgical pretreatment may include sweating, burning of insulation (especialh from
wire scrap), and drying  (burning off oil and volatiles) in rotary kilns. These techniques may
cause the  formation of CDD/CDF.  Hydrometallurgical  methods include flotation and leaching
with chemical recovery.

              Pretreated scrap that contains 10 to  30 percent copper is normally smelted in a
cupola-type blast furnace. A cupola furnace is a vertical, refractory-lined cylinder open at the
                                          4-93

-------
    ENTERING THE  SYSTEM
                               LEAVING  THE SYSTEM
    LOW GRADE SCRAP.
     (SLAG.  SKIMMINGS.
DROSS. CHIPS.  BORINGS)

                  FUEL

                   AIR
                  FLUX

                  FUEL

                   AIR
                  FLUX

                  FUEL
                  FLUX

                  FUEL
                  AIR

                 FUEL

        , C IN G MEDIUM
             (P C J N G;
PYROMETALLURGICAL
   PRETREATMENT
     (DRYING)
  (SCC 3-04-002-07)
                                     TREATED
                                     SCRAP
      CUPOLA

   (SCC 3-04-002-10)
 SMELTING FURNACE
  (REVERBERATORY)
   (SCC 3-04-002-14)
                                     SEPARATED
                                     COPPER
                                                  SLAG
                                        CONVER'tH
                                       (SCC 3-04-002-50)
                           BLIS'ER
                                                         BLISTE
                PRC
               (SCC 3-04-002-39)
GASES. DUST. METAL  OXIDES
TO CONTROL EQUIPMENT


BLACK SLAG'
COPPER

                                CARBON MONOXIDE  PARTICULATE DOST,
                             _^ METAL OXIDES  TO AFTERBURNER  AKD
                                PARTICULATE CONTROL
                                SLAG TO  DISPOSAL
GASES AND METAL OXIDES
TO CONTROL EQUIPMENT
                                      AND ME~AL OXIDES
                                TO CONTROL EQUIPMENT
                                                                         FUGP'XE  ME-A. CXiDES FROM
                                     _^ GASES METAL  oi'S*
                                       TO  CON'ROL  DEV.C;
                                REFINED  COPPEP
                 Figure 4-17.  Secondary Copper Recovery Process Flow Diagram

 Source: Reference 57.
                                                  4-94

-------
top and equipped with vents at the bottom.  Alternative charges of scrap, coke, and limestone
are placed on top of a burning bed of coke. As the scrap is heated, the metal melts and is
drawn off through a tap hole and spout at the bottom of the furnace; the combustibles are
burned off and combustion gases and PM exit the furnace. Oxides of copper and heavy
metals are chemically reduced. Various impurities, such as iron, combine to form a slag,
which collects on top of the molten metal and can be drawn off separately.

              In a typical system, further smelting and refining are accomplished using a
reverberator/ holding furnace, a converter, and a reverberatory or rotary refining furnace.
The holding  furnace retains the melt until a sufficient batch is accumulated to form a charge
to the converter, and allows for tapping the slag.  (An electric arc furnace can also be used
for this purpose.) Feed with a low-copper value can also be smelted in electric crucible or
pot furnaces, where pure oxygen  is used in place of air for oxidation.

              A converter consists of a cylindrical steel shell that can be rotated about its
longitudinal axis. An opening in one side emits the molten charge and vents gases.  Air is
blown through the melt by means of a  horizontal row of pipes with openings (tuyeres) that
are belcm the liquid metal when the  furnace is rotated.   A silica flux is added to remove iron
from the metal, whereas zinc and any sulfur are converted to their respective oxides b> the air
that is blown in.

              The product from the converter is blister copper,  usually 90 to 99 percent pure
This material may be poured and cast into  ingots or it may be transferred while molten to
another furnace for a final pyrometallurgical process  known  as fire refining.

              Blister copper is typically purified further by fire refining to about  99.9 percent
copper.  Fire-refined copper is cast into wirebar as well as ingots.  The refining processes  are
essentially the same in secondary smelting  as in primary  copper smelting.  Fire-refining
furnaces are  typically reverberatory or  rotary furnaces.  In both  furnaces, air is blown through
the molten metal to oxidize impurities  that are removed as oxides in  the slag, which is
skimmed or poured off.  Copper oxide, formed to the extent of less than 1 percent of copper,
                                          4-95

-------
is reduced by "poling" (submerging wooden poles in the molten metal) or by supplying a
reducing atmosphere of gas (by fuel-rich combustion).  The usual sequence of events in fire
refining is (1) charging, (2) melting, (3) skimming,  (4) blowing,  (5) adding fluxes,
(6) reducing, (7) reskimming, and (8) pouring.

              Electrolytic refining may be done as an additional  step to produce electrolytic
copper. Electrolytic refining separates impurities from the copper by electrolysis in a solution
containing both  copper sulfate and sulfuric acid. Metallic impurities form a sludge that can
be removed and treated for recovery of precious metals.

              Emission Control Techniques

              Generally, afterburners (usually natural gas-fired) are located at the top of the
cupola furnace and serve to complete the combustion of the exhaust gases.  These afterburners
control emissions of unburned combustible PM  and  organic compounds.

              Exhaust gases from the furnace after the afterburners are typically cooled with
\\ater in a spray chamber and mixed with ventilation gases from  the furnace charge floor
and or ambient air  Generally, this gas stream is then passed through a fabric filter
(baghouse) before  release to the atmosphere.

              Emission Factors

              Emission factors were identified for a secondary copper recovery cupola
furnace fmng scrap materials that included shredded telephone equipment, other copper-
bearing metallic scrap, metallurgical  slags, and plant revert, along with  coke and limestone.58
The shredded telephone equipment was composed of circuit boards, switching gear, telephone
parts, and other  miscellaneous plastic parts.  Some of the plastic  contained in the scrap was
PVC.  The total  amount of telephone scrap processed accounted for 22  percent by weight of
the total scrap feed.  No other scrap materials contained plastic materials.
                                          4-96

-------
             The facility was equipped with natural gas-fired afterburners and a fabric filter.
Flue gas temperature after the afterburners averaged 1,610°F (877°C).  After water spray
cooling, flue gas temperatures at the inlet to the fabric filter averaged 320°F (146°C).

             Emissions of TCDD/TCDF through OCDD/OCDF were measured after the
fabric filter at the stack outlet.  Table 4-23 presents CDD/CDF emissions on a flue
gas-concentration basis and  as emission factors. Emission factors are based on the total
weight of scrap metal (plastic and nonplastic-bearing metal) fed to the furnace.

4.5.2         Secondary Aluminum Production

             Secondary aluminum producers recycle aluminum from aluminum-containing
scrap, while primary aluminum producers convert bauxite ore into  aluminum.  The secondary
aluminum industry was responsible  for 27.5  percent of domestic aluminum produced in 1989.
There are approximately 116 plants with a recovery capacity of approximately 2.6 million
tons (2.4 million megagrams) of aluminum per year.  Actual total secondary aluminum
production was  relatively  constant during the 1980s.  However, increased demand for
aluminum by the automobile industry has doubled in the last 10  years to an average of
P3 pounds (78.5 kilograms) per car. Recycling of used aluminum beverage cans (UBC)
increased more  than 26 percent from 1986 to 1989.  In  1989, 1.4 million tons  (1.3 million
megagrams) of  UBCs were recycled, representing over 60 percent of cans  shipped   Recycling
a ton of aluminum requires  only 5 percent of the energy required to refine a ton of primary
aluminum from bauxite ore, making the secondary aluminum economically viable

             Process Description

             Secondary aluminum production involves two  general categories of operations-
scrap pretreatment and smelting/refining. Pretreatment operations include sorting, processing,
and cleaning scrap.  Smelting/refining operations include cleaning,  melting, refining, alloying.
and pouring of  aluminum recovered from scrap. The processes used to convert scrap
aluminum to products such as lightweight aluminum alloys for industrial castings are in
                                         4-97

-------
  TABLE 4-23.  CDD/CDF EMISSION CONCENTRATIONS AND EMISSION FACTORS
                    FOR SECONDARY COPPER SMELTING -
                    COPPER RECOVERY CUPOLA FURNACE

                                SCC 3-04-002-11
                         FACTOR QUALITY RATING: D
Isomer
DIOXINS
2,3,7,8-TCDD
Total Other TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
FURANS
2.3.".8-TCDF
Total Other TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
TOTAL CDD ''CDF
Flue Gas
Concentration in Ib/ft3
(Hg/dscm at 3% 02)

1.45x10-* (232)
7.12xlO-8(l,140)
1.13xlO-7 (1,810)
1.45xlO-7 (2,320)
2.43xlO'7 (3,890)
1.57xlO'7 (2,520)
7.43xlO'7 (11,900)

3.17x10'" (5,070)
1.29xlO'( (20.600)
1.00x10'" (16.100)
4.86x10'" (7,790)
3.98x10'" (6,380)
2.93x10" (4,700)
3 79x10'" (60.700)
4.53x10'° (72.600)
Emission Factor in Ib/ton
(fig/kg scrap feed)1^

2.54xlO'7 (0.127)
1. 22x1 0-6 (0.609)
1.94X10'6 (0.970)
2.52xlO-6(1.26)
4.16xlO-6 (2.08)
2. 70xlO'(' (1.35)
1.28xlO'5 (6.39)

5.44x10'° (2 72)
2. 20x10' ' (11.0)
1.73x10-' (8 64)
848x10'" (4.24)
6.84x10'° (3.42)
5.04x10'" (2.52)
6. 50x10 • (32.5)
7.78xlO'5 (38.89)
Source  Reference 58

a Emissions measured in the stack gases after an afterburner and a fabric filter
k Includes all scrap feed (plastic and nonplastic-bearmg) including coke and limestone
c The feed scrap for this furnace contained 22 percent by weight plastic-containing scrap
                                     4-98

-------
involved at any one facility.  Some steps may be combined or reordered, depending on the
type of scrap used (see Figures 4-18a and 4-18b).  Some or all the steps in these figures may
be quality, source of scrap, auxiliary equipment available, furnace design, and product
specifications. Plant configuration, scrap type usage, and product output varies throughout the
secondary aluminum industry.

             Scrap Pretreatment-Aluminum scrap comes from a variety of sources.  "New"
scrap is generated by pre-consumer sources, such as drilling and machining of aluminum
castings, scrap from aluminum fabrication and manufacturing operations, and aluminum
bearing residual material (dross) skimmed off molten aluminum during smelting operations.
"Old" aluminum scrap is material that has been used by the consumer and discarded.
Examples of old scrap include used appliances, aluminum foil, automobile and airplane parts,
aluminum siding, and beverage cans.59

             Scrap pretreatment involves sorting and processing scrap to remove
contaminants and to prepare  the material for smelting.  Sorting and processing separates the
aluminum from other metals, dirt, oil, plastics, and paint.  Pretreatment cleaning processes are
based on mechanical, pyrometallurgical, and hydrometallurgical techniques.

             Mechanical Cleaning:  Mechanical cleaning includes the physical separation of
aluminum from other scrap,  with hammer mills, ring rushers, and other machines to break
scrap containing aluminum into smaller pieces.  This improves the efficiency of downstream
recovery by magnetic removal of iron.  Other recovery' processes  include  vibratory screens
and air classifiers.

             An example of mechanical cleaning is the dry milling  process.  Cold
aluminum-laden dross and other residues are processed by milling and screening to obtain a
product containing at least 60 to 70 percent aluminum.  Ball, rod, or hammer mills can be
used to reduce oxides and nonmetallic particles to  fine powders for ease of removal during
screening.
                                         4-99

-------
                                               PRETREATMENT
                                 OLD SHEET
                                 ». CASTINGS L
                                  CLIPPINGS
                                      J CRUSHING _
                                       (SCREENING
                                                        •BALING-
                                       ;NEW CLIPPINGS.
                                           FORCINGS
  BULK SCRAP-
 INSPECTION_
'  SORTING
                                    -> CABLE
                                    SHREDDING/
                                   "CLASSIFYING"

                                          FUEL—-
                                 ^BORINGS,
                                 "TURNINGS"
                               CRUSHING,
                              'SCREENING'
                               (SCC^fMXIJX,
 BURNING,
'  DRYING ~
 (sec vo4-oc:-oti
                                         HEAVY METALLIC
                                              SKIMS
                                            FUEL
                                  RESIDUES
                                    > HOT DROS£
                                     PROCESSING
                                      (SCC 3-04-001 J37,
                                                                     FLUX
                                                     • DRY  MILLING	>J	
                                                      (SCC 3-CH-OC'.-1C
                                           WATER
                                   FOIL
                                       LEACHING  	
                                       (SCC3-04O01 It
                                       J

                                      ~       —FUEL
                                         	I	
                                      	> ROASTING -
                                          (SCC 3*4-001 -!l)
 TREATED
ALUMINUM
  SCRAP
                                     .HIGH IRON_
                                        SCRAP
                                       -> SWEATING	
                                          (SCC J04-OC1-01I
                                         	A
                                                   FUEL
   Figure 4-18a.  Process Diagram for a Typical  Secondary Aluminum Processing Industry
                         (Source Classification Codes in parentheses)

Source:  Reference 59.
                                             4-100

-------
SMELTING/REFINING
                                                         PRODUCT
   TREATED
  ALUMINUM
    SCRAP
                                        CHLORINE
                                        FLUX
  REVERBERATORY
     (CHLORINE)
SMELTING/REFINING
     (SCO JMH-001-04)
                                        FLUOR.NE
                                        FLUX
                                      -FUEL
                                  YVT
  REVERBERATORY
     (FLUORINE)
SMELTING/REFINING
     (SCO 3-04-001 -05)
                                        FLUX
                                        FUEL
                              CRUCIBLE
                         SMELTING/REFINING
                              (SCC 3-04-001-02)
                                                                 ALLOY
NOTCHED
  BARS

  SHOT


   HOT
  METAL
                              INDUCTION
                         SMELTING/REFINING
                                        FLUX
                                        ELECTRICITY
       Figure 18-b. Process for a Typical Secondary Aluminum Processing Industry
                    (Source Classification Codes in parentheses)
Source: Reference 59.
                                    4-101

-------
              Pyrometallurgical Cleaning:  Pyrometallurgical techniques (called drying in the
industry) use heat to separate aluminum from contaminates and other metals.
Pyrometallurgical techniques include roasting and sweating.  The roasting process involves
heating aluminum scrap that contains organic contaminates in  rotary dryers to temperatures
•high enough to vaporize or carbonize organic contaminates,  but not high enough to melt
aluminum (1,220°F [660°C]).  An example of roasting is the APROS delacquering and
preheating process used during the processing of used beverage cans (shown in Figure 4-19).
The sweating process involves heating aluminum scrap containing other metals  in a sweat
furnace to temperatures above the melting temperature of aluminum, but below that of the
other  metal.59

              In addition to roasting and sweating, a catalytic technique may also be used to
clean  aluminum dross. Dross is a layer of impurities and semisolid flux that has been
skimmed from the surface of molten aluminum.  Aluminum may be recovered from dross by
batch  fluxing with a salt, cryolite mixture  in a mechanically rotated, refractory-lined barrel
furnace.  Cryolite acts as a catalyst that decreases aluminum surface tension and therefore
increases reccnery rates.  Aluminum is tapped periodically through  a hole in the base of the
furnace

             H\drometallurgical Cleaning. H\drometallurgical  techniques use water to
clear,  and process aluminum scrap.  H\drometallurgical techniques include  leaching and heavy
metal  separation  Leaching is used to reco\er aluminum from dross, furnace skimmings, and
slag  li requires wet milling, screening, drying,  and finally magnetic separation to  remove
fluxing salts and other waste products from the aluminum.

             The heavy metal separation hydrometallurgical process separates  high density
metal  from lo\\ density metal using  a viscous medium, such as copper and  iron, from
aluminum.  Heavy metal separation  has been used to concentrate  aluminum recovered  from
shredded cars.  The cars are shredded after  large aluminum components have been  removed
(shredded material contains approximately 30 percent aluminum)  and processed in heavy
media  to further concentrate aluminum to 80 percent or more.
                                         4-102

-------
   Scrap                                        Dust   Collector
   Aluminum                                    -  	
   Inlet                                                 .—
       Airlock
                                                     t

                                                  Fines  ^^^^

                                                        ^^^     Hot   Gas
                          Kiln           ',                        Recycle   Fan
                                  Discharge
                                  Airlock
                         Reverberotory   —          '

                              Furnace      Aluminum
                   Exhaust




    i e a * e a   Recycle Gas



                               Combustor
                                               Fuel
               Figure 4-19.  APROS Delacquering and Preheating Process


Source: Reference 59.


                                    4-103

-------
              Smelting/Refining-After scrap pretreatment, smelting and refining is
performed.  Smelting and refining in secondary aluminum recovery takes place primarily in
reverberatory furnaces.  These furnaces are brick-lined and constructed with a curved roof.
The term reverberatory is used because heat rising from ignited fuel is reflected (reverberated)
back down from the curved furnace roof and into the melted charge.  A typical reverberatory
furnace has an enclosed melt area where the flame heatsource operates directly above the
molten aluminum.  The furnace charging well is connected to the melt area by channels
through which molten aluminum is pumped from the melt area into the charging well.
Aluminum flows back into the melt section of the furnace under gravity.

             Most secondary aluminum recovery facilities use batch processing in smelting
and refining operations.59  It is common for one large melting reverberatory furnace to support
the flow requirements for two or more smaller holding  furnaces. The melting furnace is used
to melt the scrap, and remove impurities and entrained gases.  The molten  aluminum  is then
pumped into a holding furnace. Holding furnaces are better suited for final alloying,  and  for
making  any additional adjustments necessary to ensure that the aluminum meets product
specifications.  Pouring takes place from holding furnaces, either into molds or as feedstock
for continuous casters.

             Smelting and refining operations can involve the  following steps:  charging,
melting,  fluxing, demaggmg. degassing, alloying,  skimming, and pouring.

             The crucible smelting refining process is  used to melt small  batches of
aluminum scrap, generally limited  to 1.100 Ib (500 kg)  or less.  The metal-treating process
steps are essentially the same as those of reverberator) furnaces.

             The induction smelting and refining process is designed to produce aluminum
alloys with increased  strength and hardness by blending aluminum and hardening agents in an
electric  induction furnace.  The process steps include charging scrap, melting, adding  and
blending the hardening agent, skimming, pTmring. and casting into notched  bars.  Hardening
agents include manganese and silicon.
                                         4-104

-------
              Emissions

              The major sources of emissions from scrap prerreatment processes are scrap
driers, sweat furnaces, and UBC delacquering systems.60 Tables 4-24 and 4-25 present
CDD/CDF emission factors for two separate delacquering systems.  Control devices included
a venturi scrubber (Table 4-24) and multiple cyclones (Table 4-25).

              Controls—Mechanical cleaning techniques involve crushing, shredding, and
screening and  produce metallic and nonmetallic particulates.  Burning and drying operations
(pyrometallurgic techniques) emit particulates and organic vapors.  Emissions from
reverberatory furnaces represent a significant fraction of the total particulate and gaseous
effluent generated in the secondary aluminum industry.  Afterburners are frequently used to
incinerate unburned VOCs. Oxidized aluminum fines blown out of the dryer by the
combustion gases contain particulate  emissions.  Wet scrubbers or fabric  filters are sometimes
used in conjunction with afterburners.

              Mechanically generated dust from rotating barrel dross furnaces constitutes the
main air  emission of hot dross processing. Some fumes are produced from the fluxing
reactions.  Fugitive emissions are controlled by enclosing the barrel furnace in a hood system
and by ducting the emissions to a fabric filter.  Furnace offgas emissions, mainh fluxing salt
fume, are often controlled by a venturi  scrubber.

4.5.3          Secondary1 Lead Production

              In  1990, primary and secondary smelters  in the United States produced
1,380,000 tons (1,255,000 Mg) of lead.  Secondary lead smelters produced 948,000 tons
(860,000 Mg)  or about 69 percent of the total refined lead produced in 1990; primary smelters
produced 434.000 tons (395,000 Mg).62  Table 4-26 lists U.S. secondary  lead smelters
according to their annual lead production capacity.
                                         4-105

-------
TABLE 4-24. CDD/CDF EMISSION FACTORS FOR SECONDARY ALUMINUM
   SHREDDING AND DELACQUERING SYSTEM - SCRUBBER OUTLET
            CONTROL DEVICE - VENTURI SCRUBBER
                      SCC 3-04-001-09
                FACTOR QUALITY RATING: D

Pollutant
2,3,7,8-TCDD
Total TCDD
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,4,7.8-HxCDD
1,2,3.6.7,8-HxCDD
1,2.3.7,8,9-HxCDD
Total HxCDD
12^467 8-HpCDD
Total HpCDD
Total OCDD
2.3.-.S-TCDF
loiai TCDF
: 2 :• ~.S-PeCDF
: 3 4 ->-PeCDF
Iota' PtrCDF
I :.?.4.-.S-HxCDF
l.:.?.6~.S-HxCDF
1 : ~.~.S Q-HxCDF
2 3 4.6 ".S-HxCDF
Total HxCDF
1 2 ~ 4 6 7 8-HpCDF
12^478 9-HpCDF
Total HpCDF
Total OCDF
Total CDD
Total CDF
Source. Reference 60
3 Emission factors are Ib (k
Average Emission
Ib/ton
3.94x10"'
9.56x1 0'8
1.42xlO-8
1.28x10"
8.52x10''
1.06xlO'8
1.06xlO-s
1.56xlOT
5 78x10'*
1.17x10"
6.64xlOs
4 64\10 '
1 24\10'
6 ^6x10 '
960x10'
1 ITxRi-'
922x10'
922x10'
4 40x ) 0 '
7 80x10'
1.03x10'
2 44x10"
5 42x10*
4.94x10"
1.21x10"
562x10"
4.04x1 0'(1

g) of pollutant emitted per ton (Me) of aluminum produce^
Factor3
kg/Mg
1.97x10'"
4.78xlO'8
7.1 Ox 10-'
6.40x1 0'11
4.26x10"
5.30x1 0'"
5.30x10"
7.80x10'"
2.89x10"
5.85x10-'
3.32x10''
2.32x10'
6 20x10"
3 38x10 '
4 80x10 v
5 85x10"
461x10'
4.61x10'
2 20x10 '
3 90x 1 0 '
5 15x10"
1.22x10"
2.71xlOs
247x10"
6.05x10'
2.81x10'
202x10-'

d.
                          4-106

-------
   TABLE 4-25.  CDD/CDF EMISSION FACTORS FOR SECONDARY ALUMINUM
                SHREDDING AND DELACQUERING SYSTEM
                CONTROL DEVICE - MULTIPLE CYCLONES
                            SCC 3-04-001-09
                      FACTOR QUALITY RATING: D

Pollutant
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Total HxCDD
1,2,3,4,6,7,8-HpCDD
Total OCDD
2.3,7.8-TCDF
1.2,3.7,S-PeCDF
2.3.4.7.8-PeCDF
Total PeCDF
l.Vv4.7.S-H\CDF
1,2, 3. 6. "\S-H\CDF
1. \3.~.8.9-H\CDF
2.3.4.G.~.S-HxCDF
Total HxCDF
1. 2,3.4,6.7, 8-HpCDF
1,2,3.4,7,8,9-HpCDF
Total HpCDF
Total OCDF
Total CDD
Total CDF
Average
Ib/ton
1.69X10'9
7.28xlO'9
5.64x10-"
8.24x1 0'9
4.04x1 0'9
1.79xlO'8
3.86xlO-8
4.86xlO'8
9.68xlO-9
2.36xlO'8
4.66x10-"
".02xlO's
3.52xlO-s
3.38x10-'
2.70x10''"'
3.20x10-'
1.04x10-"
8.52x10'"
1.24x10''
9.76xlO'8
5.90x10-'
1.14xlO-7
3.40xlO'7
Emission Factora
kg/Mg
8.45x10-'°
3.64xlO-l)
2.82xlO'9
4.12x10-'
2.02x10-"
8.95x10-"
1.93x10-"
2.43x10-"
4.84x10-''
l.lSxlO'"
2.33x10"
3.51xl()-v
1 76x10"
1.69x10"
1.35x10"''
1.60xlC>-s
5.20x10"
4. 26x1 0'"
6.20x10'"
4.88x10'"
2.95x10-"
5.70x10-"
1.70x10-"
Source1 Reference 61.

a Emission factors are Ib (kg) of pollutant emitted per ton (Mg) of aluminum produced.
                                4-107

-------
                 TABLE 4-26. U.S. SECONDARY LEAD SMELTERS
 Smelter
      Location
 Small Capacity: less than 22,000 ton/yr (20,000 Mg/yr)
 Delatte Metals
 General Smelting and Refining Company
 Master Metals, Inc.
 Metals Control of Kansas
 Metals Control of Oklahoma

 Medium Capacity; 22.000 to 82,000 ton/yr (20,000 to
 75,000 Mg/yr)
 Doe Run Company
 East Perm Manufacturing Company
 Exide Corporation
 Exide Corporation
 GXB. Inc.
 GXB. Inc.
 Gulf Coast Recycling. Inc.
 Refined Metals Corporation
 Refined Metals Corporation
 RSR  Corporation
 RSR  Corporation
 Schinlkill Metals Corporation
 Tejas Resources. Inc

 Lar^e CapaciU:  greater than 82.000 tonyr (75.000 Mg yr)
 Gopher Smelting and Refining. Inc
 GNB. Inc.
 "RSR  Corporation
 Sanders Lead  Company
 Schuylkill Metals Corporation
Ponchatoula, LA
College Grove, TN
Cleveland, OH  .
Hillsboro, KS
Muskogee, OK
Boss, MO
Lyon Station. PA
Muncie. IN
Reading. PA
Columbus. GA
Frisco.  TX
Tampa,  FL
Beech Grove. IX
Memphis. TX
Cit>  of Industn. CA
Middletown. NY
Forest Cit\. MO
Terrell. TX
Eagen, MN
Vernon, CA
Indianapolis. IX
Troy. AL
Baton Rouee. LA
Source  Reference 62
                                      4-108

-------
              Process Description

              The secondary lead smelting industry produces elemental lead and lead alloys
by reclaiming lead, mainly from scrap automobile batteries. Blast, reverberatory, rotary, and
electric furnaces are used for smelting scrap lead and producing secondary lead. Smelting is
the reduction of lead compounds to elemental lead in a high-temperature furnace.  It requires
higher temperatures  (2,200 to 2,300°F [1,200 to 1,260°C])  than those required for  melting
elemental lead (621°F [327°C]).  Secondary lead may be refined to produce soft lead (which
is nearly pure lead) or alloyed to produce hard lead alloys.  Fifty percent of the lead produced
by secondary lead smelters is hard lead, and fifty percent is soft.  About 80 percent of all lead
in the United States  goes to producing new batteries.62

              Lead-acid batteries represent about 90 percent of the raw materials at a typical
secondary' lead smelter,  although this percentage may vary  from one  plant to the next.  These
batteries contain approximately 18 Ib (8.2 kg) of lead per battery' consisting of 40  percent lead
alloys and 60 percent lead oxide.   Other types of lead-bearing raw materials recycled by
secondary lead smelters include drosses (lead-containing byproducts of lead refining). v.hich
may be purchased from companies that perform lead alloying or refining but not smelting:
battery plant scrap, such as defective grids or paste: and scrap lead, such as old pipes or roof
flashing   Other scrap lead sources include cable sheathing, solder, and babbitt metal L-

              As illustrated in Figure 4-20. the normal sequence of operations in a secondary
lead smelter is scrap receiving, charge  preparation, furnace  smelting,  and lead refining and
alloying.  In all plants, scrap batteries are first sawed or broken open to remove the lead alloy
plates and lead oxide paste material. At blast furnace smelters,  a slow-speed saw is used to
remove the top of the case, the plates are  dumped from the case, and whole  grids are charged
to the furnace.  At other types of smelters, hammermills or other crushing/shredding devices
are used to break open the battery cases.   Float/sink separation systems are typically  used to
separate plastic batten' parts, lead terminals, lead oxide paste, and hard rubber used in older
batteries.  The majority  of lead smelters recover the crushed polypropylene plastic  materials
for recycling.  Hard  rubber materials are  usually charged to the furnace.
                                         4-109

-------

Polypropylene^
Plastic to "*
Recycling .,
AftM 4n
ACIO ID
Water Treatment
or Recycling

Other Lead-
and Scrap











Batteries Arrive
by Truck
I


Breaking
i
Grid Metal,
Hard Rubber,
Separators
1
Materials
Storage
?
Charge
Preparation
I
Smelting
Furnace
i
?
Refining/
Alloying
T
Casting
T
Finished
Product
OPTIONAL
i
i


\ <
Paste
Desulfurization


1 — 	




•* oldy ^* DlbpUbdl






        Figure 4-20.  Simplified Process Flow Diagram for Secondary Lead Smelting




Source: Reference 62.



                                         4-110

-------
              Paste desulfurization is an optional feed material processing step used by some
secondary lead smelters. It involves the separation of lead sulfate and lead oxide paste from
the lead grid metal, polypropylene plastic cases, separators, and hard rubber battery cases.
The next step is the chemical conversion of lead sulfate in the lead battery paste to lead oxide.
This process improves furnace  efficiency by reducing the need for fluxing agents to reduce
lead-sulfur compounds to lead metal.  The process also reduces SO2 furnace emissions.
However, S02 emissions reduction is usually a less important consideration because many
plants that perform paste desulfurization are also equipped with SO2 scrubbers.  About half of
all smelters perform paste desulfurization.62

              After removing the lead components from the  scrap batteries, the lead scrap is
combined with other charge materials such as refining drosses, flue dust, furnace slag, coke,
limestone,  and sand and fed to  either a reverberator/, blast, rotary or  electric smelting furnace.
Smelting furnaces are used to produce crude  lead bullion, which is refined and'or alloyed into
final lead products.

              Refining, the final step in secondary lead production, consists of removing
impurities and adding alloying  metals to the molten lead obtained  from the smelting furnaces
to meet a customer's specifications.  Refining kettles are used for the purifying and alloying
of molten lead.

              Blast and reverberator^ furnaces are currently  the most common types of
smelting furnaces in the industry, although some ne\\  plants  are  using rotary furnaces.  There
are approximately 15 reverberators' furnaces, 24 blast furnaces. 5 rotary furnaces, and
1 electric furnace in the secondary lead  industry'.62  The following discussion provides process
descriptions of these four types of secondary lead smelters.

              Reverberating Furnaces-A reverberator}' furnace as shown in Figure 4-21, is a
rectangular refractory-lined furnace. Reverberator/ furnaces are operated on a continuous
basis.  Natural gas- or  fuel oil-fired jets located at one end or at the sides of the  furnace are
used to heat the furnace and  charge material to an operating  temperature of about 2,000°F
                                          4-111

-------
                                                  o
                                                  u
                                                  ta
                                                   i
                                                  tn
                                                  t/i

                                                  O


                                                  CJ
                                                  fS
                                                  8
                                                  3
                                                         O
                                                         c
                                                         Si

                                                         o
                                                        V.
4-112

-------
(1,100°C). Oxygen enrichment may be used to decrease the combustion air requirements.
Reverberatory furnaces are maintained at negative pressure by an induced draft fan.

              Reverberatory furnace charge materials include battery grids  and paste, battery
plant scrap, rerun reverberatory furnace slag, flue dust, drosses, iron, silica, and coke.  A
typical charge over one hour may include 9.3 tons (8.4 Mg) of grids and paste to produce
6.2 tons (5.6 Mg) of lead.62

              Charge materials are often fed to a natural gas-  or oil-fired rotary drying kiln,
which dries the material before it reaches the reverberatory furnace.  The temperature of the
drying kiln is about 400°F (200°C), and the drying kiln exhaust is drawn directly into the
reverberatory furnace or ventilated to a control device. From the rotary drying kiln, the feed
is either dropped into the top of the furnace through a charging chute, or fed into the furnace
at fixed intervals with a hydraulic ram.  In furnaces that use a feed chute, a hydraulic ram is
often used as a stoker to move the material down the  furnace.62

              Reverberatory furnaces are used to produce a  soft (nearly pure) lead product
and a lead-bearing slag. This is done by controlling the reducing conditions in the furnace so
thai lead components  are reduced to metallic lead bullion and the alloying elements
(antimon>. tin. arsenic) in the battery grids, posts, straps, and connectors are oxidized and
removed in the slag.  The reduction of PbSO.. and PbO is promoted by the  carbon-containing
coke added to the charge material:

                             PbSO4  - C -» Pb - CO: - SO,

                                2PbO + C ->  2Pb  -  CO2

              The PbSO.4  and PbO also react with the alloying elements to  form lead bullion
and oxides of the alloying elements; the latter are removed in the slag.
                                         4-113

-------
              The molten lead collects in a pool at the lowest part of the hearth.  Slag
collects in a layer on top of this pool and retards further oxidation of the lead.  The slag is
made up of molten fluxing agents such as iron, silica, and lime, and typically has significant
quantities of lead.  Slag is usually tapped continuously and lead is tapped intermittently. The
slag is tapped into a crucible. The slag tap and crucible are hooded  and vented  to a control
device. Reverberatory furnace slag usually has a high lead .content (as much as  70 percent by
weight) and is used as feed material in a  blast or electric furnace to recover the  lead content.
Reverberatory furnace slag may also be rerun through the reverberatory furnace  during special
slag campaigns before being sent to a blast or electric furnace.  Lead may be tapped into a
crucible or  directly into a holding kettle.  The lead tap is usually hooded and vented to  a
control device.62

              Blast Furnaces—A blast furnace as shown in Figure 4-22, is a vertical furnace
that  consists of a crucible with a vertical  cylinder affixed to  the top.  The crucible is
refractory-lined and the vertical  cylinder consists of a steel water jacket  Oxygen-enriched
combustion air is introduced into the furnace through tuyeres located around the base of the
cylinder

              Charge materials are pre-weighed to ensure the proper mixture and then
introduced into the top  of the cylinder using a skip hoist, a conveyor, or a front-end loader.
The-  charge  fills nearh  the entire cylinder.  Charge material is added periodical]},' to keep the
le\e! of the charge at a consistent  working height  while lead and slag are tapped from the
crucible   Coke is added to the charge as  the primary fuel, although natural gas jets ma\ be
used to start the combustion process.  Combustion is self-sustaining as  long as there is
sufficient  coke in the charge material.  Combustion  occurs in the layer  of the charge nearest
the tuyeres.

              At plants that operate only blast furnaces, the  lead-bearing charge materials
may include broken battery components, drosses from the refining kettles, agglomerated flue
dust, and lead-bearing slag. A typical charge over one hour may  include 4.8 tons  (4.4 Mg) of
grids and paste.  0.3 tons (0.3 Mg) of coke, 0.1 tons (0.1 Mg) of calcium  carbonate, 0.07 tons
                                          4-114

-------
                                                  Charge Hopper
                                                              Exhaust Offtake to Afterburner
                      Charge
         Cool Water
          Hot Water
          Cool Water
  Lead Spout
 Lead Well and Siphon
                                                                            Working Height
                                                                               of Charge
                                                                              2.4 to 3.0 m
                                                                     Average Level of Charge
                                       Diameter at Tuyeres
                                       •—68 to 120 cm —•
                                                                        Slag Spout

                                                                 Drain Tap       Lead
                    Figure 4-22.  Cross-section of a Typical Blast Furnace
Source: Reference 62.
                                             4-115

-------
(0.06 Mg) of silica, 0'.5 tons (0.4 Mg) of cast iron, and 0.2 tons (0.2 Mg) of rerun blast
furnace slag, to produce 3.7 tons (3.3 Mg) of lead. At plants that also have a reverberatory
furnace, the charge materials will also include lead-bearing reverberatory furnace slag.62

              Blast furnaces are designed and operated to produce a hard (high alloy content)
lead product by achieving more reducing furnace conditions than those typically found in a
Teverberatory furnace. Fluxing agents include iron, soda ash, limestone, and silica (sand).
The oxidation of the  iron, limestone, and silica promotes the reduction of lead compounds and
prevents oxidation of the lead and other metals. The soda ash enhances the reaction of PbSO4
and PbO with carbon from the coke to reduce these compounds to lead metal.

              Lead tapped from a blast furnace has a higher content of alloying metals (up to
25 percent) than lead produced by a reverberatory furnace.  In addition, much less of the lead
and alloying  metals are oxidized and removed in the slag, so the slag has a low metal content
(e.g., 1 to 3 percent)  and frequently qualifies as a  nonhazardous solid waste.

              Because air is introduced into the blast furnace at the tuyeres,  blast furnaces are
operated at positive pressure.  The operating temperature at  the combustion layer of the
charge is between 2.200 and 2.600°F (1.200 and 1.400CC). but the temperature of the  gases
exiting the top  of the charge material is onl\  between "50 and 950°F (400 and 500°O

              Molten lead collects in the crucible  beneath a layer of molten slag  As  in a
re\ erberatory furnace, the slag inhibits the further  oxidation of the molten metal. Lead is
tapped continuously and slag is tapped intermittently, slightly before it reaches the level of the
tuyeres  If the tuyeres become blocked with  slag,  they  are manually or automatically
"punched"  to clear the slag.  A sight glass on the tuyeres allows the furnace operator to
monitor the slag level and ensure that they are  clear of slag.  At most facilities,  the slag tap is
temporarily sealed with a clay plug, which is driven out to begin the flow of slag from the tap
into  a crucible.  The slag tap and crucible are enclosed  in a hood, which is vented to a control
device.
                                          4-116

-------
              A weir dam and siphon in the furnace are used to remove the lead from
beneath the slag layer.  Lead is tapped from a blast furnace into either a crucible or directly to
a refining kettle designated as a holding kettle. The lead in the holding kettle is kept molten
before being pumped to a refining kettle for refining and alloying.  The lead tap on a blast
furnace is hooded and vented to a control device.

              Rotary Furnaces—As noted above, rotary furnaces (sometimes referred to as
rotary reverberatory furnaces)  (Figure 4-23) are used at only  a few recently constructed
secondary lead smelters in the United States.  Rotary furnaces have two advantages over other
furnace types:  it is easier to adjust the relative amount of fluxing agents because the furnaces
are operated on a batch rather than a continuous basis, and they achieve better mixing of the
charge materials than do blast or reverberatory furnaces.62

              A rotary  furnace consists of a refractory-lined  steel drum mounted on rollers.
Variable-speed motors are  used to rotate the drum.  An oxygen-enriched natural gas or  fuel
oil jet at one end of the furnace heats the charge material and the refractory lining of the
drum. The connection to the flue  is located at the same end  as the jet.  A sliding door  at the
end of the  furnace opposite from the jet allows charging of material to the furnace.  Charge
materials are typically placed in the furnace using a retractable conveyor or charge bucket.
although other methods are possible.

              Lead-bearing raw materials charged to rotary furnaces include broken batten-
components, flue dust, and drosses. Rotary furnaces can use  the  same lead-bearing raw
materials as reverberatory furnaces, but they produce slag that is relatively free of lead, less
than 2 percent. As a result, a blast furnace is not needed for recovering  lead from the slag.
which may be disposed of as a nonhazardous waste.

              Fluxing agents for rotary furnaces may include iron, silica, soda ash, limestone,
and coke.  The fluxing  agents are  added  to promote the conversion of lead compounds to lead
metal.  Coke is used as a reducing agent rather than as a primary fuel. A typical charge may
consist  of 12 tons (11 Mg) of wet battery scrap, 0.8 tons (0.7 Mg) of soda ash, 0.6 tons
                                          4-117

-------
                                                                  Hygiene Hood
                                               Rotary Furnace Shell       /
                                Drive Tram
              Figure 4-23.  Side-view of a Typical Rotary Reverbatory Furnace
Source: Reference 62.
                                            4-118

-------
(0.5 Mg) of coke, and 0.6 tons (0.5 Mg) of iron. This charge will yield approximately 9 tons
(8 Mg) of lead product.62

              The lead produced by rotary furnaces is a semi-soft lead with an antimony
content somewhere between that of lead from reverberatory and blast furnaces.  Lead and slag
are tapped from the furnace at the conclusion of the smelting cycle.  Each batch takes 5 to
12 hours to process, depending on the size of the furnace.  Like reverberatory furnaces, rotary
furnaces  are operated at a slightly negative pressure.

              Electric Furnaces-An electric furnace consists of a large, steel, kettle-shaped
container that is refractory-lined (Figure 4-24).  A cathode extends downward into the
container and an anode is located in the bottom  of the container. Second-run reverberatory
furnace slag is charged into the top of the furnace.  Lead and slag are tapped  from the bottom
and side of the furnace, respectively.  A fume hood covering the top of the furnace is vented
to a control device.

              In an electric furnace, electric current flows  from the cathode to the anode
through the scrap charge.  The electrical resistance of the charge causes the charge to heat up
and become molten.  There is no combustion process involved  in an electric  furnace11".

              There is only one electric  furnace in operation in the U.S. secondary lead
industry.  It is used to process second-run reverberatory furnace slag, and it fulfills the  same
role as a blast furnace used in conjunction with  a reverberatory furnace.  However, the
electric furnace has two advantages over a blast  furnace. First, because there  are no
combustion gases, ventilation  requirements are much lower than for blast  or reverberatory
furnaces, and  the potential  for formation of organics is greatly reduced.  Second, the electric
furnace is extremely reducing, and produces  a glass-like, nearly lead-free  slag that is
nonhazardous.
                                          4-119

-------
                                                       I
fS
-------
              Emission Control Techniques

              Controls used to reduce organic emissions from smelting furnaces in the
secondary lead smelting industry include afterburners on blast furnaces and combined blast
and reverberatory exhausts.  Reverberatory  and rotary furnaces have minimal dioxin/furan
emissions because of high exhaust temperatures and turbulence, which promote complete
combustion of organics.  No controls for total hydrocarbons (THC) are necessary for these
process configurations.62

              CDD/CDF emissions  from blast furnaces are dependent on the type of add-on
control used.  An afterburner operated at 1,300°F (700°C) achieves about 84 percent
destruction efficiency of total hydrocarbons (THC).62 Several facilities with blast and
reverberatory furnaces combine the exhaust streams and vent the combined stream to an
afterburner. The higher operating temperature of the reverberatory furnace reduces the  fuel
needs of the afterburner so that the afterburner is essentially "idling." Any temperature
increase measured across the afterburner is  due to the heating value of organic compounds in
the blast furnace exhaust.  A combined reverberatory and blast furnace exhaust stream ducted
to an  afterburner with an exit temperature of  1.700°F (930°C) can achieve 99 percent
destruction efficiency for THC fc:

              Additional controls used by  secondary lead smelters include baghouses for
paniculate and metal control, hooding and  ventilation to  a baghouse for process fugimes. and
scrubbers for HC1  and SO: control L~

              Emission Factors

              Process emissions (i.e., those emitted from the smelting furnace's main exhaust)
contain metals, organics (including dioxins/furans),  hydrogen chloride (HC1). and chlorine
(Cl;).   Process emissions also contain criteria  pollutants,  including PM, VOCs, CO, and SO:.
The primary source of CDD/CDF at secondary lead smelters is PVC used as separators  in
lead-acid batteries.  The Battery Council International (BCI) recently provided EPA with data
                                         4-121
-------
(gathered in 1995) showing that less than one-tenth of one percent of U.S.-manufactured
batteries were found to contain PVC separators.63  It is important to note here that BCI also
reports that no U.S. manufacturer of lead-acid automotive batteries currently uses PVC in
production.63

              Blast furnaces are greater sources of dioxin/furan emissions than reverberatory
or rotary furnaces.  Low exhaust temperatures from the charge column [about 800°F (430°C)}
result in the formation of products of incomplete combustion (PIC) from the  organic material
in the feed material. Uncontrolled THC emissions (which  correlate closely with organic
pollutant emissions) from a typical 55,000-ton/yr (50,000 Mg/yr) blast furnace are about 309
tons/yr (280 Mg/yr).62

              The EPA does not have sufficient data to  link dioxin/furan emissions to specific
control technologies currently in use in the industry.02  Rotary and reverberatory furnaces have
much higher exhaust temperatures than blast furnaces, about 1,800 to 2,200°F (980  to
1.200°C). and much lower  THC emissions because of more complete combustion. Total
hydrocarbon emissions from a typical rotary furnace (16.500 toiyyr [15,000 Mg'yr]  capacityj
are about  38 ton yr (34 Mg/yr).  The majority of these emissions occur during furnace
charging,  when the furnace's burner is cut back and the temperature is reduced.  Emissions
drop off sharply when charging is completed  and the  furnace is brought to normal operating
temperature.0"  CDD CDF emissions from reverberatory furnaces  are even lower than those
from rotary furnaces because reverberatory furnaces are operated  continuously rather than  on a
batch basis

              Three test  reports from three secondary lead smelters were used to develop
CDD CDF emission factors.64"66 All testing was conducted in  support of the EPA's  Secondary
Lead National Emission Standards  for Hazardous Air Pollutants (NESHAP) program. The
three facilities tested represent the following process configurations:  a rotary smelting furnace
equipped with a baghouse and S02 scrubber; a blast furnace equipped with an afterburner,
baghouse. and SO: scrubber; and a reverberatory  and  blast  furnace with exhaust from each
furnace combined prior to a single afterburner, baghouse, and  S02 scrubber.
                                         4-122
-------
              Emissions were measured at all three facilities, and emission factors were
derived from the test reports of the three facilities representing the three principal furnace
types in use. These emission factors represent emissions with a baghouse and scrubber, and
are shown in Table 4-27.  However, the effect of these controls on CDD/CDF emissions is
unclear.

4.5.4         Scrap Metal Incinerators

              Scrap metal incinerators are used to bum off combustible contaminants
(e.g., plastics, rubber, paper, oils) contained in scrap metal.  This process renders a cleaner
metal scrap  that can be further processed into a refined, saleable metal product.   Scrap metal
incinerators  operate in an  oxidizing atmosphere, as opposed to metal smelters, which operate
in a reducing atmosphere.  The purpose of a scrap metal incinerator is simply to burn off
contaminants prior to smelting.

              Process  Description

              Many types of scrap materials are processed in incinerators prior  to smelting.
including wire and cable,  drained transformer cores, automobile bodies, electric  motors, and
various other types  of metal-bearing scrap. The combustible portion of scrap metal comprises
a great vaneu of materials, including rubber, paper, cotton,  asphalt-impregnated fabrics, silk.
and plastics  such as polyethylene, polypropylene and PYC.  Additionally, the metals
themselves may have baked-on coatings of plastic, paint, or  varnish.  The chlorine present in
PVC wire insulation or automobile parts and other sources of chlorinated organic materials
provide sufficient chlorine to produce CDD/CDF from  the combustion of these  materials.6S

              Figure 4-25 shows a process flow diagram of a scrap metal reclamation
incinerator.  There are many different designs of scrap  metal reclamation incinerators;
however, there are some  commonalities.  A typical scrap metal reclamation incinerator
consists of one or more chambers and an afterburner connected  to a stack.  The older designs
are normally limited to a  single primary or charging chamber and afterburner. Newer designs
                                         4-123
-------
  TABLE 4-27.  CDD/CDF EMISSION FACTORS FOR SECONDARY LEAD SMELTING

                             SCC 3-04-004-02, -03, -04
                          FACTOR QUALITY RATING:  D
                                               Emission Factor
                                               Ib/ton (kg/Mg)a
 Isomer
  Baghouse Outlet
                   Scrubber Outlet
ROTARY FURNACE (3-04-004-04)
      2,3,7,8-TCDD           3.
      2,3,7,8-TCDF
      2,3,7,8-TCDD TEQ
      Total CDD
      Total CDF
                                2.00x1O'9 (1.
                                1.42xlO-9(7.
                                1.49xlO-8(7.
                                5.16xlO'8 (2,
           .58x10-'°)
           OOxl O'9)
           lOxlO'10)
           45x10-9)
           58xlO'8)
                3.96xlO-10(1.98xlO-10)
                 2.00x10-9(1.00xlO-")
                1.21X10'10 (6.05x10-")
                 1.85x10 "(9.25x10-'°)
                 5.16xlO-8(2.58xlO'8)
 BLAST FURNACE (3-04-004-03)
        2,3J,8-TCDD
        2,3,7.8-TCDF
        2,3,7.8-TCDD TEQ
        Total CDD
        Total CDF
4.46x10'9
1.85x10-'
1.76xlO's
2.94x10'"
5.10x10"
.23x10-9)
25x10-")
.80x10-'O
,47x10")
55x10'")
                                                         5.38x10''° (2.69x10-'°)
                                                          1.97x10'" (9.85x10-'°)
                                                          1.68xlO''; (8.40x10'10)
                                                          2.26x10'" (1.13x10-*)
                                                          4.74x10'' (2.37x10-')
 BLAST 'REVERB FURNACE (3-04-004-02)
       2.3.-.S-TCDD
       2.3.~.8-TCDF
       2.3.~.8-TCDD TEQ
       Total CDD
       Total CDF
1.4SxlO'K'
 834x10'°
 2.68x10"
 1.12x10'-
 •766x10-'
40x10'")
6"xlO'J)
.34x10'")
,60x10")
83x10'')
                                                         1.75x10'"- (S.T5xlO'!1)
                                                          2.88xlO''J (1.44x10'")
                                                         8.14x10'° (407x10-'")
                                                          1.42x10"' (7.10x10'")
                                                          3.16x10'" (1.58x10'")
Souuc References 64-6"

a Emission factors are in Ib (kgI of pollutant emitted per ton (Mg) of lead produced
                                      4-124
-------
      Scrap Metal
 Supplemental Fuel

   (Natural Gas)




Bume
toSn
d Scrap
netting
Open Air Cooling
Stack
t
Primary
Char
1
i
Burned Scrap
f-umace
Exhaust
Furnace Gas ^Settling \ __ After Burner


i
t
Settling Chamber
Ash
                             Primary Chamber
                                  Ash
Natural Gas
                 Figure 4-25. Scrap Metal Incinerator Process Flow Diagram


Source: Reference 69.


                                            4-125
-------
generally incorporate a secondary or settling chamber prior to the afterburner.  The designs
also differ in the placement of burners and use of water sprays for quenching.69

              A typical scrap reclamation incinerator is operated in batch mode 8 hours per
day, 5 days per week.  However, operation is variable and largely dependent on scrap
availability. At  the beginning of a batch, a charge of scrap material is placed in the primary
chamber and is ignited using  paper or the primary chamber burner,  if one exists.  Gases from
the primary chamber flow through the secondary chamber, where some settling of large
particulate occurs,  and then to the afterburner, where the flue gases are heated to 1,800° to
2,000°F (980° to 1090°C) to  control emissions prior to discharge to the atmosphere.  Natural
gas is typically used as the auxiliary fuel for a scrap incinerator; however, liquid propane or
No. 2 fuel oil  can  be used.

              Most incinerators operate with very little or no instrumentation to measure
temperature or control draft and oxygen level.  Combustion conditions can be controlled by
varying the amount of air allowed into the primary chamber during combustion.  The amount
of air is controlled by opening or closing the doors and the draft registers.  The primary
chamber temperature can go as high  as 800 to  1.200°F  (42~ to 649°C) when an auxiliary
burner is used  However, many operators restrict the temperatures and amount ofox\gen in
ordei  10 increase yield c"

              Emission  Control Techniques

              Most scrap metal incinerators use afterburners to complete the combustion of
the exhaust gases,  thereby controlling emissions of PM  and gaseous organic  compounds.
These afterburners are typically fired with natural gas,  and temperatures of 1,800 to 2.000°F
(980 to 1,090°C) are achieved.  Some scrap metal incinerators may  be equipped with
paniculate collection devices  such as fabric filters, but most have no additional controls other
than the afterburner.os
                                         4-126
-------
             Emission Factors

             Emission factors were identified for a scrap metal incinerator that burns
combustibles from scrap wire and drained transformer cores.68  The scrap wire burned at the
facility contains some PVC plastic insulation, and the drained transformer cores contain
transformer oil  residues containing less than 500 ppm of PCBs.

             Temperatures in the primary chamber furnace during incineration were about
1,050°F (570°C).  The facility is equipped with a natural gas-fired afterburner that achieves
temperatures  of 1,800 to 2,000°F (980 to  1,090°C).68

             Emissions of TCDD/TCDF through OCDD/OCDF were measured in the stack
after the afterburner. Table 4-28 presents emissions on a flue gas-concentration basis and as
emission factors.  Emission factors are based on  the total weight of wire and transformer
scrap feed to the furnace.  The HpCDD/HpCDF  and OCDD/OCDF were the primary species
present, but measurable quantities of the TCDD/TCDF through HxCDD/HxCDF were also
present.

4 5.5         Drum and Barrel Reclamation Furnaces

             Drum and barrel reclamation  facilities recondition used steel drums for resale
Combustion  is used to remove drum paints,  interior linings, labels, residual liquids in the
drum, and other contaminants. Residual materials include organic solvents,  inks, paints, food.
and a variety of other products.

             Process Description

             Figure 4-26 shows a flow diagram of a typical drum reclamation facility. Most
facilities use  a tunnel furnace to bum contaminants. The tunnel furnace is equipped with
multiple natural gas burners on each side.  Dirty drums are loaded onto a conveyor and
conveyed to the furnace.  Before entering  the furnace, any free contents in the drums are
                                         4-127
-------
  TABLE 4-28  CDD/CDF FLUE GAS CONCENTRATIONS AND EMISSION FACTORS
             FOR A SCRAP WIRE AND TRANSFORMER INCINERATOR
                                 SCC 3-04-900-13
                         FACTOR QUALITY RATING:  D
 Isomer
       Flue Gas
     Concentration
Ib/ft3 (fig/dscm) at 3% 02a
    Emission Factor
Ib/ton (|ig/kg) scrap feed
 DIOXINS
       2,3,7,8-TCDD
       Total Other TCDD
       Total PeCDD
       Total HxCDD
       Total HpCDD
       Total OCDD
       Total CDD

 FURAXS
       2.3.~.8-TCDF
       Total Other TCDF
       Total PeCDF
       Total HxCDF
       Total HpCDF
       Total OCDF
       Total CDF

 TOTAL CDD/CDF
  7.86x10-° (1.26x10'')
    9.37xlO-5(1.50)
    3.22xlO'4(5.15)
    2.05x10° (32.8)
     1.04x10°- (167)
     3.11x10° (498)
     4.40xlO-2 (705)
  5.70x10'' (9.13x10"
    2 80x10-' (44.8)
    2.62x10-"' (42.0)
    5.92x10° (94.8)
     1.83xlO-: (293)
    244x10° (390)
    541x10- (866)

    9.81x10- (1,571)
 7.47x10-'° (3.74X10-4)
  8.10x10-" (4.05x10°)
  2.74xlO'8 (1.37x10°)
  1.42xlO-7(7.11xlO-2)
  6.94xlO'7 (3.47xlO'!)
    2.00x10-° (1.0)
    2.88x10-° (1.44)
  5.34x10'" (2.67x10°)
  2.08x10'' (1.04x10-)
  1.95x10-" (9."4xlCr2)
  4.06x10-" (2 03x10 ')
  1.25x10-° (6.23x10-)
  1.61x10-° (8 07x10-')
    3.68xlO'(' (1.84)

    6.56x10-° (3.28)
Source Reference 68
Note  The composition and combustible portion of the scrap metal was not stated in this report
a Emissions measured in the stack gas after an afterburner.
                                      4-128
-------
Dirty Drums
                   Gas-

                    Air-
Burning Furnace
                                      Furnace
                                       Ash
                               Furnace Rue
                                  Gas
                         Afterburner
                         Afterburner
                            Ash
                             To Atmosphere



                                 Stack !
                  Figure 4-26.  Drum and Barrel Incinerator Process Flow Diagram


    Source: Reference 69.


                                               4-129
-------
drained into a collection vessel.  The drums pass through the furnace, where temperatures
reach about 1,000°F (540°C), and are air-cooled as they exit the furnace.69  After cooling, the
drums are shotblasted with an abrasive to clean the drum to bare metal.  The drums are then
repainted prior to sale.

              Emission Control Techniques

              Natural gas-fired afterburners are typically used to combust unburned
hydrocarbons in the exhaust gases from the furnace, thereby controlling emissions of gaseous
organic compounds.

              Emission Factors

              Emission factors were identified for a drum and barrel reclamation furnace that
processes drums previously containing  lacquer, organic solvents, inks, enamel-type paints and
other materials69 The residue in the drums was analyzed for total organic hahdes (TOX).
The  TOX  content of the residue in the drums during testing was about 800  ppm.

              The facility is equipped \\ith an afterburner.  The afterburner operated at an
a\erage of 1.500~F (827°C) during testing.  Emissions of TCDDTCDF through
OCDD OCDF \\ere measured both before and after the afterburner.  Table 4-29 presents
emissions  on a flue gas-concentration basis  and as emission factors  Emission factors are on a
per-drum basis (55-gallon drum).  These data sho\\ that  the afterburner achieved greater than
95 percent  control of CDD/CDF  emissions

              Source Locations

              Approximately 2.8 to 6.4 million 55-gallon drums  are reconditioned annually in
the United  States.4  This estimate is based on the assumptions that there are 23  to
26 incinerators currently in operation, each  incinerator handles 500 to 1,000 drums per day,
and each incinerator  operates 5 days a week with  14 days down time for maintenance.4"
                                         4-130
-------
 TABLE 4-29. CDD/CDF EMISSION CONCENTRATIONS AND EMISSION FACTORS
            FOR A DRUM AND BARREL RECLAMATION FACILITY

                              SCC 3-09-025-01
                       FACTOR QUALITY RATING:  D
Isomer
DIOXINS
2,3,7,8-TCDD
Total Other TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
FURANS
2.3.7, 8-TCDF
Total Other TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
TOTAL CDD/CDF
Before Afterburner
Flue Gas
Concentration Emission Factor
Ib/ft3 (ug/dscm) Ib/drum
at 3% 02 (ug/drum)a

1. 024x1 0'9 (16.4)
4.78x1 0'9 (76.6)
6.49x10'' (104)
8.49x10-' (136)
1.66X10'8 (266)
5.43x10'' (86.9)
4.29x10-* (687)

3.90xlO'9 (62.5)
S.SlxlO'8 (930)
3.89xlO-s (610)
1.46xlO's (234)
1.60xlO'8 (256)
4.61x10-° (73.8)
1.35x10" (2,170)
1.78x10-" (2.857)

4.61xlO-'° (0.209)
2.01x10-' (0.912)
2.69x10"' (1.22)
3.29x10"' (1.49)
8.11x10-' (3.68)
2.78x10'' (1.26)
1.93xlO-8 (8.78)

1.67x10-" (0.756)
2.38xlO'8 (108)
1 62xlO-8 (7.34)
5.67xl09 (2.57)
7.74x10-' (3.51)
2.31xlO'9 (1.05)
5.73xlO'8 (26.0)
7.67xlO'8 (34.8)
After Afterburner
Flue Gas
Concentration Emission Factor
Ib/ft3 (Mg/dscm) Ib/drum
at 3% O2 (ng/drum)a

3.22xlO'12 (0.0516)
7.43x10'" (1.19)
4.49x10-" (0.719)
4.95x10'" (0.793)
8.18x10-" (1.31)
5.74x10'" (0.919)
3.11xlO-'°(4.98)

5.60x10-" (0.897)
8.93x10-'° (14.3)
3.87x10-'° (6.2)
1.87x10-'° (2.99)
1.26x10-'° (2.02)
343x10-" (0.549)
1 69x10-' (27.0)
2.00x10-' (32.0)

4.61xlO'12 (2.09xlO-5)
1.06x10-'° (0.0482)
6.44x10'" (0.0292)
7.10x10-" (0.0322)
l.lSxlO'10 (0.0534)
8.27x10'" (0.0375)
4.48x10-'° (0.203)

8.05x10-" (0.0365)
1.29x10-' (0586)
5.58x10-'° (0.253)
2.69x10-'° (0.122)
1.81x10-'° (00822)
4.94x10'" (0.0224)
242x10-' (1 10)
2.87xlO-9 (1.3)
Source: Reference 69.

Note:  These emissions would originate from drums previously storing chlorine-containing lacquers, solvent, etc

a Based on a 55-gallon drum.
                                   4-131
-------
Exact locations of the incinerators were not confirmed at the time this document was
developed.

4.6           PULP AND PAPER PRODUCTION - KRAFT RECOVERY BOILERS

              Chemical wood pulping involves the extraction of cellulose from wood by
dissolving the lignin that binds the cellulose together.  Kraft pulping is the major form of
chemical wood pulping in the United States, accounting for over 80 percent of the chemically
produced pulp, and is expected to continue as the dominant pulp process.70'71  The following
sections focus on the pulp mill thermal chemical recovery processes associated with
CDD/CDF emissions.

              Black liquor is a digestion byproduct of the kraft pulping process that consists
of soluble lignin and cooking chemicals.  Concentrated black liquor is fired in a recovery
furnace primarily to recover  inorganic chemicals for reuse in the kraft process and.
secondarily, to provide heat for process steam.  Relative to other sources, particular!},' waste
incineration, the combustion  of black liquor has minimal potential for  CDD CDF emissions. :

4 o 1          Process Description

              The kraft  pulping process mvohes the cooking or digesting of  wood chips at an
ele\ated temperature 340 to 360°F (about 175°C) and pressure (100 to  135 psig) in "\\hite
liquor." \\hich is a \\ater solution of sodium sulfide (Na;S) and  sodium hydroxide  (NaOH).
The lignin that binds the cellulose fibers together is chemically  dissolved  b> the white liquor
in a digester   This process breaks the wood into soluble lignin  and alkali-soluble
hemicellulose and  insoluble cellulose or pulp.  A typical kraft sulfite pulping  and recovery
process is shown in Figure 4-27.

              Two types of digester systems are used in chemical pulping: batch and
continuous.  In a batch digester, the  contents of the digester  are transferred to an atmospheric
tank, usually referred to  as a blow tank, after cooking is completed (2 to  6 hours).  In a
                                         4-132
-------
                                              u
                                              o
                                              u

                                              o
                                             •o

                                              ca

                                              BO
                                              a.
                                              re
                                              u
                                             CM
                                                    c
                                                    u
                                                    o
4-133
-------
continuous digester,- wood chips and white liquor continuously enter the system from the top
while pulp is continuously withdrawn from the bottom into a blow tank.  In both types of
digesters, the entire contents of the blow tank are diluted and pumped to a series of brown-
stock washers, where the spent cooking liquor (called black liquor) is separated from the pulp.
The pulp, which may then be bleached, is pressed and dried into the finished product.
                                                                                    *.
              The balance of the kraft process is designed to recover the  cooking chemicals
and heat.  The diluted spent cooking liquor, or weak black liquor, which is 12 to 18 percent
dissolved solids, is extracted from the brownstock washers and concentrated in a multiple-
effect evaporator system to about 55 percent solids.   The liquor is then further concentrated to
65 percent solids (strong black liquor) in a direct contact evaporator (DCE) or a nondirect
contact evaporator (NDCE), depending on the configuration of the recovery furnace in which
the liquor is combusted.  (DCE and NDCE recovery furnace schematics are shown  in Figures
4-28 and 4-29. respectively.)

             In older recovery furnaces, the  furnace's hot combustion gases concentrate  the
black liquor in a DCE prior to combustion.  NDCEs include most furnaces built since  the
earlx  1970s and modified older furnaces that have incorporated recovery systems that
eliminate the conventional direct contact evaporators  These NDCEs  use a concentrator rather
than a DCE to concentrate the black liquor prior to  combustion   In another type of NDCE
SNstem. the multiple  effect evaporator system  is extended to replace the direct contact system.

             The strong black liquor is sprayed into a recovery furnace with air control to
create both reducing and oxidizing zones within the  furnace chamber.  The combustion of the
organics dissolved  in the black liquor provides heat  for generating process steam and. more
importantly, reduces sodium sulfate to sodium  sulfide to be reused in the cooking process.
Sodium sulfate. which constitutes the bulk of the particulates in the furnace flue gas. may be
recovered by an ESP and recycled.  After combustion, most of the inorganic  chemicals
present in the black liquor collect as a molten smelt  in the form of sodium carbonate
                                         4-134
-------
                                               1
tn
CL
                                             ra
                                          ± ID
                                          Q
                                                                          §
                                                                          u
                                                                          o
                                                                          o
                                                                          2
                                                                          o
                                                                          I
                                                                          o
                                                                          ce
                                                                          •*-<
                                                                          O
                                                                          U
                                                                          *-•
                                                                          o
                                                                          oo
                                                                          (N
                                                                           00
                                                                                 o
                                                                                 f-
                                                                                 u
                                                                                 
-------


is
1:1
15
UJ Q-
i
«n ?
CO t
O
1 I-
| »
O
0

*l
1J
1
CD
i
V)
'S
CO


h
o co
•*•«§•
Q> 5
%UJ
^C


1
)H
w i
2
8
cc
CC
~1
'
4
10
^O
"o
CO
r
^A
ffl
w
1
                         £
                         r
                         on

                         Q
                         CJ
                         2
                         c
                         c
                         U
                         o

                         •5
                         c
                         2
                         •-


                         3
rrr
                             o
                             r^
                             u
.t
<
       0
       CO
4-136
-------
       j) and sodium sulfide at the bottom of the furnace, where they are continuously
withdrawn into a smelt-dissolving tank.

             CDD/CDF emissions from black liquor combustion will be affected by furnace
emission control devices as well as recovery process operating characteristics, furnace design
and operation, and the characteristics of the black liquor feed.  Furnace design and operation
affect combustion efficiency, which is  inversely related to CDD/CDF emissions.  The black
liquor recovery process determines the concentration of solids in the black liquor feed.  Feeds
containing a greater concentration of organic compounds  will exhibit better combustion
properties.  Preliminary emissions test  results from kraft recovery  furnaces indicate that the
CDD/CDF levels range from extremely low to nondetectable.9'41

             Organic and inorganic chlorine inputs to the black liquor circuit increase the
probability of CDD/CDF formation. The two primary sources of  chlorine that enter the black
liquor circuit are caustic makeup and mill water.72

             In addition to straight kraft process liquor, semi-chemical pulping process spent
liquor, known as brown liquor, may also be recovered in kraft recovery furnaces.  The  semi-
chemical pulping process is a combination  of chemical and mechanical pulping processes that
was developed to produce high-yield chemical pulps.  In  the semi-chemical process,  wood
chips are partially digested with cooking chemicals  to weaken  the bonds between the lignin
and the wood. Oversize particles are removed  from the softened wood chips and the chips are
mechanically reduced to pulp by grinding them in a refiner.

              The most common type of semi-chemical pulping is referred to as neutral
sulfite semi-chemical (NSSC). The major  difference between  the semi-chemical process and
kraft/sulfite pulping process is that the semi-chemical digestion process is shorter and wood
chips are only partially delignified.  Based on a survey conducted by EPA in 1993 under the
pulp and paper industry MACT standard development program, no U.S. semi-chemical mills
currently practice chemical recovery.  However, some semi-chemical pulp mills are, as of
                                         4-137
-------
 1997, using chemical recovery.73  Also, some mills combine spent liquor from on-site
 semi-chemical process with spent liquor from adjacent kraft process for chemical recovery.39

4.6.2         Emission Control Techniques

              Particulate emissions from the kraft recovery process are regulated by standards
of performance in 40 CFR 60, Subpart BB.  Particulate emissions consist primarily of sodium
sulfate and sodium carbonate, with some sodium chloride and potentially trace quantities of
CDD/CDF.  Particulate control is provided on recovery furnaces in a variety of ways, which
should have an effect on CDD/CDF emissions since CDD/CDF usually condense on
paniculate.  Further paniculate control is necessary for direct contact evaporators equipped
with either a cyclonic scrubber or cascade evaporator because these devices are  generally only
20 to 50 percent efficient for particulates.70  Most often in these cases, an ESP is employed
after the direct contact evaporator, for an overall paniculate  control efficiency range of
85 percent to more than 99 percent. At  existing mills, auxiliary scrubbers may  be added to
supplement older and less efficient primary paniculate control devices.

              The most common!} used control device on NDCE recover) furnaces is an
ESP  Both \\et and dr\  bottom ESPs are in  use \\ithin the industry.  The  control deMces
yeneralh emplo\  mechanisms to return captured paniculate to the process, thus improving the
efficienc} of chemical recover}1.

4 6 3          Emission  Factors

              Emissions from kraft pulp and paper mills will van,' with variations in the krafi
pulping processes and the type of wood pulped.39  The National  Council of the Paper Industry
for Air  and Stream Improvement. Inc. (NCAS1) recently evaluated and summarized
CDD'CDF emissions data from  seven mills burning black liquor. The individual test reports
and site locations are considered confidential and were not made available for inclusion in this
report.  The information presented here was taken from NCASI's technical project summary.
All TEFs provided in the summary and discussed in this section are based on the l-TEF/89
                                         4-138
-------
scheme, which has been adopted by EPA as an interim procedure for assessing the risks
associated with CDD/CDF exposure.

             Detailed  specific emissions data for CDD/CDF were reported for seven kraft
recovery furnaces.  The furnaces at five mills were NDCE furnaces, while the other two mills
had direct contact furnaces. All seven furnaces were controlled with ESPs and two (one of
each type) furnaces were followed by wet scrubbers (probably for added paniculate emissions
control).

             The total CDD/CDF TEQs for all furnaces were small (<0.01 ng/dscm). One
exception showed an average emission of about 1.6 ng TEQ/dscm at 8 percent O2.  When
subjected to various quality control criteria, the data from this test were found unacceptable
and were discarded.  For the remaining six furnaces tested,  average emissions were estimated
at about 0.002 ng TEQ/dscm  at 8 percent O2.  Assuming nominal conversion factors of
9,000 dscf/10° Btu, and 13,000 Btu/kg of black liquor solids (bis), an average emission factor
of l.lxlO'5 ng TEQ/kg  bis is obtained.41  Table 4-30 presents the summary of CDD/CDF
emissions as reported by NCASI.

4 d 4         Source Locations

             The distribution of kraft pulp mills in the United States in 1997 is shown in
Table 4-31.  Kraft pulp mills are located primarily in the southeast, whose  forests provide
oxer 60 percent of U.S. pulpwood.  Other areas of concentration include the Great Lakes
region of the  midwest and  the Pacific northwest.

4.7          ON-ROAD MOBILE SOURCES

             Internal combustion engines can emit gas-phase polycyclic and polyhalogenated
compounds (e.g., CDD  and CDF) and organic PM as products of incomplete combustion.
The combustion process variables specific to  internal combustion engines are described in this
section.
                                       4-139
-------
 O
 cs:
 u.

 C/3

 o

.u

 u.


 O

 So
 ?/:

 I
 w

 D

 c/;



   i
 TJ-

 u.1

 s


 H
c
5

1



u
8P
fc
>
^^
^<


"b
u.
{*


u
u.
u.
a:



w
u.
*

o
Q
[i.
C£



.£
U
0£




r3
ffl
U.
Bi









Q
'E
b b
it
* iC""
b b
es ^>

^^ ^^
^ **i ^ vi
b b b b
*p x *JJ x
•<->"•: ^*'

* *C^
b £
x ^
^ ^
*~ ^^
* ^r^
o b
X X
q n
^ -^—s

r~ • — -
b b
x ><
o\' --
c ^"^
b ^
x x
C^ ' — •_
^^ ^ ^

•C f^
*^~ *~~
5, -?

~ — '

s£ r^^
b C;
— "^
f^j O"
__; p^j
^^



rN
^*) yj
o c ~5
•a ^ ^-v
Ic ° k °°
£ oo o ^
» — o ob
C" ra i2 s
fel |§
H "gk E u
-i S m H




•ri
U
in
4)
CA
'c
9
•o
C
CD
e
_W
'5

uf
'i
u
h.
i£
is
"c
o
•a
o
u
u
re
£
u
c
re

z
o
"
r
-

c'
^/
—
o
IL>
c-
Psl

re
•c
•J-
o
u
w
4J
•? U
« *
g ^2
K "5
* S
0 >-
05 =
<
u
a u
0 0
c^ Z


























^J
™
C
1
|
'
,1
E
^c
—
re
C
•5: c

r» c
^^ g
2; 2
k-T "

!§ 7
E i
t_)
•c S
c _
CL Cu

U UJ
f £
1 'i
u: w
%%
Z Z
ffl X)



























i)
3
0
u
|
o
,' E
n C
re ,<"
7 ^^
O w
^ ^ Z

)>" ^ ^
^ Tr-
5 *" i;
Z °^ ™
c — o-
~" u.' ^
~ ^ e."

EUJ
• - u
^ if. C
— *o £
&.'«'!
C^ A
tu c/3 T!
•£
'^ ~ ~
« ft> OJ
2 o o
CJ W w
Q U U
Z Q D
U ^3 CJ







































&J
^r
"T~
2
c
—
ii
^j
c
c>
2
-
CA
ul
JC
irnace wil
C
<—



















































•
•r;
~o
tr.
c
O"
u
re
"
ll
15
                                                    4-140
-------
             TABLE 4-31.  DISTRIBUTION OF KRAFT PULP MILLS IN THE
                                UNITED STATES (1997)
State
Alabama
Arizona
Arkansas
California
Florida
Georgia
Idaho
Kentucky
Louisiana
Maine
Maryland
Michigan
Minnesota
Mississippi

Kraft Pulp Mills
14
1
7
2
7
12
1
2
10
7
1
3
2
6

State
Montana
New Hampshire
New York
North Carolina
Ohio
Oklahoma
Oregon
Pennsylvania
South Carolina
Tennessee
Texas
Virginia
Washington
Wisconsin
Total
Kraft Pulp Mills
1
1
1
6
1
1
7
3
6
2
6
4
6
4
124
    Source. Reference 72.
4 7.;
Process Description
             Combustion generates the heat that takes place inside the combustion chamber
Internal combustion engines are generally fueled by gasoline, diesel fuel,  or gasoline oil
mixtures, and can make use of either a two-stroke or a four-stroke cycle.

             In a four-stroke cycle, the piston  strokes are intake, compression, power, and
exhaust.  The two-stroke cycle gasoline engine is designed to eliminate the intake and  exhaust
strokes of the four-stroke cycle.  The two-stroke cycle engine operates on a mixture of oil and
gas, with the oil in the gas being the sole source of lubrication for the system.7"
                                        4-141
-------
              Most passenger cars and some trucks are gasoline-fueled, but large trucks,
buses, and farm and heavy equipment are usually diesel-fueled. Motorcycles, outboard
motors, lawn mowers, and chain saws are examples of equipment that typically use two-cycle
engines.

              Temperatures in the combustion chamber and exhaust system and volume flow
fates influence CDD/CDF formation in internal combustion engines."'74  Gasoline engine
combustion occurs at temperatures around 6,332°F (3,500°C) at near-stoichiometric oxygen
levels. Gasoline engine exhaust temperatures generally range between 752 and 1,112°F (400
and 600°C).  Diesel engines operate at combustion temperatures of about 3,657.6°F (2,000°C)
with an excess of oxygen. Diesel engine exhaust temperatures range between 392 and 752°F
(200 and 400°C).

              The gasoline engine derives its power from the explosion of a mixture of air
and gasoline, whereas in the diesel engine the fuel burns rather than explodes.  The air-fuel
mixture, when ignited, expands rapidly in a cylinder, forcing a piston from the top of the
cylinder to the bottom. The exhaust gases from  internal combustion engines are potential
sources of CDD CDF emissions.1'  After exhaust is released from  a vehicle, it is diluted
approximately  1.000-fold in the  first fe\\ seconds and cools very rapidly

              Paniculate emissions from diesels contain a significant amount of organic
carbon  CDD CDF may accumulate in  engine oil and be  emitted when the oil leaks into the
combustion chamber or exhaust  system and survives the emission  process/1  (

              A number of factors may affect CDD'CDF emissions from gasoline
automobiles and  trucks, including'"

              •      Air-to-fuel ratio;
              •      Mode of vehicle operation;
              •      Vehicle mileage;
                                        4-142
-------
              •     Fuel content;
              •     Presence of additives or lubricants; and
              •     Presence of emission controls.

              Air-to-fuel ratios less than stoichiometric promote incomplete combustion and
increase emissions.  The effect of vehicle operation mode is related to the air-to-fuel ratio.
Cold-start operation will cause higher emissions because the engine is operating in a fuel-rich
condition.  Higher engine load may also increase emissions during cold starts.  Frequent
engine start-ups and shut-downs will decrease the air-to-fuel ratio, thereby decreasing the
amount of fuel oxidized."

              CDD/CDF emissions are expected to increase with vehicle mileage, primarily
because of increased oil consumption. The higher quantities of oil consumed in older, more
worn cylinders provide more intermediates for CDD/CDF formation; in addition,  the
CDD/CDF become concentrated in the oil.76  Another cause of increased CDD/CDF emissions
with increased mileage is the formation of deposits in the combustion chamber.  Emissions
increase \\ith mileage until  the deposits become stabilized.

              Sex eral studies have identified strong correlations between chlorinated additives
in gasoline and motor oil and CDD/CDF emissions during combustion tests.""5' ''  Unleaded
gasoline may have a chlorine content of approximately 0.2 Ib/ton (10 ppm). \\hereas the
chlorine content m leaded gasoline may be 5 to 10 times higher." In addition, it has been
suggested that the concentration of aromatics in the fuel may contribute to these emissions

              CDD'CDF emissions are higher  m cars using leaded gasoline.^  One reason
may be that leaded gasoline contains chlorine in the form  of dichloroethane, which is added
as a "lead scavenger."  However, the amount of lead in leaded gasoline has decreased, and
leaded  gasoline was totally  phased out in 1996.
                                         4-143
-------
 4.7.2         Emission Control Techniques

              Emission control devices such as catalytic converters have reduced automobile
 emissions significantly since the early 1970s.76  In one study, no dioxins were detected in cars
.equipped with catalytic converters using unleaded gasoline.76 A subsequent study revealed
 some CDD and CDF in catalyst-equipped cars, but at much lower levels than in the cars using
 leaded gasoline.77  It is theoretically possible that the CDD/CDF formed in cars using
 unleaded gasoline could be destroyed in the catalytic converter.76  However, the lower levels
 of CDD/CDF found in cars  equipped with catalytic converters using unleaded gasoline cannot
 be attributed solely to the catalytic converter. It appears the combination of the unleaded
 gasoline and the catalytic  converter lowers CDD/CDF levels.

 4.7,3         Emission Factors

              CDD/CDF  can be formed from mobile sources.  However, emission  factors
 relevant  to the United States are not  readily available. A low confidence or quality rating is
 assigned to the results and emission factors derived from European studies because  the fuels
 and control  technologies used in these cars most hkeh differ from U.S. fuels  and
 technologies

              In a  19S~ study in Sweden, automobile exhaust emissions \\ere analyzed for
 CDD CDF/  No CDD CDF were identified from cars equipped  with catalytic com  erters using
 unleaded gasoline  (representame of cars in the United States)   The reported results from cars
 \Mthout catalytic converters  and burning leaded  gasoline is not representative of cars in the
 United States because the  normal scavenger mixture of dichloro- and dibromoethane were not
 used  Reported CDD/CDF emissions were approximate!) 2.6x10"** to 4.7x10'  Ib'ton (13 to
 235  ng'kg) or 8.6x10'" to 1.6xlO'9 Ib/gal (39 to  704 ng gal) of gasoline burned.

              A 1991 Norwegian study reported CDD/CDF emission factors for on-road
 vehicles  measured in a tunnel experiment.  The length of the tunnel was not specified in the
 report, but since complex  ventilation was not stated, the  length was probably relatively short.
                                          4-144
-------
Measurements of traffic density, traffic composition, and ventilation rate were also performed.
Most of the emissions were observed up a grade and were not measured on a flat road.  Road
tunnel studies can represent an opportunity to obtain exhaust emission factors which can be
representative for the car population and the various traffic conditions that usually prevail in a
tunnel.

             The data presented in Table 4-32 differentiates between emission factors for
light duty (LDV) and heavy duty diesel vehicles (HDDV).  Although weekday and weekend
sampling occurred, the study did not differentiate between the fraction of light- and
heavy-duty fleets sampled during this period.  Depending upon driving conditions, the
estimated emission factors were in the order of 1.4xlO~13 to l.SxlO"12 Ib/mile (0.04 to
0.5 ng/km) 2,3,7,8-TEQ (Nordic Model) for LDV and 2.5xlO'10 to 3.4x10'"  Ib/mile (0.7 to
9.5 ng/km) for HDDV.  The emission factors for LDV, expressed as ng/km 2,3,7,8-TEQ,
were obtained by dynamometer experiments using leaded gasoline with dichloroethane  added
as a scavenger, which is not representative of on-road motor vehicles in the United States.

     TABLE 4-32.  EMISSION FACTORS FOR ON-ROAD MOBILE SOURCES
                                 AMS 22-01-001-000
                           FACTOR QUALITY RATING:  U

                          2.3,7,8-TCDD TEQa       2,3,7,8-TCDD         2,3.7.8-TCDF
                               Ib/VMT               Ib'VMT               Ib/VMT
    Source                     (ngA/kmT)             (ng^TonT)             (ng'VkmT)
    Light Duty Vehicles      1.35xlO'13 - 1.84xlO'i:
                          (3.80xlO': - 5.20x10'')
    Heavy Duty Diesel       2.55xlO'12 - 3.37x10-"
    Vehicles                 (7.20x10-' - 9.5)
    Total on-road                8.85xlO'14              3.6xlO'15              5.65xlO'14
    vehiclesb	(2.5QxlO';)	(l.OOxlO'3)	(1.60xlO'2)
   a Source: Reference 78.
   b Source: Reference 79.
   VMT =  vehicle miles travelled.
   VkmT =  vehicle kilometers travelled.
                                         4-145
-------
              The reader is cautioned in using the emission factors presented in Table 4-32
and should recognize that this experiment was conducted with a mixture of new and old cars
{which are not always properly maintained).  At the time of the measurements, the average
age of the car was about 9 years, and unleaded  gasoline usage was only 25 to 30 percent of
total consumption, which also is not representative of on-road vehicles in the United States.

4.7.4         Source Locations

              Internal combustion engines can be found in numerous vehicles, including
passenger cars, small and large trucks, buses, motorcycles, trains, ships, aircraft, farm
machinery, and military vehicles. Because these vehicles can be found nationwide, attempting
to list specific source/sites is not feasible.  It may be reasonable to assume, however, that
there is a direct correlation between population  density and the number of mobile vehicles in
an area (i.e.,  there would be more vehicles in a densely populated area than in  a rural area).

4.8           CARBON REGENERATION

              Activated  carbon is used primarily for adsorbing pollutants from water or air
(e.g.. in industrial or municipal wastewater treatment plants).  Because of increasing
environmental awareness and tighter regulations, the demand for activated carbon is
increasing. The consumption of activated carbon in water and  wastewater treatment
operations in 1990 was reported at 1.44xl08 Ib  (71,900 tons) (6.54x10" kg).80

              Used carbon can be regenerated (reactivated) by essentially the same process as
used for the original activation.  The regeneration process  creates the potential  for CDD/CDF
formation.
                                         4-146
-------
4.8.1          Process Description

              In the regeneration process, organics adsorbed on carbon during use are burned
off by placing the spent carbon in continuous internally or externally fired rotary retorts or,
most commonly, in multiple-hearth furnaces.81  Figure 4-30 shows a cross-section of a typical
multiple-hearth furnace.  In this type of furnace, the charge (carbon) is stirred and moved
from one hearth to the next lower hearth by rotating rabble arms.  For smaller-scale
regeneration operations, fluidized-bed and infrared furnaces can be used.  The various furnace
types used for carbon regeneration and the approximate number of furnaces of each type are
shown in Table 4-33.82

              In a typical regeneration process, spent carbon in a water slurry form is fed
from a surge tank  to a dewatering screw, which feeds the spent carbon to the top of the
furnace.  In the furnace, the spent carbon is  dried and the organics on the carbon are
volatilized  and burned as the carbon is regenerated. The regenerated carbon drops from the
bottom hearth of the furnace to a quench tank and is stored as a slurry.81 A flow diagram  of
the carbon regeneration process is shown in  Figure 4-3 l.M

              A hot gas, such as  steam or carbon dioxide, is  introduced into the furnace at
temperatures of approximately 1498 to 1858°F (800 to 1.000°C). although some excess
oxygen is typically present throughout the furnace.*"  The regeneration process is exothermic.
using  the heating value of the volatile carbon plus heat supplied from  supplemental fuel (e.g..
natural gas).  A typical furnace may fire an  average of 459.089 cubic  feet (13.000 cubic
meters/day) of natural gas.81
                                         4-147
-------
                                              Furnace Exhaust
                                               to Afterburner
 Pyro lysis
  Gases
                                                                                    FMd
                                                                                   Matsrial
                                 Cooling and Combustion
                                          Air
              Figure 4-30.  Cross-Section of a Typical Multiple-Hearth Furnace


Source:  Reference 18.
                                          4-148
-------


1


i
1
s
, I
1
1
f
1
(X








|« I W
•S O*
w s 1

5"
is ,
:o- *
ic
: o
;i










2
Cfl «
to O
S*
a i
I ^ i
1 <
I f 1

, ^

5r-. 5S 8§ =
= m E-= *L_
^ S ' 	 3 X
5 x u.uj
\ *
1 '
, e
E ^s o
S CO c
i=|




1








•a
|s&s
ilil
IOCTJI-
oc

1



I

£
£•*
^ CD C
0^




                                                           0-
                                                           c

                                                           'ea
                                                           o

                                                           00
                                                           o
                                                           OL

                                                           o

                                                           C3
                                                           u
                                                           C3
                                                            CO

                                                            00
                                                            •J
                                                            o
                                                            c/5
                                                            on
                                                            o
                                                            £


                                                            el
                                                            IE
4-149
-------
            TABLE 4-33.  TYPES OF EQUIPMENT USED FOR ACTIVATED
                              CARBON REGENERATION

                                                       Approximate No.
    Furnace Type	of Units in U.S.
    Multiple-hearth                                           <100
    Fluidized-bed                                     -        <20
    Indirect-fired rotary kiln                                    >50
    Direct-fired rotary kiln                                      <30
    Vertical-tube type                                          <30
    Infrared-horizontal                                           <5
    Infrared-vertical                                               4
   Source: Reference 82
              Typical industrial carbon regeneration plants may process up to 109,127 Ibs/day
(49.500 kg/day) of spent carbon from numerous industrial or municipal facilities that use
activated carbon for wastewater treatment.81  Regeneration plants may operate 24 hrs/day.
  days \vk for much of the year, with periodic shut-downs for furnace maintenance.

              Emissions from carbon activation and regeneration processes contain a number
of toxic air pollutants.  Regeneration has  an even greater potential for producing toxic
emissions because the carbon has often been used in adsorbing compounds classified as toxic
air pollutants ^

              Of special interest is the potential for CDD/CDF formation in the high-
temperature, low-oxygen environment of the regeneration furnace.  One study found no
evidence of CDD/CDF emissions from the regeneration of virgin carbon, but did  detect both
families of compounds when regenerating spent carbon from wastewater treatment facilities."
The data indicate that these byproducts formed from the adsorbed organics on the spent
carbon rather than from impurities in the virgin carbon.
                                        4-150
-------
4.8.2         Emission Control Techniques

              The primary point source of emissions from the carbon regeneration process is
the furnace exhaust. These emissions are typically controlled by afterburners followed by
water scrubbers.81  The afterburner may consist of a short vertical section with natural
gas-fired burners and a long horizontal section of refractory-lined duct with no burners.
Afterburner combustion temperatures of 1822°F (980°C) or greater and residence times in
excess of two seconds are typical.83  Temperatures greater than 1625°F (871°C)  and residence
times longer than 0.5 seconds are recommended.81  There are no available data on destruction
removal efficiency (DRE) for an afterburner control system in this application.   However, the
conditions and configuration are similar to those used for controlling hazardous  waste
incinerator emissions, where DREs of 99.99 percent are typical.83

              Exhaust gases from the afterburner can be cooled by an alkaline (e.g., sodium
carbonate) spray cooler in which an  atomized  dilute  alkaline solution is mixed with the
exhaust gas.  The alkaline medium neutralizes acid gases to permit  compliance with regulatory
emission limits.8' From the spray cooler,  the exhaust gases may enter centrifugal or fabric
filter (baghouse) collectors,  which are used to  control particulate and reaction products from
upstream components.  Collection efficiencies  of 65  percent for  centrifugal collection and
99 percent for fabric filtration have been reported.s?  The collected particulate is ultimately
disposed of in a landfill.

4 8.3         Emissions and Emission Factors

              Several studies have been conducted to test CDD/CDF emissions  from carbon
regeneration  facilities.  Table 4-34 summarizes the results of two studies in which emissions
from a fluidized-bed system were tested.9  The first study tested emissions from  the system
before an afterburner was installed; the second study took place  after its installation.  The
carbon regenerated during the first study had been  in service  for approximately one year, and
the carbon in the second study for 200 days.
                                          4-151
-------








z
00
<
_. tu
eS
^ _1
r £
^"
fe ^
*^ M ~
r~ ^
^ —
£ ~
P ?r
\^ t""*
<

^ *£
f. S
4 C
> "••
j ^
rvi
<
t—














tu
o
u
f
H


0
Q
U
re
O
f-









—
•—
^
~
oc
r-
P^,
r^i







1.

c"e
'Z 1>
60 3,
1


t
£^
i!
V
%
i£ ^
C"E
•:|>
M ^5.
n
o:


.**
W-
ic ^
e 1
'i "
TO s_ ^
U


'^
£ ^-^
f~ "c
c c
"."I1
5^ -— -
re
Oi


c:
•^ ^^
c E
'e §•
r: >_-
o
5
t£.

i
c-
re
00


^

o
n
U.
u-
c
u
p.
S
- ~ *> 5
b^o ^
-- ui "3 c
X .-; JS c
S? 8l5?
£ oo * i N
, o ' - '
o o D O
>ri ^ O


*> •*: 'an
b^ fi-S
H ^ ^ c
So" ?§
- 8-

— '•J — 60
£ o ^ c
06 • 2 ^
— VO ^ ^
, q £ '
»•> 8, — ^
^ £~


c
° ^^
"28? 2S
x — ; * c
^. >oo — O
^S^^t-:2- » ^
•^ oo »s —
o •~> o ^
^« ^* ^^ <^
X ^ X ^
S'" ?T
7o °?g
Q 5, Q ^
z z



wi «ri ^ ^
2^2^22 "2 P
-. 2 ^- J5d ^o
06 ^^ e> "^ r^: ^^ fi
ON O" ^ n














^V S** ^ ^K
Z Z Z Z





it 1-1-
— o r" 5 u r- ? °
r/- -C; U^uC'^U
^ £- ti-s^'i-fc1"
n r^ e *~ a: ^
t ^ < <
UH >--^ ^.^.


"8 g
ca E
•o *
si
^ c
'= <
E "$






























j^
"c
c

^\ I/".
•^ PJ
c -
= EL
-------
             In the first study, concentrations of 2,3,7,8-TCDD in the flue gas ranged from
6.24x10-" to l.SlxlO'14 Ib/ft3 (0.01 to 0.21 ng/m3), with an average of 6.24X1005 Ib/ft3
(0.1 ng/m3). In the second study, emissions from the stack and afterburner were tested.
2,3,7,8-TCDD was not detected in any of the samples.

             In another study, a horizontal infrared regeneration furnace was fitted with an
afterburner designed to provide a 20-second residence time at a temperature of 1876°F
(1010°C).  Table 4-35 shows CDD/CDF emissions from this study.84

             Results of CDD/CDF emissions testing at a carbon regeneration facility with a
multiple-hearth furnace are shown in Table 4-36.81  The emissions at this facility were
controlled by an afterburner, a sodium carbonate spray cooler, and a baghouse.  Sampling was
performed at the spray cooler inlet location and the baghouse outlet exhaust stack. Samples
of the baghouse dust were also collected and analyzed for CDD/CDF.  In addition, ambient
air sampling was performed near the atomizing air intake point at the spray cooler.

             Detectable quantities of all targeted dioxin and furan species except
2,3,7,8-TCDD and 2,3,7,8-TCDF were found in the stack  gas at the baghouse outlet exhaust
stack.  At the spray cooler inlet, all targeted CDD/CDF species were detected.  Dioxin and
furan homologues except 2,3,7,8-TCDD  were detected at low concentrations in the baghouse
ash.

              Results of 1991  emissions tests performed at  a county water treatment facility
in California were recently made  available. The tests were  conducted on the lime recalcing
unit  and the charcoal furnace.  Each unit was tested for speciated organic compounds,
including dioxins and furans.  Emission  factors developed from the test results are presented
in Table 4-37.  Note that the configuration and type of furnace tested is not known.
However, the test report  did state that the  furnace was controlled by an afterburner and a
scrubber.85
                                         4-153
-------
 TABLE 4-35.  CDD/CDF CONCENTRATIONS AND EMISSION FACTORS FOR A
       HORIZONTAL INFRARED CARBON REGENERATION FURNACEa-b
                                SCC 3-99-999-93
                        FACTOR QUALITY RATING:  U
 Isomer
Average Concentration in
   Ib/dscf (ng/dscm)
Emission Factors in
  Ib/ton (mg/kg) of
 carbon regenerated
 DIOXINS
      2,3,7,8-TCDD
      Total Other TCDD
      Total PeCDD
      Total HxCDD
      Total HpCDD
      Total OCDD
      Total CDD
         ND
  7.50X10'14 (1.20X10'2)
         ND
  2.80xlO'16 (4.48x10-')
  2.00xlO-15 (3.20xlO-2)
  1.56xKrlfc(2.50xlO-')
  7.74xlO'N (7.42x10-')
4.80X10'5 (2.40xlO'2)

1.80x10° (9.00x10'')
1.20xlO'4 (6.40x10'2)
1.28x10-' (5.02x10-')
  2.10x10--' (1.49)
 FURANS
      2.3.7.8-TCDF
      Total Other TCDF
      Total PeCDF
      Total HxCDF
      Total HpCDF
      Total OCDF
      Total CDF
  4 20x10•'•• (6.80xlCr:)
  1.18x10-'" (1.89x10-')
  4.70xlO-!? (~.50xlO-2)
  1.87x10 '• (3.00xl02)
  2.68x10'" (4.30x10°)
  2 06x10-'-' (3.30xlO:)
  1 56xl(V14 (4.38x10-')
2.74xlO-4 (1.37x10-')
7.60xUr4 (3.80x10-')
3.02xlO-4 (1.51x10-')
1.20x10'' (6.00.x 10':)
1.72X10"1 (8.60xlO-2)
1.32x10"* (6 60xlO':)
1.76x10-' (8 8x10-')
Source  Reference 84
a Combustion gas flow rate was 196 dscm.hr  Facility operated 7.000 ni'yr Operating rate of system
  was 97.5 kg of spent activated carbon per hour
k Control device consists of afterburner sized for a 0 3-mmute (20-second) residence time at 1,850°F
ND  =  Not detected
I"   =  Unratable
                                    4-154
-------
    TABLE 4-36. CDD/CDF EMISSIONS DATA FROM A MULTIPLE-HEARTH
                   CARBON REGENERATION FURNACE3
                              SCC 3-99-999-93
                      FACTOR QUALITY RATING:  D
 Isomer
  Concentration in
  Ib/dscf (ng/dscm)
 Emission Factor in
   Ib/ton (ng/kg)
 INLET:
       2,3,7,8-TCDD
       Total CDD
       Total CDF
5.62xlO-15 (9.00xlO-2)
  1.79xlO-18(28.8)
  3.13xlO-18(50.1)
1.78xlO-9 (8.90x10-')
  6.00x10-" (300)
  1.40xlO'i: (700)
 OUTLET:
       2,3,7,8-TCDD
       Total CDD
       Total CDF
       ND
  2.30xlO'17 (3.69)
  2.07xlO'17 (3.32)
       ND
  6.26x10-'° (31.3)
  5.46\10-Ul (27.3)
 BAGHOUSE ASH:
       Total CDD
       Total CDF
  6.86x10-'° (1.1)
3.12x10-'° (5.00x10'
       XR
       NR
Source.  Reference 81.
a Control devices consist of afterburner, sodium carbonate spray cooler, and baghouse
ND =  Not detected.
NR =  Not reported.
                                   4-155
-------
       TABLE 4-37.  CARBON REGENERATION FURNACE EMISSION FACTORS
                                    SCC 3-99-999-93
                            FACTOR QUALITY RATING:  U

Pollutant
2,3,7,8-TCDD
2,3,7,8-TCDF
2.3,7,8-TCDD TEQ
Total CDD
Total CDF
Emission
Ib/tori carbon
reactivated
2.10x10-'°
1.36xlO-9
3.46x10-°
4.64x1 0's
4.76xlO'8
Factor
Kg/Mg carbon
reactivated
l.OSxlO'10
6.80x10-'°
1.73x10-"
2.32xlO'8
2.38x10-*
   Source Reference 85  The type of configuration of the furnace were not specified  Control devices used
         were an afterburner and a scrubber.
              In summary, the studies indicated that, in most cases, detectable quantities of
targeted dioxm and furan species were found at various locations (stack outlets, spray cooler
inlets, ambient air) at carbon regeneration facilities  Ho\\ever, emission control devices
reduced CDD CDF emissions  The sites chosen for these  studies were  considered
representatne of other carbon regeneration facilities in the United States, therefore, the
emission factors de\ eloped from the data are considered reliable.

4 S 4         Source Locations

              Activated carbon is used primarily to adsorb orgamcs from water at industrial
or municipal wastewater treatment plants.  Carbon regeneration may be performed at the site
where the carbon  was used (on-site regeneration) or at a commercial regeneration facility that
processes spent carbon from multiple industries.  Because  of the numbers of potential
individual emission sources, listing specific sites in this document is not feasible.
                                         4-156
-------
4.9           OPEN BURNING AND ACCIDENTAL FIRES

              This section describes CDD/CDF formation from forest fires, agricultural and
open refuse burning, and structure (building) and PCB fires, and their associated emission
factors.

4.9.1          Forest Fires and Agricultural Burning

              Process Description

              The burning of forest lands occurs through controlled prescribed burning and
through uncontrolled accidental forest fires.  Prescribed burning is the application and
confinement of fire under specified weather, fuel moisture, and soil moisture conditions to
accomplish planned benefits such as fire hazard reduction, control of undesired species,
seedbed and site preparation, wildlife habitat improvement, and tree disease control.
Uncontrolled forest fires (wildfires) are fires that are started naturally (e.g., by lightening),
accidentally, or intentionally that bum and spread in generally unpredictable patterns.

              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:

              •      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 seeds;
                                          4-157
-------
             •      Indirect weed control by providing clean soil surface for soil-active
                    herbicides; and
             •      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.

             Emission Factors

             Although the potential for emissions exists, CDD/CDF emission factors have
not been identified for agricultural burning.   As discussed above, CDD/CDF have been
detected  in wood-fired boiler emissions and in the ash  from residential wood stoves.
Although the combustion processes that take place in agricultural burning are different from
those in wood-fired boilers or wood stoves, the  fuels are of similar composition.  Reported
chloride  concentrations range from 100 to 10,000 ppm in wood and agricultural vegetative
matter.86  Emission factors based on the mass of pollutant emitted per mass of material
combusted would be expected to be low for  agricultural burning; however, total emissions
could be substantial because of the large amounts of materials combusted.

              Two separate studies reviewed indicated that wood burned in  forest fires may
reasonably be considered a source of CDD/CDF.87'8* Another study reviewed reported direct
measurements of CDD/CDF in the actual emissions from forest fires at detected levels of  15
to 400 pg<;m3 for total CDD/CDF.89  These concentrations cannot accurately be converted to
an emission factor because the rate of wood  combustion is not known. However, an
alternative approach  assumes that the emission factor for residential wood burning (using
natural wood and an open door) applies to forest fires.  This approach suggests an emission
factor of about 1 ng TEQ/kg of material burned for  total CDD/CDF.  It should be noted that
forest fire and wood stove combustion conditions differ significantly.   Thus, this emission
factor is considered highly uncertain and  is assigned a low quality rating of U (unratable
because  it was developed from engineering judgment based on theoretical data).

                                          4-158
-------
              Source Locations

              According to the U.S. Forest Service, the majority of prescribed burning in the
United States occurs in the southern/southeastern region of the United States (60 percent in
1984), followed by the Pacific Northwest (almost 20 percent), and California (10 percent).87-**

              The locations of uncontrolled forest fires are not as definable as those of
prescribed burning sites, but historical records of fires and a knowledge of the locations of
primary forest resources can be used to predict where the majority of forest fires are likely to
occur. The southern and western regions of the country (including  California, the Pacific
Northwest, and western mountain states) appear to represent the greatest potential for forest
wildfires.   Forest Service data for 1983 indicate that 67 percent of-the total number of acres
burned by wildfires nationally were in the southern/southeastern region. The western regions
contained  17 percent of the wildfire-burned acreage, and the northern region (Idaho. Montana.
North Dakota) contained another 6 percent.

              Agricultural burning is direct!}' tied to the agriculture industry.  Major
agricultural states-including California, Louisiana. Florida. Hawaii, North Carolina,
Mississippi, and Kansas-conduct the majority of agricultural burning.

492          Miscellaneous Open Refuse Burnine and Structure Fires

              Process  Description

              The most readily identifiable types of open-burned refuse materials are
municipal refuse, bulky items such as furniture and bedding, construction debris, and  yard
waste. Structure fires  are similar to open refuse burning in that the types of materials
eombusted are similar  (e.g., wood, paper, plastic, textiles, etc.).

              The procedure of open burning is relatively simple.   The material to be burned
is collected and  aggregated in an open space fully exposed to the atmosphere.  The materials
                                          4-159
-------
are ignited and allowed to burn and smolder until all combustible material is consumed or the
desired degree of volume reduction is achieved.  Structure fires are highly variable in nature
and often result in smoldering heaps  similar to open refuse piles.

             In open refuse burning and structure fires, combustion conditions are typically
poor and are highly variable because of variations in air flow, fuel moisture content, oxygen
levels, material configuration, and degree of exposed surface area. In addition, some refuse or
building materials may contain organic constituents that are CDD/CDF precursors or that
accelerate CDD/CDF formation.

             Emission Factors

             CDD/CDF emission factors have not been identified for miscellaneous open
burning or structure fires.  However, CDD/CDF emissions would be expected because the
composition of materials burned in these  fires may be the same or similar to that of municipal
waste combusted in MWCs.  CDD/CDF emissions from MWCs measured at the inlet to
pollution control equipment (i.e., uncontrolled emissions) were previously presented (in
Section 4.1).  The combustion processes occurring in open-burning refuse piles or structure
fires are much less efficient than those in an MWC and may provide an enhanced
environment for CDD/CDF formation.

             Open burning of municipal waste or construction debris containing chlorinated
plastics or other chlorine-containing materials would be expected  to emit levels of CDD'CDF
comparable to or higher than those from  uncontrolled MWC emissions. On the other hand,
open burning or structure fires in which the materials being burned are low in chlorine
content (e.g., wood, yard waste) would be expected to have lower CDD/CDF emissions.
                                         4-160
-------
4.9.3         Polvchlorinated Biphenvls Fires

              Process Description

              Fires involving polychlorinated biphenyls (PCB)-containing electrical
equipment such as transformers and capacitors can result in CDD/CDF formation and
emissions.  Electrical equipment containing PCB may catch fire or explode as a result of a
fire in a building containing such equipment, or during lightening strikes or electrical surges.

              Regulations established by EPA have reduced the chances of PCB fires by
(1) requiring the removal of some large networks of PCB transformers near commercial
buildings; (2) banning the installation of new PCB transformers; (3) requiring existing PCB
transformers to be  equipped with enhanced electrical protection; (4) requiring the removal of
combustible materials from PCB transformer locations; and (5) requiring that all PCB
transformers be registered with building owners and emergency response personnel.011
However, many PCB transformers and other electrical equipment are still in use and subject
to accidental fires or explosions.

              Emissions Data

              Gaseous emissions from PCB fires  have not been measured and CDD CDF
emission factors are not available.   However, the presence  of CDD CDF in soot from PCB
fires has  been  confirmed in several studies.9'  The data identified from these studies are
presented belo\\

              In Binghamton, New York, in  1981, an electrical  transformer containing about
1,100 gallons of PCB was involved in an explosion.  Total CDD/CDF in the soot was initially
found to  be as high as 4.3 Ib/ton (2,160,000 ng/g).  2,3.7,8-TCDF accounted for 0.02 Ib/ton
(12,000 ng/g) of total CDF. HxCDF alone accounted for 1.9 Ib/ton (965,000 ng/g) of total
CDF.  Total CDD were found at a concentration of 0.04 Ib'ton (20,000 ng'g), including
1.2xlO-? Ib/ton (600 ng/g) 2,3,7,8-TCDD.
                                        4-161
-------
             In January 1982, an electrical fire involving PCB occurred in a Boston,
Massachusetts, office building. One bulk soot sample contained a total of 0.23 Ib/ton
(115,000 ng/g) CDF, including 0.12 Ib/ton (60,000 ng/g) TCDF.  CDD were not detected
above an analytical detection limit of 2X10"4 (100 ng/g).91

             In Miami, Florida, in April 1982, a fire and explosion occurred when an
underground transformer vault exploded, releasing approximately 100 gallons (379 liters) of
PCB transformer oil onto the  floor.  Smoke ejector fans were set up to ventilate the vault.
Samples of soot and other residue from the fire were collected from surfaces near the fire
scene.  CDD were not detected in these samples at an analytical detection limit of
2xlO'5 Ib/ton (10 ng/g).  One soot and dust sample contained 3.4xlO'3 Ib/ton (1,710 ng/g)
TCDF through OCDF homologues and another soot sample contained  1.3xlO'3 Ib/ton
(670 ng/g) TCDF through OCDF homologues.  The 2,3,7,8-TCDF isomer was not detected at
an analytical detection limit of 2xlO"5 Ib/ton (10 ng/g).91

             A  fire in Washington State in 1984 involved transformer oil and cores.  A grab
sample of the ash was analyzed and found to contain 8.2xlO"5 Ib/ton (41.4 ng/g) CDF and
5.4xlO'6 Ib'ton (2.7 ng/g) and  5xlO'6 Ib/ton (2.5 ng/g) of the HpCDD and OCDD homologues.
respect) veh.9'

             Source Locations

             Transformers and capacitors containing  PCBs are widely distributed throughout
the United States.  They are located at electrical substations, in commercial and industrial
buildings, mounted on utility poles, in railroad locomotives, and in mining equipment motors
Although the  installation of new PCB-containing equipment has been banned and regulations
regarding existing ones have been implemented, there  are millions of existing PCB
transformers and capacitors currently in use in the United States. Table 4-38 provides
estimates of the numbers  and types  of PCB-containing electrical equipment in the United
States in 1988.
                                         4-162
-------
    TABLE 4-38.  ESTIMATES OF THE NUMBER AND TYPE OF PCB-CONTAINING
             ELECTRICAL EQUIPMENT IN THE UNITED STATES (1988)
Equipment
Transformers
Capacitors
Mineral Oil
Transformers/Capacitors
Mineral Oil
Transformers/Capacitors
Mineral Oil
Transformers/Capacitors
Other Electrical Equipment
PCB Content
70% (by wt.)
70% (by wt.)
>500 ppm
50-500 ppm
<50 ppma
<50 ppma
Approximate Units
32,000
1,500,000
200,000
1,500,000
14,920,000
700,000
  Source  Reference 91
  a These units are expected to contain small quantities of PCB.
4 10         MUNICIPAL SOLID WASTE LANDFILLS

             A municipal solid waste (MSW) landfill unit is a discrete area of land or an
e\ca\ation that receives household waste, but is not a land application unit (i.e.. for receiving
se\\age sludge)  An MSW landfill  unit may also receive other types of wastes, such as
commercial solid waste, nonhazardous sludge, and industrial solid waste.  CDD CDF
emissions from MSW landfills are expected to originate from the non-household sources of
MSW

             MSW management in the United States is dominated by disposal in landfills.
Approximate!) 67 percent of solid waste is landfilled, 16 percent is incinerated, and
17 percent is recycled or composted.  There were an estimated 5,345  active MSW landfills in
the United States in 1992.  In 1990, active landfills were receiving an estimated 130 million
                                       4-163
-------
tons (118 million Mg) of waste annually, with 55 to 60 percent reported as household waste
and 35 to 45 percent reported as commercial waste.92

4.10.1        Process Description

              There are three major designs for municipal landfills: the area method, the
trench method, and the ramp method.92  They all utilize a three-step process, which includes
spreading the waste, compacting the waste, and covering the waste with soil.  The area fill
method involves placing waste on the ground surface or landfill liner, spreading it in layers,
and compacting it with heavy equipment.  A daily soil cover is spread over the compacted
waste. The trench method entails excavating trenches designed to receive a day's worth of
waste. The soil from the excavation is often used for cover material and wind breaks.  The
ramp method is typically employed on sloping land, where  waste is spread and compacted in
a manner similar to the area method; however, the cover material  obtained is generally from
the front of the working face of the filling operation.  The trench  and ramp methods are not
commonly used, and are  not the preferred methods when liners and leachate collection
systems are utilized or required by law

              Modern landfill design  often incorporates liners  constructed  of soil
(e.g., recompacted clay) or synthetics  (e.g.. high  density polyethylene) or both to provide an
impermeable barrier to leachate (i.e., water that has passed  through the  landfill) and gas
migration from the  landfill.

4.10.2        Emission Control Techniques

              Landfill gas collection systems are either active  or passive systems.  Active
collection systems provide  a pressure  gradient in order to extract landfill gas by use of
mechanical blowers or compressors.  Passive systems  allow the natural  pressure gradient
created by the increase in landfill pressure from landfill gas generation  to mobilize the gas for
collection.
                                         4-164
-------
              Landfill gas control and treatment options include (1) combustion of the landfill
gas, and (2) purification of the landfill gas.  Combustion techniques include techniques that do
not recover energy (e.g., flares and thermal incinerators) and techniques that recover energy
and  generate electricity from the combustion of the landfill gas (e.g., gas turbines and internal
combustion engines).  Boilers can also be employed to recover energy from landfill gas in the
form of steam. Flares involve an open combustion process that requires oxygen for
combustion; the flares can be open or enclosed. Thermal incinerators heat an organic
chemical to a high enough temperature in the presence of sufficient oxygen to oxidize the
chemical to C02 and water.  Purification techniques can also be used to process raw landfill
gas to pipeline quality natural gas by using adsorption, absorption,  and membranes.

4.10.3         Emission Factors

              During the development of this document, no data were identified that indicate
CDD'CDF  are emitted  in landfill gas.  However, one test report on a landfill gas flare was
obtained that  presents CDD/CDF emissions results.93 Emission factors developed from data
presented in the report  are provided  in Table 4-39.  Results of the emissions test on the flare
indicate that combustion of landfill gas may be a source of CDD'CDF emissions.

4 11           ORGANIC CHEMICALS  MANUFACTURE AND USE

              Chemical reactions involved m the manufacture  of halogenated  organic
chemicals can produce  small quantities of dioxin and furan by-products.  These pollutants
ma\  be  lost to the air during product manufacture or emitted later during the use of the
contaminated products. This section documents potential mechanisms for CDD/CDF
formation and the potential occurrence of these contaminants in the production and use of
specific halogenated organic chemicals.  In addition, data on actual product analysis for some
of these compounds is presented. The presented information has been limited to chlorinated
and brominated compounds currently produced in the United States that are most likely to be
contaminated with dioxins and furans.
                                         4-165
-------
        TABLE 4-39. EMISSION FACTORS FROM A LANDFILL GAS
                        COMBUSTION SYSTEM*

                           SCC: 5-02-006-01
                    FACTOR QUALITY RATING:  D
Isomer
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,6,8-HxCDD
1,2,3,4,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2.3.4,7,8-HxCDF
1.2,3,6. 7. 8-HxCDF
1.2,3,7-8.9-HxCDF
2.3.4.6.7,8-HxCDF
1,2,3,4,6, 7,8-HpCDF
1.2,3.4,7,8,9-HpCDF
Total OCDF
Total CDF
Ib/MMBtu
" 2.30xlO'12
1.15x10-"
9.20x1 0'12
9.20xlO'12
3.23x10'"
9.45x10-"
5.52x10-'°
7.11x10-'°
1.76xlO'9
4.82x10-"
1.42x10-'°
1.82x10-'°
5.28x10'"
1.38xlO-n
8.52xlO'u
1.52x10-'°
9.19xlO'12
7.99x10"
2.53xlO'9
g/MJ
9.89xlO'13
4.95xlO'12
3.96xlO'12
3.96xlO'12
1.39x10'"
4.06x10-"
2.37x10-'°
3.05x10-'°
7.57x10-'°
2.07x10-"
6.11x10-"
7.83x10-"
2.27x10'"
5.93xlO'i:
3.66xlO'u
6.54xlO'i;
3.95xlO'!:
3.44x10'"
1.09xlO'9
Source: Reference 93.

a Control device is an afterburner where test was taken.
                                4-166
-------
4.11.1        General Chemical Formation Mechanisms

-,              Four major mechanisms have been postulated for the formation of halogenated
dioxins and furans in the manufacture of halogenated organic chemicals:94'95  (1) direct
halogenation of dioxins or furans (Figure 4-32a); (2) reaction of an ortho halogen with a
phenate (Figure 4-32b); (3) loss of the halogen (e.g., chlorine or bromine) from a halogenated
phenate to form halogenated furans (Figure 4-32c); and (4) reactions between ortho- and
meta-substituted halogens (Figure 4-32d).

              A number of factors influence the amount of dioxins and furans that  may be
formed in a given manufacturing process, including temperature, pH, catalyst, and reaction
kinetics.95

              The effect of temperature on the formation of halogenated dioxins and furans is
well recognized.  A mathematical relationship between temperature and dioxin formation has
been proposed to calculate the theoretical  amount of dioxins that can be expected.95 This
relationship is expressed as:
                                 7y = 0.025t2e-3|"-200)200'

where:
              y = dioxin concentration
              t = temperature

              This relationship, graphically represented in Figure 4-33, shows that dioxin
formation peaks at 392°F  (200°C) and decreases unsymmetrically with increasing temperature.
The use of this predictive model also assumes that (1)  impurities in the feedstock (including
any dioxins) are carried through to the final product, based on the  chemical stability and low
concentration of dioxins formed; (2) there is a catalyst present to promote the reaction; and
(3) there is no purification of the product.95
                                          4-167
-------
C!
                            (x+y)Cl
                           Hatogenation of Dtoxins or Furans
                                                            (x+y)HC)
                  cr  x
                                                         x  cr
                                                            M*
                                                            -X
                                                                       MX
                                                 X = Leaving Group
                                                     (e.g., Cl, Br, F, I. NO,;
                                                 M » Alkali Metal Cation
                                                 Y = Substttuent Group

                    b. Reaction of an Ortho Halogen with a Phenate
ONa   Cl
                c. Loss of Halogen (e.g., Chlorine or Bromine) from a
                  Halogenated Phenate to form Halogenated Furans
                                                   *r-
                                                £>
                                                Na


                                               Br
                                                        Y-
                                                              Na*
                                                                 .0' Br
                                                                         ^
         0  Br
       Br
                                                                          0
                  d. Reactions Between Ortho - and Meta Substituted Halogens


        Figure 4-32 (a-d). Mechanisms for Halogenated Dioxin and Furan Production


 Source:  References 94,95.

                                         4-168
-------
     1000—
o
c
o
O
o    500 —
O
100       200       300      400       500




                       Temperature fC)
                                                                   600
                    Figure 4-33.  Dioxin Concentration Versus Temperature
  Source: Reference 95.
                                            4-169
-------
4.11.2        Chlorophenols

              Formation mechanisms and potential emissions of dioxins and furans for the
production and use of chlorophenols are discussed in this subsection.

              Chlorophenol Use

              Since the 1950s, chlorophenols have been used as herbicides, insecticides,
fungicides, mold inhibitors, antiseptics, disinfectants and, most importantly,
woodpreservatives.

              Polychlorinated phenols (PCP) are currently used in the United States as
industrial wood preservatives.  The principal and most effective method of application is
pressure treating, which forces PCP into wood fibers.  The bactericide, fungicide, and
insecticide  properties  of PCP help to preserve outdoor lumber, including railroad ties, marine
pilings, highway barriers, and, primarily,  utility poles.  All other United  States uses of PCP
have been discontinued.96

              2.4-dichlorophenol (2,4-DCP) is produced commercially for use as an
intermediate in the manufacture of industrial and agricultural products. One of the primary
uses of 2,4-DCP includes feedstock for the production of 2.4-dichlorophenoxyacetic acid
(2.4-D) and derivatives, which are used as pesticides, germicides,  and  soil sterilants.  2.4-DCP
is also used in the  production of certain methyl compounds used in mothproofing, antiseptics.
and seed disinfectants. Furthermore, 2.4-DCP may be chlorinated with benzene sulfonyl
chloride to produce miticides, or further chlorinated to produce PCP.95

              Dioxin and Furan Contamination in the Manufacture and Use of Chlorophenols

              Chlorinated phenols, including DCP and PCP, are manufactured by the
chlorination of the phenols using catalysts or by the alkaline hydrolysis of a chlorobenzene.
Both of these reactions can produce CDD/CDF by-products that show  up as contaminants in
                                         4-170
-------
commercially produced chlorophenols and chlorophenol derivatives such as phenoxy acids.

Potentially contaminated chlorophenol products and their derivatives are listed in Table 4-40.
                                                                               .4
Although most of the following discussion are pertains to the formation of CDD, the reaction

mechanisms for the formation of CDF the same.94
       TABLE 4-40.  SOME COMMERCIAL CHLOROPHENOL PRODUCTS AND
    DERIVATIVES THAT MAY BE CONTAMINATED WITH DIOXINS OR FURANS


    Common Name                       Chemical Name               Primary Use

    2,4-D (esters and salts)         2,4-dichlorophenoxyacetic acid           Pesticide
                                 and esters and salts

    2,4-DB and salts               2,4-dichlorophenoxybutyric               Pesticide

    2,4-DP                        2-2,4-dichlorophenoxy propionic          Pesticide
                                 acid

    PCP and salts                 Pentachlorophenol and salts           Wood treatment

  Source  Reference 95
             Dioxins may form as a contaminant in commercial products b\  the

intermolecular condensation of polyhalophenols to polyhalodibenzo-p-dioxms.  including the

condensation of phenates with various chlorine substituents. Condensation reactions of

chlorophenols are influenced by the following factors''1"


             •      The total number of chlorine substituents, which determines the  ease of
                    chlorine removal and the ether-bond formation.

             •      In the case of solid-state reactions, the arrangement of the molecules
                    within a crystal are influenced by the metal cation involved

             •      Steric effects from molecular conformations that impact  site-specific
                    nucleophilic substitution.

             •      Electronic effects, which allow chlorine atoms in some positions to be
                    removed more readily than those in the other positions on the ring.
                                       4-171
-------
             In laboratory preparations, yields of 10-20 percent have been reported for the
Ullman-type self-condensation of 2,4-DCP to 2,7-dichlorodibenzo-p-dioxin.94-95  Another study
on the chemistry of chlorinated dioxins obtained 80 percent OCDD in the solid-state
condensation of sodium pentachlorophenate molecules.94'95  This high yield was attributed to
lattice or steric  effects of sodium pentachlorophenate, which open the oxygen atom to attack
and thus lead to dioxin formation.  The same study yielded 30 percent HxCDD in the
solid-state condensation of sodium 2,3,4,6-tetrachlorophenate.  The lower yield and the
finding of two isomers of HxCDD indicate that the tetrachlorophenate is less stereo-specific
than the pentachlorophenate.

             CDD Formation in PCP Manufacture-PCP may be manufactured commercially
by direct chlorination of phenol or as a mixture of chlorophenols, as shown below:
                      OH                                OH

                                               Cl     /\   .  C1
                                     CL
                                   AICI3,A
                                               ci       \/-       ci

                                                         Cl

              In one manufacturing process, phenol is chlorinated under anhydrous
conditions, with aluminum chloride as the catalyst.  In this process,  three to four chlorine
atoms are added to the phenol.  The off-gas from the chlorinator (primarily HC1 vrth some
chlorine) is passed through a scrubber-reactor system containing excess phenol. The
temperature is held at the point where the chlorine is almost completely reacted to give the
lower chlorinated phenols, which may be either separated or fed back to the reactor for
further chlorination.

              PCP manufacture can produce a variety of CDD  via phenoxy radical reactions.
Specifically, phenoxy radicals are produced from decomposition of
polychlorocyclohexadienones produced by excess chlorination of tri-, tetra-, or PCP. The

                                         4-172
-------
electrophilic phenoxy radical attacks electronegative sites (ortho or para positions) on a
polychlorophenol molecule to form phenoxyphenols, which undergo further reactions to form
CDD.  In the PCP manufacturing process, chlorination is normally stopped when 3 to
7 percent tetrachlorophenol remains.  Further chlorination results in increased decomposition.

              CDD/CDF Emissions from PCP Wood Treatment-Concentrations of ODD/CDF
•vary greatly over time and are a function of heat, sunlight, and co-solvents.  PCP pressure
treatment of wood varies from facility to facility.  However, the treatment method generally
involves the following steps. First, the pre-cut wood is loaded into a pressure cylinder, which
is then filled with PCP dissolved in a petroleum solvent.  The cylinder is then pressurized
with steam until the required amount of the preservative has been absorbed.  The cylinder is
then depressurized and the preservative returned to storage, and the wood is  placed in a
vacuum to remove excess preservative.  In the final step, the wood is removed from the
cylinder and allowed to cool. Figure 4-34 displays a schematic of a pressure treating plant.97

              Wood treatment cylinders emit PCP  in the steam that is  released to the
atmosphere when the cylinder is opened to remove the treated lumber.  Typically, these
cylinders are opened only once  a day for a period of roughly 30 minutes.  Evaporative losses
of PCP from the hot wood surface as well as fugitive emissions from pipes and fittings occur
but are roughh two orders of magnitude less than the losses  from  the pressure cylinder.

              Emissions data from five wood treatment facilities were used  to develop a
2.3.T.8-TCDD TEQ emission factor of 7.06x1 CT6 Ib/ton  of PCP used.9"

              CDD Formation in DCP Manufacture—Commercial manufacture of DCP
involves the alkaline hydrolysis of trichlorobenzene.9"  1,2,4-trichlorobenzene is reacted with
sodium hydroxide in methanol at approximately 93°F (200°C) to yield the sodium salt of 2,5-,
2,4-, and 3,4-DCP, followed by acidification to produce the 2,5-, 2,4-,  and 3,4-DCP products.
                                         4-173
-------
               Condenser
Air-Vacuum

  Pump
                                       Cylinder
                          Control Panel
                                                                n
                 Mix Tank
                           Transfer Pump
                         Pressure Pump
  Storage Tank
                  o
                 Manual
 O
Automatic
                 Figure 4-34. Schematic Drawing of a Pressure Treating Plant



Source: Reference 96.
                                            4-174
-------
              The alkaline hydrolysis of 1,2,4-trichlorobenzene in the manufacture of PCP
 may produce CDD such as 2,7-dichlorodibenzo-p-dioxin.95  In addition to alkaline hydrolysis
 of 1,2,4-trichlorobenzene, the use of other chlorobenzenes in the manufacture of
 chlorophenols may form CDD contaminants as shown in Table 4-41.

             TABLE 4-41.  DIOXIN CONTAMINANTS ASSOCIATED WITH
                                 CHLOROBENZENES

    Chlorobenzene                                      Dioxin Contaminant
    1,2-Dichlorobenzene                                Dibenzo-p-dioxin
    1,2,3-Trichlorobenzene                         1,6-Dichlorodibenzo-p-dioxin
    1,2,4-Trichlorobenzene                         2,7-Dichlorodibenzo-p-dioxin
    1,2,3,4-Tetrachlorobenzene                             1,2,6,7-TCDD
    1.2,3,5-Tetrachlorobenzene                             1,3,6,8-TCDD
    1.2.4.5-Tetrachlorobenzene                             2,3,7,8-TCDD
   Source. Reference 94.

4 11 3        Brommated Compounds

             Extensive research on CDD CDF has produced much information on their
cnemistn   However, much less is known about their brominated counterparts   On the basis
of laboratory  studies on the formation  of polybrommated dibenzo-p-dioxms (BDD) and
polybrommated dibenzofurans (BDF) from certain brominated compounds, an assumption can
be made that  the mechanisms of dioxin and furan formation for brominated substances are
similar to those for chlorinated substances.
-------
             Information on the formation of BDD/BDF provided by one study
demonstrated that they could be formed during the chemical synthesis of flame retardams such
as 2,4,6-tribromophenol, pentabromophenol, and tetrabromobisphenol A.  Combustion of these
flame retardants also resulted in formation of BDD/BDF. High-resolution capillary column
gas chromatography coupled with mass spectrometry (HRGC/MS) analysis of
2,4,6-tribromophenol indicated the presence of di- and tri-BDD, TBDD, PeBDF, HxBDF,
HpBDF, and OBDF.94

             Table 4-42 contains profiles of some industrial brominated chemicals, and
include information on the manufacturing process, possible contaminants, and most likely
BDD/BDF isomers.  The predicted number of possible BDD/BDF formation pathways are
given in Table 4-43.

4.12         PORTLAND  CEMENT PRODUCTION

             Most of the hydraulic cement produced in the United States is Portland cement.
which is a cementitious,  crystalline compound composed of metallic oxides. The end-product
cement, in its fused state, is  referred to as "clinker."  Raw materials used in the process can
be calcium carbonate- and aluminum-containing limestone, iron, silicon oxides, shale, clay,
and sand.9>   There are four primary components in Portland cement manufacturing,  ra\\
materials handling, kiln feed preparation, pyroprocessmg. and finished cement grinding
Pyroprocessmg. the fuel  intensive process accomplished in cement kilns, has been identified as
a potential source of CDD/CDF emissions and constitutes the primary focus of this chapter.

4.12.1        Process Description

             In Portland cement production, most raw materials typically are quarried on site
and transferred by conveyor to crushers and raw mills.  After the raw materials are reduced to
the desired particle size,  they are blended and  fed to  a large rotary kiln. The feed enters the
kiln at the elevated end,  and the burner is located at the opposite end.  The raw materials are
then changed into cementitious oxides of metal by  a  countercurrent heat exchange process.
                                        4-176
-------








00
O
"7
TFD COMPOUI
<
z

I
/— v
W/
'V'
w>

<

«


r f-
^/,
— -
z


r-,
rt



<















tn
t
O
ws
U.
O
e
Q
O
CO
>*
5
j
I




ible C'ontaminants
V.
V.
r*
L»






t/~
LT

^
^
--
Z
^
, ^

r;

^_








•o
c.
E
_c
^





c
'x
o
-Dibromodibenzo-p-di
r-.
r^T


•ibromophenol;
bromoplienol; and
' tetrabromophenol
labromophenol.
- ~^ &
*Z w. ? C"
•v V £ "g
- *. C
rsj rs C. f3

o
~
re




—
i:
_
_Z "^
c "^
= -5
— 7
~ —

— y
Z- C.




c
c
o
"E.
c
E
e
£
T
rl




C
'x
o
•B
B.
s
,7,9-Tetrabromodiben;
^
*.
~ y
*- £
— >. -^
^ ^~ ^
i s^
t- c £
i_ C- re
z - X

— "c ^
^ W. f-
i - "
£ re re
C is t

^
C
c
o
f
c.
c
t
c
i
VC
V
»N"
w
o
*D
•a S
E M
go
>.o
•go
§ *°
Itc
favorable kinetics and
iditions. Low probabi
>Fs formation. Possib
CPU
D oCD


dibromophenol .
e
S
El

^
o
_c
c
re




•-
i
f
_2
^
0 ^
3" ^
C "~

^ w
E ^
« ">,
s x
II
•c
c
TO
.•
,L
C

C
(/
C
c
S! 1?
"c. £
u*> t>
1 1
C ^
t» '™~
C D.




U
U
1
,e
^
o
I
£
•S
i
.0
1
o
E






































oil
fe'C
•£ u
°S
1 E
s s
U «9
I§
11
'£ o
U 60
Id
«2
£E
v o
^^=
* v>
£ "
E i*1
o s-
Z o


romo-4-nitrophenol
^
2
ir,
fN
r-<
b<
C
IT
\T.
SJ
U
>;
o

-C
— >-,

< JB
V
T ""
C t/
C O
0 0
"E. 2
W V
c >-»
c-Z
u- 5
^ .c
^ c
c ^~
~~ ^
P5 "~
E ^
™ ^^
P j^
S TO
e B






•4
o
P"3
c c
w flj
•e^
9 §•
* i
rs" E






I
a
|
1
i
a
tj
O

at
yloxide with varyin
of bromination.
fe <«
Is
C- OXJ
C-c
^r
«
C->
** C f •
re ^ U-
«- ^ _ ^
— u C O
r: c ra ^r
£ f :>, "~
^- X*
O C; -^ "^

•c Q.T: =
x ^ ~
JS Rl -. C
C S: # >-
~ ~- c
- C "r C
- ~ ^ I
C- ^ ^i =
^ = 0 ^
c S^ £
^ <-) _. p-
T X c
S 0 — o
1 E^ E
i |", 1.-
»— &. 1_ f ^
-^ ^ r* fc c
I?|E!
0
*o
X
o

^~\
c
U
J^
•5
o
c
i.
TO
(_j
O
C


ii
g
U
N
1
O
1
fc.
«
"3
U
1
U
i
i
B.
U
O


^£
i« N
V C
6J5
0 -o
•o c
c°c
'E-'i
« £
> i

w
l>
U. ^^
•a i—
c ^
« S
c
r-, _--

£ c
^
it
o u.
C D
u rvi
N —
c r^
u
-° £
c ^c
r-
c —
V tT
re >>
E 5
E ^
c ^
£ re
O
"C
><
_c

>,
c
0
f
•c
o
J5
re
~

O

u
,£
u ''K
•S °
§1
•0 ~
e 2
E w
«2 —
«'§
)Ds and BDFs could I
inbustion conditions si
I— n
m S


; brominated
;
.EC
>N 4J
fe-l
> £


c
Si c
S ~7
~ —
c C
E S
G C
i £
£ £
C. 0.




ft^
C
~
"e
o
o
k—
TO
C
4^
c-

IA
i 1
C ^5
|l|
S c fe
s o •*
*2 w) t!
CO T3 ^
M C S
sr v u
lorinated benzenes in 1
e specific isomers dep
ndensation of the parti
15 JC n
*• r f
u r- o






c
«
c
Q

^^


re
u
E
*"
tr
LJ
_i;
re
re
o
(«",
r-~
p-.' ~
oc o:
rs c^
TT ^













•o
«
15
g
£
£«•
e -e
tl
1*8.
g s
C (0
=5 g
2 2
^2
8-g
tf:
c c
D 'S
3 5
•• o
S'-v




































4-177
-------













,_,
D
in
P
C
U
s— '
'OMPOUNDS
W^
f*^

i—
Z
2
/*v
^X
™
»_
— .

"" - •= <> ri _ e*


o

^
"E.
c

— ^
ro »^
^ C
- <
e
V
,2

•g
p
0
^5
rt
1
TS
CO
i/l
1
1

1
V
u
C/3



mobenzene and
omotoluene.
C £
o c
u flj
CO C.





c
c
^
*~
[5
re
t/
£. u
t from the
MTiobcnzer
u r
l_^-* -*—
c —
f~ L/~
^"1
e -
V
C
K
C
(L>
o
|
•I
H

r*j
— '








u
£
N
1
o
0
i
£
Tt^
V
V
C/3

4)
C
,4-tribromoben7.e
fN
— '
u
v:
c'^'
< « "O
^ "^
^ " ^
S; C ^"
= E c
*- ^" *r
*— re
*- *j —
- E §• ^
o i § 1
C .1 rj u
o . t-
N r<-, <£ O
^~|?,
u_ u i_
C C rc Ai
C "E ** £
(^ >-^ ^^ <~
= 5 o 7
^_ — •*" ^_
§ u = -;
^r (/•-->
— re 5 —
s.
£
N.
C
X;
C
E
O
|
*_
vr;
r*~,
—







g
'§
•5
c.
J2
g
£
o
1
S



"5
f
u
ra
C
1

C
^ o
^0

rr 5
— C k-
?= o ^
C^J fx] P
c .T —
c c~ C
'~ T* C
< c c _
"C !- -^ ^
£ £ = c
r o o
_^ oc c r-
5r — CD ^
«C •— w~,
— < W":
TO C ^~
E ^ •*• £

£ -|^ ^

<
, 	
^
CJ
4:
v;
15
c
o
^
E
^
H












-































•
Qv
U
U
C
^JJ
c;
B:

u
3
O
4-178
-------
     TABLE 4-43.  BROMINATED COMPOUNDS WITH THE POTENTIAL FOR
                            BDD/BDF FORMATION
                                            Predicted Number of Possible BDD/BDF
 Chemical Name                                     Formation Pathways
 2,4-Dibromophenol                                          5
 2,4,6-Tribromophenol                             -           2
 o-, m-, and p-Bromophenol                                   2
 2,6-Dibromo-4-nitrophenol                                    1
 Decabromodiphenyloxide                                     1
 Octabromodiphenyloxide                                     1
 Bromobenzene (mono- and di-)                               Oa
 Pentabromotoluene                                          0^
 Tetrabromophthalic anhydride                                 0
 1.2,4-Tribromobenzene                                       Oa-b
 1,3,5-Tribromobenzene                                       Oa'b
 Pentabromodiphenyloxide                                     1
 Tetrabromobisphenol A                                       1
Source Reference 94
a  Under combustion  conditions, brommated benzenes could condense to BDD and BDF similar 10
  chlorinated benzenes
r  BDD and BDF could be formed under combustion conditions similar to those of chlorinated benzenes in
  PCB transformers
                                    4-179
-------
The materials are continuously and slowly moved to the low end by the rotation of the kiln
while being heated to high temperatures (2,700°F [1,482°C]) by direct firing.  In this stage,
chemical reactions occur, and a rock-like substance called "clinker" is formed. This clinker is
then cooled, crushed, and blended with gypsum to produce Portland cement.98 The cement is
then either bagged or bulk-loaded and transported out.98

              Cement may be made via a wet or a dry process.  Many older kilns use the wet
process.  In the past, wet grinding and mixing technologies provided more uniform and
consistent material mixing, resulting in a higher quality clinker. Dry process technologies
have improved, however, to the point that all of the new  kilns since 1975 use the dry
process.100  In the wet process, water is added to the mill while the raw materials are being
ground.  The resulting slurry is fed to the kiln.  In the dry process, raw materials are also
ground finely in a mill,  but no water is added and the feed enters the kiln in  a dry state.

              More fuel is required for the wet process then the dry  process to evaporate the
water from the feed.  However, for either the wet or dry  process,  Portland cement  production
is fuel-intensive.  The primary fuel burned in the kiln may be natural gas, oil, or coal.  Many
cement plants burn supplemental  fuels such as waste solvents, chipped rubber, shredded
municipal garbage, and  coke.9f A major trend in the industry1 is the increased use of waste
fuels.  In 1989. 33 plants in the United States and  Canada reported using waste fuels: the
number increased to  55  plants m  1990.98

              The increased use of hazardous waste-derived  fuels (HWDFs) for  the kilns is
attributed to lower cost  and increased availability.  As waste generators reduce or eliminate
solvents from their waste steams, the streams contain more sludge and solids.  As a result,
two new hazardous waste fueling methods have emerged  at cement kilns. The first method
pumps solids (either  slurried with liquids or dried and  ground) into the hot end of the kiln.
The  second method (patented by cement kiln processor and fuel blender Cadence, Inc.)
introduces containers of solid waste into the  calcining zone of the kiln.100
                                         4-180
-------
              The kiln system for the manufacture of Portland cement by dry process with
 preheater is shown in Figure 4-35.  The raw material enters a four-stage suspension preheater,
 where hot gases from the kiln heat the raw feed and provide about 40-percent calcination
 before the feed enters the kiln.  Some installations include a precalcining furnace, which
•provides about 85-percent calcination before the feed enters the kiln.98

              Facilities that burn HWDF are subject to the Boilers and Industrial Furnaces
 (BIF) rule under the Resource Conservation and Recovery Act (RCRA) promulgated
 February 21, 1991. The BIF rule requires that a facility that burns hazardous waste
 demonstrate a 99.99 percent destruction efficiency for principal organic hazardous constituents
 in the waste stream.  To guard against products of incomplete combustion, the BIF rule limits
 CO levels in the kiln and or total hydrocarbon levels in stack gases.100-101  In addition, a
 NESHAP for control of HAPs from Portland cement kilns is under development by the
 Emission Standards Division  of OAQPS.

 4.12.2         Emission Control Techniques

              Fuel combustion at Portland  cement plants can emit a wide range of pollutants
 in small quantities. If the combustion reactions do not reach completion. CO and YOCs can
 be emitted  When waste fuels are  burned, incomplete  combustion can lead to emissions  of
 specific HAPs. such as CDD'CDF.  These pollutants are generally emitted at lo\\ le\els

              In the pyroprocessmg units, control devices employed are  fabric filters (reverse
 air. pulse jet. or pulse plenum) and ESPs  Typical control measures for the kiln exhaust are
 reverse air fabric filters with  an air-to-cloth ratio of 0.41:1 meter per minute (nvmm)
 (1.5.1 acfm/fr) and ESPs with a net specific collecting area (SCA) of 1,140  to 1.620 square
 meters per thousand m3 (nr/1,100 m3) (350 to 500 square feet per thousand ft' [fr 1.000 ft?])
 Clinker cooler systems are controlled most  frequently with pulse jet or pulse plenum fabric
 filters.  The potential for secondary CDD/CDF formation exists within the ESP.  CDD/CDF
 formation can occur in the presence of excess oxygen  over a wide range of temperatures.
 Refer to Section 4.1.1 for a detailed discussion.
                                         4-181
-------
                                                             ra
                                                             oj
                                                             QJ
                                                             £
                                                             V)
                                                              DC
                                                              C
                                                              *ul

                                                              tj
                                                              uS
                                                               3

                                                               CO
                                                               c
                                                               OJ

                                                               CJ
                                                               u
                                                               •a
                                                                CB
                                                                 CJO
                                                                 U
                                                                 
-------
 4.12.3       Emission Factors

              The raw materials used by some facilities may contain organic compounds,
•which constitute a precursor to potential CDD/CDF formation during the heating step.
^iowever, fuel combustion to heat the kiln is believed to be the greater source of CDD/CDF
 emissions.  The data collected and presented in this section indicate that CDD/CDF are
 emitted when either fossil fuels, HWDFs, or combinations of the two are combusted in the
 kiln.98'101

              CDD/CDF emissions data for Portland cement kilns with various process, fuel,
 and control configurations were compiled by the U.S. EPA's Office of Solid Waste in 1994.10!
 Testing was conducted  at 35 Portland cement manufacturing facilities to certify compliance
 with the BIF Rule.  Emission factors  developed from the study are presented in Tables 4-44
 and 4-45.

              It should be noted that  Table  4-44 presents emission factors  for kilns that bum
 hazardous waste (HW)  and also kilns  that do not burn hazardous waste (NHW)   In addition,
 this document presents  separate  emissions estimates for HW and NHW kilns
              Source Locations
              The Portland cement manufacturing industry is dispersed geographically
 throughout the United States.  Thirty-six states have at least one facility.  As of
 December 1990, there were 112 operating Portland cement plants in the United States.
 operating 213 kilns with a total annual clinker capacity of approximately 80 million tons
 (73.7 million Mg).  The kiln population included  80 wet process kilns and 133 dry process
*kilns.9f Table 4-46 presents the number of Portland cement plants and kilns in the United
 States by State and the associated production capacities as of December 1990.
                                         4-183
-------
  TABLE 4-44. CDD/CDF EMISSION FACTORS FOR DRY PROCESS PORTLAND
                               CEMENT KILNS
                               SCC 3-05-006-06
                        FACTOR QUALITY RATING:  D
Fuel Type
Control Device
     Pollutant
  Average Emission
   Factor in Ib/ton
      (kg/Mg)a
Natural Gas
 Electrostatic
 Precipitatorb
2,3,7,8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
1.20x10-'° (6.00x10-")
4.93xlO'9 (2.47xlO-9)
1.12xlO-9(5.60xlO-10)
4.73X10'10 (2.37x10-'°)
3.07xlO-'°(1.54xlO-10)
1.98xlO'9 (9.90xlO-10)
8.81x10'° (4.41xlO'9)
2.98x10-° (1.49x10-°)
2.24xlO-8(1.12xlO-s)
4.29x10'° (2.15xlQ-9)
5.47x10-'° (2.74xlO'lu)
3.55X10'11 (1.78x10-")
6.62x10'" (3.31xlO-r'i
2.72x10^ (1.36x10^)
Coal Coke
 Electrostatic
 Precipitator
2.3,7.8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
5.72x10'" (2.86x10'")
 1.73x10'" (8.65x10'")
 2.62x10-' (1.31x10-°)
 2.53xlO"'J (1.27xlO-9)
 1.61X10'9 (8.05x10''°)
 1.42xlO-9(7.10xlO-'°)
 l.OlxlO'8 (5.05x10-°)
 4.48x10'° (2.24xlO-9)
 2.43xl04 (1.22xlQ-8)
                                    4-184
-------
   TABLE 4-44.  CDD/CDF EMISSION FACTORS FOR DRY PROCESS PORTLAND
                         CEMENT KILNS (CONTINUED)
Fuel Type
Control Device
     Pollutant
  Average Emission
   Factor in lb/ton-
      (kg/Mg)a
Coal/Coke,
continued
 Electrostatic
 Precipitator,b
  continued
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
 1.20xl(rs (6.00xl(T9)
 7.79xlO-9.(3.90xlO-*)
 2.12x10'' (1.06x10*)
2.86xlO-'°(1.43xlO-10)
 4.64xlO'8 (2.32x10-*)
Coal
 Electrostatic
 Precipitator0
2,3,7,8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
 5.14xlO"8 (2.57x10-*)
 2.54x10-°( 1.27x10-*)
 2.67x10'" (1.34x10-*)
 2.73X10'6 (1.37xlO-6)
 5.42X10'7 (2.71xlO-7)
 8.87xlO'8 (4.44xlO'8)
 8.62x10-° (4.31xlO-fa)
 3.30xlO'7 (1.65xlO-7)
 8.45x10'° (4.23x10-")
 4.53x10'° (2.27x10-")
 1.71x10-° (8.55x10-')
 3.15xlO'7 (1.58x10°)
 5.08x10'" (2.54x10-*)
 1.54xlO'5 (7.69x10'*)
Coke/Hazardous
Waste
 Electrostatic
 Precipitatord
2,3,7,8-TCDD

Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
 3.42xlO'8 (1.71xlO-8)

 5.86xlO-7 (2.93xlO'7)
 4.67x10'7 (2.34xlQ-7)
 4.59x10'7 (2.30x10'7)
 5.45xlO-K (2.73x10-*)
 l.SSxlO'8 (9.40x10'9)
 1.62xlO-6(8.10xlO-7)
                                     4-185
-------
    TABLE 4-44.  CDD/CDF EMISSION FACTORS FOR DRY PROCESS PORTLAND
                            CEMENT KILNS (CONTINUED)
Fuel Type
Coke/
Hazardous
Waste,
continued





Coke












Control Device Pollutant
Electrostatic 2,3,7,8-TCDF
Precipitator,d continued
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
Multicyclone/Fabric 2,3,7,8-TCDD
Fllter' Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
Average Emission
Factor in Ib/ton
(kg/Mg)'
2,07xlO-7 (K04xlO-7)
5.18x10-* (2.59xlO-6)
9.95xlO'7 (4.98xlO-7)
5.70xio-7 (2.85xlO-7)
2.28xlO-8(1.14xlQ-8)
7.74xlO'9 (3.87x1 0-9)
6.98x10-° (3.49xlO'6)
1.19x10-" (5.95xlO-12)
2.47x10-" (1.24xlO-9)
3.28xlO-9(1.64xlO'9)
7.67xlO'9 (3.84xlO'9)
8.52xlO-10(4.26xlO-10)
1.88xlO-8(9.40xlO'9)
3.31xlO-8(1.65xlO-8)
5.83x10-'° (2.92x10-'°)
3.80xlO'9 (1.90xlQ-9)
7.84xlO-'° (3.92x10-'°)
2.25xlO-'°0.13xlO-10)
4.68x10-" (2.34x10-")
2.81xlO-'°(1.41xlO-10)
5.72xlO'9 (2.86xlO-9)
Source: Reference 101.
a Emission factors in Ib (kg) of pollutant emitted per ton (Mg) of clinker produced.
° Kiln operating conditions: high combustion temperature; minimum electrostatic precipitator power.
c Kiln operating conditions: maximum feed, kilns 1 and 2.
d Kiln operating conditions: maximum hazardous waste feed.
e Kiln operating conditions: high combustion temperature; maximum production.

                                          4-186
-------
        TABLE 4-45. CDD/CDF EMISSION FACTORS FOR WET PROCESS
                          PORTLAND CEMENT KILNS
                                SCC 3-05-007-06
                        FACTOR QUALITY RATING:  D
Fuel Type
  Control Device
        Pollutant
  Average Emission
   Factor in Ib/ton
      (kg/Mg)a
Coal
Electrostatic
Precipitator*1
2,3,7,8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2.3,7,8-TCDF
Total TCDF
Total PeCDF
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
 6.41x10'9 (3.21X10'9)
 4.74xlO'7 (2.37xlO'7)
 6.63xlO'7 (3.32xlO'7)
 6.52xlO'7 (3.26xlO-7)
 1.02xlO'7 (S.lOxlO'8)
 2.21xlO'7 (l.llxlCr7)
 2.12xlO-6 (1.06x10-")
 3.73xlO'8 (1.87x10'*)
 1.77x10'" (8.85x10-*)
 9.52xlO'8 (4.76x10'')
 1.21x10'" (6.05x10'*)
 2.90x10'" (1 45x10'')
 7.92x10'' (3.96x10")
 4.67x10'" (2.34x10'")
Coal Coke
Electrostatic
Precipitator"
2.3,7.8-TCDD
Total TCDD
Total PeCDD
Total HxCDD
Total HpCDD
Total OCDD
Total CDD
2,3,7,8-TCDF
Total TCDF
Total PeCDF
2.65x10 a (l.
 1.13x10'" (5 65x10'*)
 1.61x10'' (8 05x10'")
 2.77x10'" (1.39x10'")
 3.50x10''(1.75xlO's)
 1.02x10-* (S.lOxlO'9)
 5.96x10'" (2.98xlO'7)
 1.55x10* (7.75xlO'9)
 8.07x10"' (4.04x10'*)
 3.37xlO-?(1.69xIO-8)
                                     4-187
-------
         TABLE 4-45.  CDD/CDF EMISSION FACTORS FOR WET PROCESS
                    PORTLAND CEMENT KILNS (CONTINUED)
 Fuel Type
  Control Device
        Pollutant
  Average Emission
   Factor in Ib/ton
      (kg/Mg)a
 Coal/Coke,
 continued
Electrostatic
Precipitator,b
continued
Total HxCDF
Total HpCDF
Total OCDF
Total CDF
 1.03xlO'8 (S.lSxlO-9)
 2.27x10-9(1.14xlO-9)
2.42x10-'° (1.21x10-'°)
 1.27xlO'7(6.35xlO-8)
Electrostatic
Precipitatorb
                                 2,3,7,8-TCDD
                                 Total PeCDD
                                 Total HxCDD
                                 Total HpCDD
                                 Total OCDD
                                 Total CDD
                                 2,3,7,8-TCDF
                                 Total TCDF
                                 Total PeCDF
                                 Total HxCDF
                                 Total HpCDF
                                 Total OCDF
                                 Total CDF
                           4.17x10-'° (2.09x10-'°)
                            2.08x10'7 (1.04xlO-7)
                            2.96x10-7(1.48xlO'7)
                            8.51xlO-7 (4.26xlO-7)
                            1.02xlO'7 (S.lOxlO-8)
                            2.54xlO'8 (1.27xlO-8)
                            1.48xlO-6 (7.41x10-")
                            2.54x10'" (1.27x10'*)
                            1.34x10'" (6.70x10'")
                            5.00x10'" (2.50x10'")
                            2.31x10'" (1.10x10'")
                            8.71x10''' (4.30x10'")
                            4.30x10'9 (2.15x10'")
                            2.46x10'" (1.23x10'")
Source  Reference 101
a Emission factors in Ib (kg) of pollutant emitted per ton (Mg) of clinker produced.
b Kiln operating conditions: high combustion temperature, minimum electrostatic precipitator power
                                       4-188
-------
TABLE 4-46. SUMMARY OF PORTLAND CEMENT
      PLANT CAPACITY INFORMATION
Location
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
lo\\a
Kansas
Kentuck\
Maine
Maryland
Michigan
Mississippi
Missouri
Montana
Nebraska
Nevada
New Mexico
New York
Ohio
Number of Plants
(kilns)
5(6)
1 (O)1
2(7)
2(5)
12 (20)
3(5)
6(8)
2(4)
1 (1)
1 (2)
4(8)
4(8)
4(7)
4(11)
1 (1)
1 (1)
3(7)
5 (9)
1 (1)
5(7)
2(2)
1 (2)
1 (2)
1 (2)
4(5)
4(5)
Capacity
103 tons/yr(103 Mg/yr)
4,260 (3,873)
0 (0)
1,770 (1,609)
1,314(1,195)
10,392 (9,447)
1,804 (1,640)
3,363 (3,057)
1,378 (1,253)
263 (239)
210 (191)
2,585 (2,350)
2.830 (2,5^3)
2.806 (2.551)
1.888 (iriG)
"24 (658)
455 (414)
1.860 (1.691)
4.898 (4.453)
504 (458)
4.677 (4.252)
592 (538)
961 (874)
415 (377)
494 (449)
3,097 (2,815)
1.703 (1,548)
                 4-189
-------
              TABLE 4-46. SUMMARY OF PORTLAND CEMENT
              PLANT CAPACITY INFORMATION (CONTINUED)
Location
Oklahoma
Oregon
Pennsylvania
South Carolina
South Dakota
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wyoming
Number of Plants
(kilns)
3(7)
1(1)
1 1 (24)
3(7)
1(3)
2(3)
12 (20)
2(3)
1 (5)
1 (1)
1 (3)
1 (1)
Capacity
103 tons/yr (103 Mg/yr)
1,887(1,715)
480 (436)
6,643 (6,039)
2,579 (2,345)
766 (696)
1,050 (955)
8,587 (7,806)
928 (844)
1,117(1,015)
473 (430)
822 (747)
461 (419)
Source  Reference 98

- >,::',r,amu plan', ortK
                                4-190
-------
                                    SECTION 5.0
                            SOURCE TEST PROCEDURES

             Several sampling and analysis techniques have been employed for determining
CDD and CDF emissions.  Measurement of CDD and CDF 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.  The purpose of
this section is to present basic  sampling and analysis principles used to gather emissions data
on CDD and CDF from stationary sources. The  presentation of non-EPA methods in this
report does not constitute endorsement or recommendation or signify that their contents
necessarily  reflect EPA's views and policies.

5.1           SAMPLE COLLECTION

             Collection of CDD and CDF from stationary sources is achieved by using  a
sampling system that captures both paniculate and  condensibles. The most prevalent method
is EPA Method 0010, also referred to as the Modified Method 5 (MM5) Sampling Tram,
\\hich is equipped with  a sorbent resin for collecting  condensibles. A schematic of the MM5
sampling tram used for  collecting CDD and CDF is shown in Figure 5-1.

             The Source Assessment Sampling  System (SASS),  a high-volume variation of
MM5 capable of sampling paniculate and vapor  emissions from stationary sources has also
been used.102-103  A schematic of the SASS train is  shown in Figure 5-2.
                                         5-1
-------
                                                        Fitter Hoktor
                                                               Sorbcnt Trap
R*y«r»e-Typ« Prtot Tub*
                                                                                             Vacuum Line
                                          Dry Gas Meter   Air Tight Pump
                      Figure 5-1.  Modified Method 5 Sampling Train Configuration

      Source: References 102,103.
                                                    5-2
-------
                                          Isolation Ball
                          Convection Oven     Valve      Bitter
S-Style Pr
                                           | J^   Sorbent
                                          Oven Cartridge
                                           T.C.
                                                               Gas
                                                              Cooler
                                                                     Gas
                                                                   Temper-
                                                                    ature
                                                                     T.C.
                                               Condensate
                                                 Collector
                                                                       IMP/Cooler
                                                                     Trace Element
                                                                        Collector
            Dry Gas Meter/Orifice Meter :
             Centralized Temperature  j~
              And Pressure Readout   !
                 Control Module
                                         Two 10 ft/mm Vacuum Pumps

                           Figure 5-2.  Schematic of a SASS Train
                                                                                             0.
                                                                                             i
Source:  References 102,103.
                                              5-3
-------
         TABLE 5-1.  COMPARISON OF MM5 AND SASS CHARACTERISTICS
Characteristics
Inert materials of construction
Percent isokineticity achievable
Typically used to traverse
Particle sizing of sample
Sample size over a 4-6 hour period (dscm)
Sampling flow rate (dscmm)
Source. References 103
a Assuming reasonably uniform, non-stratified flow.
MM5
Yes
90- 110
Yes
No
3
0.02 - 0.03


SASS
No
70 - 150a
. No
Yes
30
0.09 - 0.14


volume of gas collected.  The smaller gas volume used in the MM5 train (about 10 times
smaller than the SASS train) requires less XAD-2* resin.  Also, in a SASS train, the  filter is
normally held at a higher temperature than in the MM5 train.102'103

             In the MM5 sampling train, a water-cooled condenser and XAD-2*  resin
cartridge are placed immediately before the impinger section. XAD-2* is designed to adsorb
a broad range of volatile organic species.  The gas stream leaving the filter is cooled  and
conditioned in the condenser prior to entering the sorbent trap, which contains the XAD-21*
resin   From the sorbent trap, the  sample  gas is routed  through impingers, a pump, and a dr\
gas meter.  The MM5 train is designed to operate at flow rates of approximately 0 015 dry
standard cubic meters per minute  (dscmm) or, equivalently. 0.5 dry  standard cubic feet per
minute (dscfm) over a 4-hour sampling period.  Sample volumes of 3 dscm (100 dscf) are
typical.103

             An  advantage of using the  MM5 train is that it is constructed of inert materials.
A disadvantage is  that long sampling periods (2 hours) are required  to collect sufficient
sample for quantitative analysis.
                                         5-4
-------
               The SASS train is a multi-component sampling system designed for collecting
 participates, volatile organics, and trace metals.  The train contains three heated cyclones and
 a heated filter that allow size fractionation of the paniculate 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
 factions.  The system is designed to operate at a flow rate of 0.113 scmm (4.0 scfm), with
 typical sample volumes of 30 dscm (1,000 dscf).102

              An advantage of the SASS  train is that a large quantity of sample is collected.
 A disadvantage is that the system does not have  the ability to traverse the stack.  Also,
 because constant flow is required to ensure proper size fractionation, the SASS train is less
 amenable to compliance determinations because isokinetic conditions are not achieved.
 Another disadvantage is the potential of corrosion of the stainless steel components  of the
 SASS train by acidic stack gases."

              Other methods that have been used to collect and determine concentrations of
 CDD and CDF from stationary sources are EPA  Reference Method 23,104 California Air
 Resources Board (CARB) Method 428.'6-  and a draft ASME {American Society of Mechanical
 Engineers) protocol that was distributed in December  1984.IOe

              EPA Reference Method 23  is  a combined sampling and analytical method  that
 use? a sampling tram identical to the one described in EPA Method 5, with  the exceptions and
 modifications specified in the method.iw

              CARB Method 428 is another combined sampling and analytical method.  With
 .Method  428, paniculate and gaseous phase CDD and CDF are extracted isokinetically from a
•stack and collected on a filter, on XAD-2* resin, in  impingers,  or in upstream sampling train
-components. The sampling train in Method  428  is similar to the train in EPA Method  23
 except that the CARB Method 428 train includes one impinger containing either water or
 ethylene glycol and the EPA Method 23 train includes two impingers containing water.
                                           5-5
-------
             The draft ASME protocol assumes that all of the compounds of interest are
collected either on XAD-2* resin or in upstream sampling train (MM5) components.  The
minimum detectable stack gas concentration should generally be in the nanogram/cubic meter
(ng/m3) or lower range.

5.2          SAMPLE RECOVERY AND PREPARATION

             Quantitative recovery of CDD and CDF requires the separation of these
compounds from the remainder of the collected material, as well as efficient removal from the
collection media.  A technique commonly used for recovery of CDD and CDF from filters.
and adsorbent and liquid media is soxhlet extraction.

             Most recovery  methods entail (1) the addition of isotopically labeled internal
standards and/or surrogate compounds, (2) concentration of the sample volume to 1-5 mL or
less. (3) sample cleanup involving a multi-column procedure, and (4) concentration of the
sample to the final desired volume.

             EPA Method 23, CARB Method 428, and the draft ASME protocol employ
soxhlet extraction  for the extraction of CDD and CDF from both filter and sorbent catches.
The primary difference among these methods is in the recovery solvents used.

             EPA Method 23 sample recover.' solvent for rinsing the sample train glassware
is acetone, with a  final quality assurance rinse of toluene; however, the results from the
toluene rinse are not used  in calculating total CDD and CDF emissions. Toluene is used  for
soxhlet extraction.  The columns used for sample cleanup include silica gel, modified silica
gel, basic alumina, and carbon/celite.

             In CARB 428,  sample recovery solvents for rinsing  the sample train glassware
include methanol,  benzene, methylene chloride and distilled deionized water.  The filter and
sorbent catches are dried with sodium sulfate (NA2SO4) prior to soxhlet extraction with
                                         5-6
-------
benzene or toluene. A minimum two-column cleanup system (silica gel/alumina) is required.
A third column, the carbon/celite cleanup procedure, may be necessary.

             The draft ASME protocol specifies the use of acetone followed by hexane as
sample recovery solvents and toluene as the soxhlet extraction solvent.  Silica and alumina
column cleanup procedures are the minimum requirement, with cleanup on carbon/celite and
silica/diol columns if necessary.

             EPA Reference Methods 8280 and 8290 are analytical methods used to
determine TCDD/TCDF through OCDD/OCDF in chemical wastes, including still  bottoms,
fuel oils,  sludges,  fly ash, reactor residues, soil, and water. Both methods involve the addition
of internal standards to the sample prior to a matrix-specific extraction procedure as specified
in the method.  The extracts are submitted to an acid-base washing treatment and dried.
Following a solvent exchange step, the residue is cleaned up by column chromatography  on
neutral alumina and carbon on celite.lo: I05

             EPA Method 8280 employs seven  13C labeled standards and one 3:C1 labeled
standard.  Two are used as recover.' standards, five are used as internal standards,  and one is
used as a cleanup  standard.  There are no labeled standards for the PeCDD/PeCDF
homologues. the HxCDF homologues. and the HpCDD homologues.  This means that for
these CDD CDF homologues, the efficiency of the extraction and  cleanup procedures cannot
he measured K  The method  does not include the use of surrogate standards, which are
normal!}  added to the adsorbent trap, because Method 8280  is only an analytical method

             EPA Method 8290 employs eleven I3C  labeled standards.  Two are used as
recovery standards and nine are used as internal standards.  TCDD/TCDF through
HpCDD/HpCDF and OCDD are represented by the internal standards.  This use of standards
allows for monitoring  of all but OCDF for extraction  and cleanup recoveries.'06'10"
                                         5-1
-------
5.3          QUANTITATIVE ANALYSIS

             The analytical techniques employed to identify and quantify CDD/CDF in
environmental samples include high-resolution capillary column gas chromatography coupled
with low-resolution mass spectrometry (HRGC/LRMS) or high-resolution mass spectrometry
(HRGC/HRMS).  EPA Method 8280 and the draft ASME protocol use HRGC/LRMS; EPA
Method 23 and EPA Reference Method  8290 use HRGC/HRMS.  Either technique can be
used in CARB 428.

             Separation of isomers or series of isomers is accomplished by HRGC, and
quantification is accomplished by operating the mass spectrometer in the selected ion
monitoring (SIM) mode.  A high-resolution fused silica capillary column (60 m DB-5) is used
to resolve as many CDD/CDF isomers as possible; however, no single column is known to
resolve all isomers.

             Identification is based on  gas chromatograph retention time and correct chlorine
isotope ratio.  Strict identification criteria for CDD/CDF are listed in each individual method.
Quantification generally involves relative response factors determined from multi-level
calibration standards.  An initial calibration curve is required prior to the analysis of any
sample and then intermittent calibrations (i.e., analysis of a column performance-check
solution and  a mid-range concentration solution) are performed throughout sample analyses.

             In the draft ASME protocol, two different columns are required if data on  both
2.3.7.8-TCDD and 2,3,7,8-TCDF. as well as on total CDD/CDF by chlorinated class, are
desired.  The appropriate columns are a  fused silica capillary column (60 m DB-5) and a
30 m capillary (DB-225). The DB-5 column is used to separate several groups of
TCDD/TCDF through OCDD/OCDF.  Although this column does not resolve all of the
isomers  within each chlorinated group, it effectively resolves each of the chlorinated groups
from all of the other chlorinated groups, thereby providing data on the total concentration of
each group (that is,  total TCDD/TCDF through OCDD/OCDF). The DB-225 column is used
                                        5-8
-------
 to quantify 2,3,7,8-TCDF because it adequately resolves this isomer from the other TCDF
 isomers.106

              The capillary gas chromatographic columns recommended in CARB Method
.428 and EPA Method 23 include the 60 m DB-5 or the SP-2331.  The peak areas for the two
 ions monitored by the mass spectrometer for each analyte are summed to yield the total
response for each analyte. Each internal standard is used to quantify the CDD/CDF in its
 homologous series.

              A DB-5 (30 m) or SP-2250 capillary column is recommended for EPA
 Methods 8280 and 8290. The analytical procedures specified in EPA Method 8290 are
 similar to those in Method 23, with the addition  of the surrogate standards used to  measure
 sample collection efficiency.  Identification of the compounds for which an isotopically
 labeled standard is used is based on elution at  the exact retention time established by analysis
 of standards and simultaneous detection of the two most abundant ions in the molecular ion
 region.  Compounds for which no isotopically  labeled standard is available are identified by
 their relative retention times, which must fall within the established retention windows, and
 the simultaneous detection of the two most abundant ions in the molecular ion region in the
 correct abundance ratio.  The retention windows  are established by analysis of a
 GC performance evaluation  solution   Identification  is confirmed by comparing  the ratio of
 the integrated ion abundance of  the molecular  ion species to the theoretical abundance ratio.

             Quantification of  the individual  homologues and total CDD'CDF is based upon
 ^ multi-point calibration curve for each homologue.10"10*1  The major difference  between
 Method 8280 and Method 8290  is the resolving power of the mass spectrometer. The HRMS
proMdes a higher-quality analysis than does the LRMS because of its ability to  incorporate
additional labeled standards  to cover almost all of the TCDD/TCDF through OCDD/OCDF
 homologues.
                                         5-9
-------
                                     SECTION 6.0
                                    REFERENCES
 1.     U.S. EPA.  Compilation of Air Pollutant Emission Factors, 5th ed. (AP-42), Vol. I:
       Stationary Point and Area Sources.  Research Triangle Park, North Carolina:
       U.S. Environmental Protection Agency, 1995.

 2.     U.S. EPA. Technical Procedures for Developing AP-42 Emission Factors and
       Preparing AP-42 Sections.  EPA-454/B-93-050.  Research Triangle Park, North
       Carolina:  U.S. Environmental Protection Agency,  Office of Air Quality  Planning and
       Standards, 1993.

 3.     Pope, A.  (U.S. Environmental Protection Agency, Emission Inventory Branch); R.B.
       Jones (Midwest Research Institute);  and G. Brooks (Radian Corporation). Group
       discussion on applying "U" rating to emission factors. Research Triangle Park, North
       Carolina.  December 8, 1992.

 4.     World Health Organization.  Sources of Environmental Pollution.  Polychlonnated
       Dibenzo-p-dioxms and Dibenzofurans, Environmental Health Criteria 88.  Geneva,
       Switzerland:  World Health Organization,  1989.

 5.     U.S. EPA.  Health and Environmental Effects Profile for Tetra-, Penta-,  and
       Hexachlorodibenzofurans.  EPA-600/X-84/114.   Cincinnati, Ohio:  U.S. Environmental
       Protection Agency, Office of Health and Environmental Assessment, 1983.  p. 1-1.

 6.     U.S. EPA.  Health Assessment Document for Poly chlorinated Dibenzo-p-dwxms.
       EPA-600/8-84-014A. Washington, D.C.:  U.S.  Environmental Protection Agency,
       Office of Health and Environmental Assessment, 1984. pp. 3-5.

 7.     Podoll, R.T., H.M. Jaber, and T. Mill. Tetrachlorodibenzodioxin:  Rates of
       Volatilization and  Photolysis in the Environment. Environmental Science and
       Technology-.  20(1986):490.
«*
 8.     Lustenhouwer, J.W.A., K.  Olie, and O. Hutzinger.  Chlorinated Dibenzo-p-dioxins and
       Related Compounds in Incinerator Effluents:  A  Review of Measurements and
       Mechanisms of Formation.  Chemosphere.  9:512,1990.
                                         6-1
-------
9.     U.S. EPA.  National Dioxin Study Tier 4—Combustion Sources: Engineering Analysis
      Report. EPA-450/4-84-014h.  Research Triangle Park, North Carolina:
      U.S. Environmental Protection Agency, 1987.

10.   U.S. EPA.  National Dioxin Study Tier 4—Combustion Sources: Project Summary
      Report. EPA-450/4-84-014g.  Research Triangle Park, North Carolina:
      U.S. Environmental Protection Agency, 1987.

11.   Barnes, D.G., F.W. Kutz (U.S. Environmental Protection Agency, Office of Research
      and Development); D.P. Bottimore (Versar, Inc.).  Interim Procedures for Estimating
      Risks Associated with Exposures to Mixtures of Chlorinated Dibenzo-p-dioxins and
      Dibenzofurans (CDD and CDF)  and 1989 Update.  EPA-625/3-89/016.  Washington,
      D.C.:  U.S. Environmental Protection Agency,  1989.

12.   National Dioxin Emission Estimates from Municipal Waste Combustors.
      U.S. Environmental Protection Agency, Emission Standards Division, Office of Air
      Quality Planning and Standards,  Research Triangle Park, North Carolina.  June 1996.
      Docket A-90-45, Item VII-A-001.

13.   National Dioxin Emission from Medical Waste Incinerators.  U.S. Environmental
      Protection Agency, Emission Standards Division, Office of Air Quality  Planning and
      Standards, Research Triangle Park, North Carolina. June 1996.

14.   U.S. EPA.  Locating and Estimating Air Emissions from Sources of Poly cyclic Organic
      Matter, Final Report.  Research Triangle Park,  North  Carolina: U.S.  Environmental
      Protection Agency, November 1996.

15.   Fenn, D.A.,  and K.L. Nebel (Radian Corporation).  Memorandum to W.H. Stevenson
      (U.S. Environmental Protection Agency, Emission Standards Division) concerning
      MWC Database.  March 1992.

16.   AP-42, 5th ed., op. cit., note 3.  Section 2.1, Refuse Combustion, 1995.

17.   U.S. EPA.  Municipal Waste Combustion Assessment:  Technical Basis for Good
      Combustion Practice.  EPA-600/8-89-063.  Research Triangle Park, North Carolina:
      U.S. Environmental Protection Agency, 1989.

18.   U.S. EPA.  Locating and Estimating Air Toxics Emissions from Sewage Sludge
      Incinerators. EPA-450/2-90-009.  Research Triangle  Park, North Carolina:
      U.S. Environmental Protection Agency, 1990.

19.   U.S. EPA.  Standards of Performance for New Stationary Sources: Municipal Waste
      Combustors, 59 FR 48216, September 20, 1994.
                                         6-2
-------
20.    U.S. EPA. Municipal Waste Combustors-Background Information for Proposed
       Standards: Post-Combustion Technology Performance.  EPA-450/3-89-27c.  Research
       Triangle Park, North Carolina: U.S. Environmental Protection Agency, 1989.

21.    U.S. EPA. Locating and Estimating Air Toxic Emissions from Medical Waste
       Incinerators.  EPA-454/R-93-053.  Research Triangle Park, North Carolina:
       U.S. Environmental Protection Agency, 1993.

22.    AP-42, 5th ed., op. cit., note 3. Section 2.3, Medical Waste Incineration, 1995.

23.    Moody, T.K. (Radian Corporation), and R. Copland (U.S. Environmental Protection
       Agency, Emission Standards Division).  Teleconference concerning hospital
       incinerators.  July 17, 1995.

24.    Radian Corporation.  Site 4 Final Emission Test Report,  Sewage Sludge Test Program.
       Research Triangle Park, North Carolina:  U.S. Environmental Protection Agency,  1989.

25.    ERT.  Non-Criteria Emissions  Monitoring Program for the  Envirotech Nine-Hearth
       Sewage Sludge Incinerator at  the Metropolitan Wastewater  Treatment Facility.
       St. Paul, Minnesota:  Metropolitan  Waste Control Commission, 1986.

26.    U.S. EPA. National Dioxin Study  Tier 4--Combustion Sources, Final Test Report--
       Site 1  Sewage Sludge Incinerator SSI-A.  EPA-450/4-84-014J.  Research Triangle Park,
       North  Carolina:  U.S. Environmental  Protection Agency, 1987.

27.    U.S. EPA. National Dioxin Study  Tier 4--Combustion Sources, Final Test Report--
       Site 3  Sewage Sludge Incinerator SSI-B.  EPA-450/4-84-0141.  Research Triangle Park,
       North  Carolina:  U.S. Environmental  Protection Agency, 1987.

28.    U.S. EPA. National Dioxin Study  Tier 4—Combustion Sources, Final Test Report-
       Site 12 Sewage Sludge Incinerator  SSI-C. EPA-450/4-84-0141.  Research Triangle
       Park, North Carolina: U.S. Environmental Protection Agency, 1987.

29     Acurex Corporation.  1990 Source  Test Data for the Sewage Sludge Incinerator.
       Prepared for the Palo Alto Water Quality Control Plant.  Palo Alto, California:
       Acurex Corporation,  1991.

30.    Entropy Environmentalists, Inc. Emissions of Metals, Chromium and Nickel Species,
       and Organics from Municipal  Wastewater Sludge Incinerators, Vol. I:  Summary
       Report.  Cincinnati, Ohio:  U.S. Environmental Protection Agency, 1992.

31.    AP-42, 5th ed., op. cit., note 3. Section 2.2, Sewage Sludge Incineration,  1995.
                                         6-3
-------
32.    Whitworth, W.E., and L.R. Waterland. Pilot-Scale Incineration of PCB-Contaminated
      Sediments from the Hot Spot of the New Bedford Harbor Superfund Site.  Jefferson,
      Arkansas: Acurex Corporation, 1992.

33.    Oppelt, E.T.  Incineration of Hazardous Waste~A  Critical Review. Journal of Air
      Pollution Control Association.  37(5):558-586, 1987.

34.    Standards for Owners and Operators of Hazardous Waste Incinerators and Burning of
      Hazardous Wastes in Boilers and Industrial Furnaces.  40 CFR Parts 260, 261, 264,
      and 270.  Federal Register.  55(82):17-93, April 27, 1990.

35.    U.S. EPA. Permit Writer's Guide to Test Burn Data, Hazardous  Waste Incineration.
      EPA-625/6-86-012.  Cincinnati, Ohio:  U.S. Environmental Protection Agency, 1986.

36.    U.S. EPA. National Dioxin Study Tier 4-Combustion Sources, Final Test Report--
      Site 2 Industrial Solid Waste Incinerator ISW-A. EPA-450/4-84-014k.  Research
      Triangle Park, North Carolina: U.S. Environmental Protection Agency, 1987.

37.    Radian Corporation.  Summary of Trace Emissions from and Recommendations of Risk
      Assessment Methodologies for Coal and Oil Combustion Sources.
      DCN No. 86-203-024-41-10.  Research Triangle Park, North Carolina:
      U.S. Environmental Protection Agency, 1986.

38.    Cole, J. (Research Triangle Institute).  Memorandum to W. Maxwell
      (U.S. Environmental Protection Agency) on HAP Emission Factors.  January 19, 1995.

39.    Someshwar, A.V. and J.E. Pinkerton.  Air Pollution Engineering Manual, Chapter 18:
      Wood Processing Industry.  Air and Waste Management Association.  New York, New
      York:  Van Nostrand Reinhold, 1992.

40.    AP-42, 5th ed., op.  cil., note 3. Section 1.6, Wood Waste Combustion Boilers,  1995.

41.    Someshwar, A.V.  Summary of Mill Test Reports  on  Wood Residue Combustion,
      Black  Liquor Combustion, and Kraft Mill Sludge Burning. NCASI Files.  Gainesville,
      Florida: National Council of the Paper Industry for Air and Stream Improvement,
      Inc., 1995.

42.    AP-42, 5th ed., op.  cit., note 3. Section 1.10, Residential Wood Stoves, 1995.

43.    Johnson, N.D., and M.T. Schultz (ORTECH Corporation). MOE Toxic Chemical
      Emissions Inventory for Ontario and Eastern North America.  Final Report
      No. P92-T61-5429/OG.  Rexdale, Ontario:  Ontario Ministry of the Environment, Air
      Resources Branch, 1992.

44.    AP-42, 5th ed,, op.  cit., note 3. Section  1.9, Residential Fireplaces, 1995.


                                         6-4
-------
45.    AP-42, 5th*ed., op. dr., note 3. Section 1.3, Fuel Oil Combustion, 1995.

46.    U.S. EPA.  National Dioxin Study Tier 4~Combustion Sources: Final Test Report--
       Site 13 Residential Wood Stove WS-A.  EPA-450/4-84-014v. Research Triangle Park,
       North Carolina:  U.S. Environmental Protection Agency, 1987.

47.    U.S. EPA.   Estimating Exposure to Dioxin-Like Compounds, Volume II:  Properties,
       Sources, Occurrence, and Background Exposures, External Review Draft.
       EPA-600/6-88-005bC.  Washington, D.C.:  U.S. Environmental Protection Agency,
       Exposure Assessment Group, Office of Health and Environmental Assessment, June
       1994.

48.    U.S. EPA.  Markets for Scrap  Tires.  EPA/530-SW-90-074A. Washington, D.C.:
       U.S. Environmental Protection Agency, Office of Solid Waste and Emergency
       Response,  1991.

49.    Scrap Tire Management Council.  Scrap Tire Use/Disposal Study, 1992 Update.  1992.

50.    Radian Corporation. Modesto  Energy Company  Waste Tire-to-Energy Facility,
       Westley, California, Final Emission Test Report. Report No. 243-047-20. Research
       Triangle Park, North Carolina:   Radian  Corporation, 1988.

51.    California Air Resources Board. Emissions Testing of a Propane Fired Incinerator  at
       a Crematorium.  Report No. ERC-39. Sacramento, California: California Air
       Resources Board, 1992.

.52     Cremation Association of North America.  Cremation Statistics.  Crematiomsi.
       Chicago, Illinois:  Cremation Association of North America, 1992.

53     U.S.  EPA.  Emission Factors for Iron Foundries—Criterion  and Toxic Pollutants.
       EPA-60,2-90-044.   Cincinnati,  Ohio:  U.S. Environmental Protection Agency, Control
       Technology Center, Office of Research  and Development, 1990.

54     U.S.  EPA.  Electric Arc Furnaces in  Ferrous Foundries, Background Information for
       Proposed Standards. Report No. 3-80-020a. Research Triangle Park, North Carolina:
       U.S.  Environmental Protection  Agency, Office of Air Quality Planning and Standards.
       1980.

55.    California Air Resources Board. Emissions Measurement of Toxic Compounds from a
       Cupola Baghouse at a Steel Foundry.  Report No. ERC-61.  Sacramento, California:
       California Air Resources Board, 1993.

56.    Keller, P.A. (Radian Corporation).  Teleconference with J. Maysilles,
       (U.S. Environmental Protection Agency).  April 2, 1994.
                                         6-5
-------
57.    AP-42, 5th ed., op. cit., note 3.  Section 12.9, Secondary Copper Smelting and
       Alloying, 1995.

58.    U.S. EPA. National Dioxin Study Tier 4—Combustion Sources: Final Test Report--
       Site 10 Secondary Copper Recovery Cupola Furnace MET-A. EPA-450/4-84-0145.
       Research Triangle Park, North Carolina: U.S. Environmental Protection Agency, 1987.

59.    AP-42, 5th ed., op. cit., note 3.  Section 12.8, Secondary Aluminum Operations, 1995.

60.    California Air Resources Board. Dioxin/Furan Emissions Estimates from a Secondary
       Aluminum Facility.  Report No. ERC-9. Sacramento, California: California Air
       Resources Board, 1992.

61.    California Air Resources Board, Emissions Measurements on Delacquering System.
       Report No. ERC-32.  Sacramento, California:  California Air Resources Board,  1992.

62.    U.S. EPA. Secondary Lead Smelting Background Information Document For
       Proposed Standards, Vol. 1.  EPA-450/R-94-024a. Research Triangle Park, North
       Carolina: U.S. Environmental Protection Agency, Office of Air Quality Planning and
       Standards, 1994.

63.    Steinwurtzel, R.N. (Counsel for Association of Battery Recyclers), and D.W. Safreit
       (U.S.  Environmental Protection Agency, Emission Factor and Inventory Group).
       Personal communication.  May  4, 1995.

64.    U.S. EPA.  HAP Emission Testing on Selected Sources a: a Secondary Lead Smelter—
       Tejas Resources, Inc.. Draft  Emission Test Report.  Prepared by Roy F. Weston. Inc.
       EPA Contract No. 68-D1-0104.  Research Triangle Park: U.S. Environmental
       Protection Agency, 1992.

65.    U.S. EPA.  HAP Emission Testing on Selected Sources at a Secondary Lead
       Smelter—Schuylkill Metals Corporation, Draft Emission Test Report. Prepared by Roy
       F. Weston, Inc. EPA Contract No. 68-D1-0104.  Research Triangle Park:
       U.S. Environmental Protection Agency, 1992.

66.    U.S. EPA.  HAP Emission Testing on Selected Sources at a Secondary Lead
       Smeller—East Penn Manufacturing Company, Draft Emission Test Report. Prepared
       by Roy F. Weston, Inc. EPA Contract No.  68-D1-0104.  Research  Triangle Park:
       U.S. Environmental Protection Agency, 1992.

67.    U.S. EPA.  Secondary Lead Smelting Background Information Document For
       Proposed Standards, Vol. 2:  Appendices.  EPA-450/R-94-024b.  Research Triangle
       Park,  North Carolina:  U.S. Environmental Protection Agency, Office of Air Quality
       Planning and Standards, 1994.
                                         6-6
-------
68.    U.S. EPA. National Dioxin Study Tier 4—Combustion Sources:  Final Tesi Repon--
       Site 6 Wire Reclamation Incinerator WR1-A.  EPA-450/4-84-0140. Research Triangle
       Park, North Carolina: U.S. Environmental Protection Agency, 1987.

69.    U.S. EPA. National Dioxin Study Tier 4--Combustion Sources:  Final Test Report-
       Site 11 Drum and Barrel Reclamation Furnace DBR-A.  EPA-450/4-8-0144. Research
       Triangle Park, North Carolina: U.S. Environmental Protection Agency, 1987.

70.    AP-42, 5th ed., op. dr., note 3.  Section 10.2, Chemical  Wood Pulping, 1995.

71.    Dyer, H., S. Gajita, and D.  Fisher.  1992 Lockwood-Post's Directory of the Pulp,
       Paper and Allied Trades. Miller Freeman Publications.  San Francisco, California,
       1991.

72.    Dyer, H., S. Gajita, and D.  Fisher.  1997 Lockwood-Post's Directory of the Pulp,
       Paper and Allied Trades. Miller Freeman Publications.  San Francisco, California,
       1997.

73.    National  Council of the Paper Industry for Air and Stream Improvement, Inc. A Study
       of Kraft Recovery Furnace Hydrochloric Acid Emissions.  Technical Bulletin No. 674.
       Gainesville, Florida:  National Council of the Paper Industry for Air and Stream
       Improvement, Inc., August,  1994.

74.    Carruth,  G., ed.  The Volume Library, Energy and Power Sources. Nashville,
       Tennessee: The Southwestern Company, 1986.

75.    Ballschmiter, K., et al.  Automobile Exhausts Versus Municipal Waste Incineration as
       Sources of the Polychloro-dibenzodioxins (PCDD) and -furans (PCDF) Found in the
       Environment.  Chemosphere.  15(7):901-915, 1986.

76.    Marklund, S., C. Rappe,  M. Tysklind, and K. Egeback.  Identification of
       Polychlorinated Dibenzofurans and Dioxins in Exhausts from Cars Run on Leaded
       Gasoline. Chemosphere.  16(l):29-36, 1987.

77.    Marklund, S., et al.  Emissions of PCDDs and PCDFs  in Gasoline and Diesel Fueled
       Cars.   Chemosphere.  20(5):553-561, 1990.

78.    Oehme, M., S. Larson, and  E.M.  Bvevik.  Emission Factors of PCDD and PCDF for
       Road Vehicles Obtained by  Tunnel Experiment.   Chemosphere.   23(11/12)1699-1708,
       1991.
                                         6-7
-------
79.    U.S. Environmetnal Protection Agency. Emissions Inventory of Section 112(c)(6)
       Pollutants:  Polycyclic Organic Matter (POM), 2,3,7,8-Tetrachlorodibenzo-P-Dioxin
       (TCDD)/2,3,7,8,-Tetrachlorodibenzofuran (TCDF), Poly chlorinated Biphenyl
       Compounds (PCBs), Hexachlorobenzene, Mercury, and Alkylated Lead.  Draft Report.
       Office of Air Quality Planning and Standards, Research Triangle Park, North
       Carolina, 1996.

80.    Humer, C.   Activated  Carbon Plant Starts for American Novit.  Chemical Marketing
       Reporter.  June 8,  1992.  p. 2.

81.    U.S. EPA.   National Dioxin Study,  Tier 4--Combustion Sources, Final Test Report--
       Site 9, Carbon Regeneration Furnace CRF-A.  EPA-450/4-84-014r.  Research Triangle
       Park, North Carolina:  U.S. Environmental Protection Agency,  Office of Air Quality
       Planning and Standards, 1987.

82.    Schuliger, W.G. and L.G. Knapil.  Reactivation Systems.  Paper presented at the
       annual conference of the American Water Works Association, June 1990.

83.    Byers, W.D. Charcoal/Activated Carbon.  Air Pollution Engineering Manual.
       A.J. Buonicore  and W.J. Davis, eds. New York, New York: Van Nostrand Reinhold,
       1991.

84.    Koffskey, W.E. and B.W. Lykins,  Jr.  GAC  Adsorption and Infrared Reactivation:
       Case Study. Journal of American  Water Works Association.  (January 1990) 48-56.

85.    BTC Environmental, Inc. AB-2588 Toxic Emission Testing.  Ventura, CA: Orange
       County Water District, 1991.

86.    Raven, P.H.  Biology of Plants. New York, New York: Worth Publishers, Inc., 1980.

87.    Bacher, R.,  M.  Swerev, and K. Ballschmiter. Profile and Pattern of Monochloro-
       through Octachlorodibenzodioxins  and -dibenzofurans in Chimney  Deposits from
       Wood Burning.  Environmental Science and Technology.  26(8):1649-1655, 1992.

88.    Nestrick, T.J., and  L.L. Lamparski.  Assessment of Chlorinated Dibenzo-p-dioxin
       Formation and Potential Emission to the Environment from Wood  Combustion.
       Chemosphere.  12(4/5):617-626, 1983.

89.    Tashiro, C.  et al.  Preliminary Report:  Dioxins and Furans in Prescribed Burns.
       Chemosphere.  20:1533-1536, 1990.

90.    U.S. Code of Federal Regulations.   Title 40, Protection of the Environment, Part 761-
       Polychorinated Bi-phenyls (PCBs)  Manufacturing, Processing, Distribution in
       Commerce,  and Use Prohibitions.  Washington, D.C.:  U.S.  Government Printing
       Office,  1994.
                                         6-8
-------
91.    U.S. EPA.  National Dioxin Study Tier 4—Combustion Sources:  Final Literature
       Review.  EPA-450/4-84-0141.  Research Triangle Park, North Carolina:
       U.S. Environmental Protection Agency, 1986.

92.    AP-42, 5th  ed., op. cit., note 3.  Section 2.4, Landfills, 1995.

93.    California Air Resources Board.  Compliance Testing for Non-Criteria Pollutants at a
       Landfill Flare. Report No. ERC-2. Sacramento, California:  California Air Resources
       Board, 1990.

94.    Lee, A. et al.  Dioxins and Furan Contamination in the Manufacture of Halogenated
       Organic Chemicals.  EPA/600/2-86/101.  Cincinnati, Ohio: U.S. Environmental
       Protection Agency, 1986.

95.    Sittig, M. Handbook of Toxic and Hazardous Chemicals and Carcinogens, 2nd ed.
       Park Ridge, New Jersey: Noyer Publications, 1985.

96.    Wilkinson, J. (Vulcan Chemicals).  Letter to D.  Safriet (U.S. Environmental Protection
       Agency, Emission Factor and Inventory Group)  concerning dioxin exposure assessment
       comments, technical information staff.  June 20, 1995.

97.    Chinkin, L.  et al.  Inventory of Chlorophenol Use in  the Forest Products Industry* and
       Investigation of Related Emissions  of Chlorinated Dibenzodioxins and Dibenzofurans.
       SYSAPP-87/078.  Sacramento, California: California Air Resources Board, 1987.

98     AP-42. 5th ed., op. cit.,  note 3.  Section 11.6, Portland Cement Manufacturing, 1995.

99     Pierson. T. (Research Triangle Institute).  Summary of Portland Cement M ACT Data.
       Report to T. Lahre, (U.S. Environmental Protection Agency, Office of Air Quality
       Planning and Standards). Research Triangle Park, North  Carolina:  Research Triangle
       Institute. April 25. 1994

100    Kim, I  Incinerators and Cement Kilns Face Off.  Chemical Engineering.  April 1994.

101    U.S. EPA.  Technical Support for Revision of the Hazardous Waste Combustion
       Regulations  for Cement Kilns and Other Thermal Treatment Devices, Second Draft.
       Prepared by Energy and Environmental Research Corporation,  Irvine, California.
       Washington, D.C.:  U.S. Environmental Protection Agency, Office of Solid Waste.
       May 17, 1994.

102.    U.S. EPA.  IERL-RTP Procedure Manual:  Level 1 Environmental Assessment
       (2nd ed.). EPA-600/7-78-201.  Interagency Energy/Environment R&D Program
       Report. U.S. Environmental Protection Agency, October 1978.
                                         6-9
-------
103.   Cooke, M., F. DeRoos, and B. Rising. Hot Flue Gas Spiking and Recovery- Study for
       Tetrachlorodibenzodioxins (TCDD) Using Modified Method 5 and SASS (Source
       Assessment Sampling System) Sampling with a Simulated Incinerator.
       EPA-600/2-84/159. Research Triangle Park, North Carolina: U.S. Environmental
       Protection Agency, Industrial Environmental Research Laboratory, 1984.

104.   Standards of Performance for New Stationary Sources, Appendix A, Test Method 23  -
       Determination of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated
       Dibenzofurans from Stationary Sources.  Federal Register, 60(104)28378-28426,
       May 31, 1995.

105.   California Air Resources Board.  Stationary Source Test Methods, Volume III:
       Methods for Determining Emissions of Toxic Air Contaminants from Stationary
       Sources. Sacramento, California:  California Air Resources Board, Monitoring and
       Laboratory Division, 1989.

106.   American Society of Mechanical Engineers.  Analytical Procedures to Assay Stack
       Effluent Samples and Residual Combustion Products for Polychlorinated
       Dibenzo-p-Dioxins (PCDD) and Polychlorinated Dibenzofurans (PCDF).  Washington,
       D.C.:  U.S. Department of Energy and U.S. Environmental Protection Agency, 1984.

107.   U.S. EPA.  Test Methods for Evaluating Solid Waste, 3rd ed. Method 8280 - The
       Analysis of Polychlorinated Dibenzo-p-Dioxin and Polychlorinated Dibenzofurans.
       SW-846.  Washington, D.C.:  Office of Solid Waste,  1986.

108.   U.S. EPA.  Test Methods for Evaluating Solid Waste, 3rd ed. Method 8290 - The
       Analysis of Polychlorinated Dibenzo-p-Dioxin and Polychlorinated Dibenzofurans by
       High-Resolution Gas Chromatography/High-Resolution Mass  Spectrometry.  SW-846.
       Washington. D.C.:  Office of Solid Waste, 1986.
                                        6-10
-------
                   APPENDIX A




METHODS FOR ESTIMATING NATIONAL CDD/CDF EMISSIONS
-------
Note:  The national emissions estimates presented here are those that were available at the
       time this document was published.  Ongoing efforts and studies by the
       U.S. Environmental Protection Agency will most likely generate new estimates.  The
       reader should contact the U.S. Environmental Protection Agency for the most recent
       estimates.
                                          n
-------
                               TABLE OF CONTENTS






Municipal Waste Combustion	A-l




Medical Waste Incineration  	A-14




.Sewage Sludge Incineration	A-20




Hazardous Waste Incineration	A-21




Lightweight Aggregate Kilns (LWAKS)  	A-23




Portland Cement	 A-24




Waste Tire Incineration	 A-26




Utility Coal Combustion  	A-27




Utility Residual  Oil Combustion	A-28




Industrial Wood Combustion 	 A-29




Residential  Coal Combustion 	 A-31




Residential  Distillate Fuel Oil Combustion	 A-34




Residential  Wood Combustion  	 A-36




Iron and Steel Foundries	A-38




Secondary Copper Smelters  	 A-40




Secondary Lead Smelters	A-41




Secondary Aluminum Smelters	A-43




Drum and Barrel Reclamation/Incineration	 A-45




On-road Mobile Sources  	 A-46




Pulp and Paper—Kraft Recovery Furnaces  	A-48




Wood Treatment  	^	 A-49




Carbon Regeneration/Reactivation	A-50







                                         iii
-------
                       TABLE OF CONTENTS (Continued)




Forest Fires	 A-52




Crematories	 A-53




Remaining Source Categories	 A-54




References	A-55
                                      w
-------
 MUNICIPAL WASTE COMBUSTION

 Basis for Calculation

TThe national dioxin/furan emissions estimates for MWCs were obtained directly from work
 done by EPA's Emission Standards Division (BSD) to support MACT standards for this
 source category.'   The estimates are based on dioxin data collected by the EPA during its
 "MWC Survey" in 1994.  The dioxin emissions data that were collected were combined with
 MWC plant design data,  plant annual utilization rate data, and flue gas flow rate conversion
 factors to calculate annual dioxin emission estimates. ESD calculated estimates for three
 different time periods (1993, 1995, and 2000), and the 1995 estimate is presented in  this
 document because it is the most recent estimate.  A summary  of the  methods used to
 determine the estimates are provided here. More detailed information on the derivation  of the
 estimates can be obtained from the cited reference  I.1

 There are three main types of municipal waste incinerators in  the United States:  mass bum
 (MB), refuse derived fuel (RDF), and modular combustors (MOD).  Mass bum combustors
.are the mosi common type of combustor, representing 54  percent of  all municipal waste
 combustors (MWCs) in the United States, followed by modular facilities (32 percent) and
 RDF facilities (13 percent).'

 According to the  1994 Maximum Achievable Control Technologies (MACT) Study there are
 158 existing MWC facilities with design capacities above  38.6 tons/day.  The facilities
 designed to burn  less than 38.6 tons/day account for less than  one percent of the total waste
 flow to MWC facilities.2   Of the total MWC capacity in the United States, about 58  percent
 of municipal  waste is treated in mass bum facilities, 29 percent in RDF-fired facilities,
 9 percent in modular combustors, and 4 percent in  other MWC designs.2

 Dioxin test results for 1993 were compiled from emission source tests performed between
 1985 and 1993 (see Table A-l).  In a limited number of cases, test results from 1994 on
 individual units were used if there were no changes in MWC unit operation or air pollution
                                          A-l
-------

•*•
 v>
u:

a
 OJ

f-

1-4
O
C bP
CO ^
E r-

if
0 8
(5 5
£
u
o
3
i
o
U





Air
Pollution






2 c
r- £1















b.
£




a
H

H

'o sS"
3-1
u

1 6
_g •§
E =>
o Z
U
Control
Devices
^
o
3 8.
E H
o
U
t4_ ^
o S
Number
Combus
it
re "— •

a
oo





u
re
Z
'E
c
00
u


O^




o
oo
O
O

"

m
•*
ro



fS



Q
u-

^
ca
^

(N
1


<

ro
U.
E

CJ
00
•o
k.
Huntsville Refuse F

rr


oo
ON
ON




Z


o

m



^^



Q a:
uu Z
U. 00

z
CD
*

— '
OC

<
U

u
re
b—

an
c
IX!
c
Commerce Refuse-t

ON


00
S




Z


M-

1



	



Q a
oo u
(b Z
u. oo
.^
^
^
2

m
g
^
,,
^






^_
Ct
&
u
nj
GJ
03
ex
c
C
i— J

ON


fn
t>




z


m

8
•^B



__.



Q
oo
2:
u.

z
CD
;>

CM
OC

•f







	
Stanislaus (Modesto

ON


**>
g




Z


*

8



JS



K:
j
n























*




VO
—

t-

ro



	 ,



Q
u.
v
S^
CO
^

(N
c
vC

f-
U







U.
t£
~o
tr
CO

_



o!




§
o

oo

r)



^


























on
00




Z




r~



__.



a
00
u.

s
c^


m
X
c
fN
l_
0





-
•^
Mid-Connecticut Pr

*^



S!




z


oo
t~-

|



__



Q
00
tt
u.
^
UJ
C
O
S

m
S

f-








Wallingford RRF

—*



g




z


g

o



«s



























s;




z


i—

0



m



























&




Z


1

O



__.



a.
00
u

U.
Q
(X


"*
|
r-j
,
U.







U.
o
u
u
re
Q

T)-



§;




z


00

1



	 ,



a.
00

£
m
^

m
8
"
,
u.





u.

Hillsborough Co. R

^•



§; si §;


^^

^ e 8
-------
        •O
        tv
        S
        e
        o

        U

        V!

        U
       I
        tf)
        O)
_      o
 ^     *••

"w     C
u.     _
        ffS

.0
"r?
£ ©
O t/3
O 5
e

2
WJ
S
.O

o
U





Air
Pollution





r^ Z-












n
u


UJ
H
3
o
H
^
If
« ^
U

c ,_
= I
F 1
o Z
U
Control
Devices
c
o
U
., i/;
o 5
•£ .c
u —
M
!j
ffi




CJ
re
Z
'E
c
o
OJj
u
Bi
SN


Z

s
m

8








c.
u.
w
en
^


ts
1

=





k.
c
u.
c
1
.£-
ra

O


§


Z

NO


8








CL,
CO
m
^


"*
I

—





Northwest WTE


m

ON


§
0
ts


00








a
oo
u.
B/WW
2


m
m

Z





Indianapolis RRF


in

S


O
o
o
-


t—
00
r-



























§


Z

VO


O








ESP/FSI
u.
O
tt



"
c

2





: Haverhill Lawrence RDI


"~*

1


Z

-------
<
        1
         B
         O
        U
        u
        I
         t/5
         0>
        •*—
         CQ
         E
         X
         o
         C3
         C
        a
         08
         •*«
         C3

-~
K°
2 t—
3
" C
o 8
SM
s



o
a
1
o
u





Air
Pollution






- =
r-C-











U.
u


2>
H
S

H


'o S-
s.1
u

o i_
•c E
E 3
o Z
U
Control
Devices
w
s«-
E H
o
U
.. v;
C ^
S *

3 ~
?i

8J
£


i!
re
E

c
o
OO
Bt
m



Z

-

•"*


8



_



CH

U.
Q
«

m

c
fi
,__,
±



u.
a:
oi
'c
a
u
LI
O


in


Si


Z


£
CQ

(N

vn
NO
,_ ^
2



o
£
UJ
s
u
u


in


£


O
S
0
-




1



„



Q
1

£
CQ

(N

R
(N
7,

-^
rt
L.
U
§
a:
OJ
c
c
t


in

«s
£


o
0
c.




1



«s



J
2
J
yi



















^
£


Z

ON




8



„



&.
CO
u
.^
^
n

(N

e
fN
z
s



>,
u
£
Olmstead WTE


m


S


Z


1
i
n
C£


in


&


§
—
ON




NO



_



CU
to
u
<
Q
O

(S

fS

Z
2
u
re
U.
u.
'5
CQ
u
to
re

c?
"8
o:


m


£


Z

VO




o



(M



to

U.
Q



Z



u.
Oi
o
u
'E


•*

^
00
ON


Z

oo




8



_



FF/DSI
>
^
cc
2

fS

R

M.
Z



u.
Oi
Oi
SES Claremont


—

^
00
ON


Z

«M




8



rg
























ON
ON


Z

m




0
m



_



ESP/SD
.^
^
cc
2

rr>

o
0

z




Camden RRF


fS

00
oo
ON


i
o
oo




8
ts



_



Q
U.
>
^
CQ
2

(N

O
0


z



u.
DO
i;
c
UJ
c
t
ra


-------
       •O
        a>
        s
       JS
       '•^t
        e
        o
       U

••—      o
i     5

.0
3
U. ©
O M
5 3
S


o
s

E
0
U




c
_G
< =
£



- -
•OH ^_














S

&
H

3
t**

x
•5 5-

re "^
rj

C i_
II
F i
c Z
O &J
— .u
U Q
Combustor
Type
ll
Z £
I ?
« ._-
~
ra





U
ra
Z
E

c
o
ui
g

Z


—•



£








Q
to
u.
MB/WW
(N

C
r-

Z







L.
e<
a
c
o
!>
ro
cc


(S
g

Z


-•



r—








Q
to
al
u.
MB/WW
M

in
f^

Z







U.
cc
c
_o
m
CO


r-i
00

Z


^



R








to
Q
U.
MB/RC
(VI

c
TT

z







u.
a
a
O
I/:
LI
f
U
C


fS
i

z


00



8
«s




























M

o
o
o

*—



J^
oo







G
to
E.
u.
MB/WW
m

r-'

(N
Z







•o
re
u
v;
C.
1^


(S
ON 9> C^i
ON ON ON

§ R |
O O O

r-j r- r>



O O O
v> w> m








to u
£ z
tt. to
ca
f.

c
r-

Z







u.
a
C£
c
c


ts
s

I
"•

S



8








0.
IS

MB/WW
— 1

R


z







u.
a:
Cd
re
00
c


^
i

z


"r



g;








Lo
Q
U.
MB/RC

^
t«
00
w
C
tu
TJ
u
«j
U
c
£



i

z


i
,
u


in
^^ c
00 C
^^ c

<; <


11
*A i
*f} V

s ?








a.
to
U-

MB/RC/RFF
fn

£
^

~




^
Oi
f
c
c
>*
k.
u
3*
c


in
^ rr
D O\
^ Ov

C §
- 00
-•

^ ^^
T OO
1

§ 8





























                                                   A-5
-------
 en
H
        TJ
        V
        3
        C
        o
        U
        WJ
       u
       ^

        -«— i
        (A.
        t-
        c
        fi
        ON

        ^
        v:
        .
F H
0
U
. 1/5
C C
U
1 «
|i

^
0
ffi



V
E
2
'£





c
o
oc
u
as
1


Z


•a
©
£
in

8



_*



a.
w
u.
u
U
C£
EQ
s


o

R
ON

•Y*
C


a.
a
.n
3
O
C/2
U
E
c
oc
c
c

;



rvi

0
f^.

Cw






H
3
X>


«3
X

to


f*% pn t*}
ON ON ON
^^ ON ON


O O Q
OO t- O
f^l ^^ f*^
r) — —

T — ON
— NO •V
~"


888



— «S t*>



Q
to
u.
^
^
CD
2>



n

8

~
O.






U.
a:
OS
a
53
fc?
en

ra

m


g g g


z z z


m oo oo




OO OO OO



- C4 m



O
to
u.
1
D
cc
CO
2


m

^
r-i
~
C-






o
O
Ci
-
f )


C

m


s
ON


Z


^.
^-



8



ts



as/asa
^
^
CQ
^



rs

S5
NO

U


OS
c
to
^i
U
b-,
OJ
Si
>

O

rp

M-


1


z


p")
-^-
~~


o



m



Cu
U
^
^
m
^



^

c
o

Z
f-
c.
o
U

C
CQ
Thermal T
u
^J
>



^


                                                      A-6
-------
       ts
        u
        s
        B
        O
       I
        cs


_     -^
^     _=

—      o
 «     5
       _c
       •«M
        ^5
       Z



       ON

        U


       •O


       ~^

        ec


       C

.0
3
^ E

.S jg
c

o
3
1
5





Air
Pollution





- -s
















8

fc

^
1
H
^
'o ^
£.1;
U
w
O i_
•c E
E §
o Z
U
Control
Devices
£
1 5-
u
^ £
c c
II
11


o
re
(75


u
E
re
Z

_c

c
o
OO
U
a:
i

<


u
•Tt


s


	



isa/asa
^
£
ca
S

<~>
r-'


$


L.
Ct


.1°
Alexandria/Ar


fi


i

Z


u


in
CO


CO



~






















i

8




X
•2
re
U.
U.
o:
OJj
k.
u
c
U
in


m


i

R

o
NC


o
>n


cs


























1

8

o
oo


g


n


























i

s

o
-


P


^.


























F^ r^ ^T f^
^ O\ ON Ox
ON ON ON CTN

ON *O ^5 ^?
r* O\ *n O

o\ r- *n o
fsi r**) m ^
-^ 
m


5



u
n
U.
U
H
6
U
X
E
u


m



o
tu
i_
<£
1
Si
K

s
ra
•D
C
.2
w
nncentr
u
Cx
UJ
u
.c



i
x>
jq
re
Si
1
f
1
Si
u
•^
?
?
E
U
tf
-a
DJ)
C
•u
re


i/-
re
E
re
e

E


5
ai.
c
c
^
.t:
c

.E
£
c
"§
w
—
£
c


E
Dioxin concen
re
£
T3
O
u

C
c
u
o

V?
3

£
re

5
re
C
C
CJ

re
re

CN
LU


u
o
^
~_

re


—
£
re
re
re

u
^
•^
re
re
"3
O
UJ
3J
0
JC
(/-,
C
c

re
u
5 a
£ H
.£ o
(/; M
If
a> -o




































fe-
re
t»

C
c
CO

5


CJ
c

*^
^
—
0
re"
re
C-
Qj
w
1
*"
O
C£
re
i—
>
re

Pinellas data ai

1
c
Cv
u
V.
Z

^
^^

I
Si
C~
c
c
^
<
.V


^
re
~
CJ
t/
o
C.
CJ

OJ
"1
z
ftj
p
s
w
re
•o
re
o
"~
o^
E
j^

b
1
T3
C
C













^
4J

^'
C

c
c
c
c
"2
re
X
1
E
i
o
•o
o
re
4>

re
u.
>
re
c
...


re
•£
Jr
z

re


,
Z
"E
^^
^_
^c

H.
u
&
re
~
—
3
C
"£

Z
Montegomery
^2 u "O

•3
P
3
i/-
(/•
re
v
n

t^
in
r~
u-
c
c
_o
o
3
•o
Si
T5
c
_c
^
•o
re
c

„
c
o
s
c

re
u
1
'E
3
4J
.C
1
ir.
c
.c
u
^
d
C


•^
re
3


g

C

^
—
re
C
C
.^"

£
S
£•
3
(/•
U
^~

'"
Alexandria dat
u

J


























































«J
X:
re

>
o
c
a
Q
n
z
                                                  A-7
-------
control device (APCD) configurations since 1993.  Where the emission test date in the table is
noted as 1985 through 1989, the data were gathered by OAQPS to develop the new source
performance standards and emissions guidelines proposed for MWCs in 1989 (December
1989). Where the emission test date is noted as 1990 through 1994, the data were gathered
by OAQPS, or submitted to OAQPS, as part of one of the following efforts:
  1.    Data gathered by OAQPS to develop the revised new source performance standards
       and emission guidelines for MWCs proposed on September 20, 1994;
  2.    Data submitted to OAQPS in response to the "MWC Survey" of dioxin emissions; or
  3.    Data submitted to the docket (A-90-45) as public comments on the revised new source
       performance standards and emission guidelines proposed on September 20, 1994.
For facilities with dioxin test data, the following criteria were used to determine the most
representative test results for each facility for 1993:
       For facilities with more than one test result, the most recent test was used.  These
       results were chosen such that they were representative of plant operation at the end of
       1993.  Exceptions to this were in cases when two or more tests were conducted at a
       MWC over a relatively short period of time. Then, the average of these tests was used
       to represent the  1993 emissions.
       At facilities with multiple units where not all units were tested, an average  of the test
       results from the  tested  units was used as a representative value for the untested units.
Test results were obtained for approximately 55 percent of the domestic MWCs.  For the
other 45 percent, a set of default values was created and used to estimate dioxin
concentrations.  The set of default values was compiled from test data, AP-42 emission
factors, and from the EPA document EPA-450/3-89-27e "Municipal Waste Combustors -
Background Information for Proposed Guidelines for Existing Facilities"3 (this document
presented the results of a study on APCD retrofits on MWC units). Default values were
generated for every MWC combustor type and APCD configuration for which a default  value
was needed.   Test data available to OAQPS, as of January 15, 1995, were organized into
groups based  on MWC combustor type and APCD configurations and averaged. These

                                         A-8
-------
 averages were designated as the default values for the corresponding combustor/APCD
 configurations.  For the combustor/APCD configurations where there were no test data
•available, default values were obtained from the MWC section in AP-424, or from EPA
 document EPA-450/3-89-27e. A summary of the default values is given in Table A-2.

 To estimate annual emissions, a capacity factor for each unit is needed.  This factor represents
 the percentage of operational time a plant has operated during  one year.  By using the
 capacity factor and unit capacity, the annual throughput (combustion) of MSW or RDF can be
 calculated.  Some  facilities provided data to OAQPS on the tonnage of municipal waste
 burned in year 1993. For these facilities,  a capacity factor was estimated by dividing the
 tonnage burned in one year by the unit's rated yearly capacity  and was used for the  1993
 emissions calculation.  For most units, however, the capacity factors used were default values
 taken from EPA document EPA-450/3-89-27e.  For all units except  modular/starved-air
 combustors, the default capacity factor was 91 percent (0.91).  For modular/starved-air units,
 the default capacity factor was 74 percent  (0.74).

 For many test results and  for all dioxin default concentrations,  TEQ  concentrations were not
 available. Similar to the development of the  default dioxin concentrations, a default total
 mass-to-TEQ ratio was also developed.  Test results from units for which there were both
 total mass and TEQ results available were used to develop a default  ratio of total dioxin
 concentration to  TEQ concentration. The  total:TEQ ratios from these units were averaged,
 resulting in a default ratio of 50:1.  This ratio was used for estimating TEQ emissions for all
 sources where TEQ test data  were not available.

 The F
-------
                                          Table A-2
                         Default Dioxin/Furan Emission Levels
                              From MWC Configurations3
Combustor Type
MB/WW
RDF (all except FB)
RDF/FB
MB/RC/WW
MB/REF
MOD/SA
MOD;EA
APCDType
ESP*
DSI/ESP
DSI/FF
SD/ESP
SD/FF
ESP*
DSI/FF
SD/ESP
SD/FF
DSI/EGB
ESP*
DSI/ESP
DSI/FF
SD/ESP
SD/FF
ESP*
DSI/ESP
DSI/FF
SD/ESP
SD/FF
Uncontrolled
ESP*
DSI/ESP
DSI/FF
SD/ESP
SD/FF
Uncontrolled
ESP*
DSL/ESP
DSI/FF
SD/FF
Average Dioxm Value
(ng/dscm @ 7% 62- total mass)
222
60
35
70
16
240
17
9
8
63
400
100
7
40
5
500
57
17
40
5
300
288
98
8
40
5
200
468
50
8
92
a Values presented in this table are averages of available data for various combustor type/APCD type
  combinations Values were estimated based on a compilation of the MWC survey data, background information
  for the 1991 and 1994 MWC rulemakings, public comments received on the 1994 MWC rulemaking. and
  AP-42 (5th edition).  None of the data listed provides credit for supplemental dioxm control (polishing) by
  carbon injection  The use of carbon injection typically reduces dioxm emissions by an additional 75 percent or
  more  (See Air Docket A-90-45, items VI-B-013  and VI-B-014)
* ESP operated at less than 440°F.

APCD  =  air pollution control device
DSI    =  duct sorbent injection
EA    =  excess air
EGB   =  electrified gravel bed
ESP    =  electrostatic precipitator
FB     =  fluidized bed
FF     =  fabric filter
MB    =  mass burn
MOD  =  modular
RC    =  rotary combustor
RDF   =  refuse derived fuel
REF   =  refractory wall
SA    =  starved air
SD    =  spray dryer
WW   =  waterwall
                                              A-10
-------
40 CFR 60, Appendix A, Method 19.  The specific F^ factor for municipal waste combustion,
given by Method 19, is 9,570 dry standard cubic feet of flue gas per million Btu
(dscf/MMBtu) of municipal waste combustion.  This flow rate is based on 0 percent C>2 in the
flue  gases.

Average heating values for fuels derived from municipal wastes are given in the Refuse
Combustion section of AP-42.  For unprocessed municipal solid waste (MSW), the heating
value is 4,500 Btu per pound.  For RDF, the heating value is 5,500 Btu per pound.  The
heating value for RDF is higher than general MSW because RDF goes through some degree
of pre-processing to remove non-combustible materials.

The  first step in calculating annual dioxin emissions from MWCs was to calculate the
emissions  from individual units. This task was accomplished with plant-specific information
such as dioxin emission concentration,  unit size, unit capacity factor, fuel heating value, and
the F(j factor.

The  following equation was used to convert dioxin stack concentrations (total and TEQ) to
grams per  year (g-Vr) emitted:

                                      el    C x V x T x CF
                         Emissions
                                     yr
where'
       Emissions    =      Annual dioxin emissions (g'yr)
             C     =      Flue Gas Dioxin Concentration (ng/dscm @7 percent 02)
             V     =      Volumetric Flow Factor (dscm @7 percent Cb/ton waste fired)
             T     =      Tons of MSW burned/year (@100 percent capacity for 365
                           days/year)
             CF    =      Capacity Factor (unitless)
                                         A-ll
-------
The volumetric flow factor (V) is calculated as follows:
                      V  -
                         ~
                              x ^ x 200° x 10"6     20'9
                  35.31
                                                     20.9-7
where:
      HV = Heating Value (4,500 — for MSW and 5,500 — for RDF)
                                Ib                      Ib
       V  =
              9,570
  dscf @ 0% 02

     MMBtu
4,500
                  2,000 JL
            Ib           ton
                                                                    20.9
                       35.31
            dscf

            dscm
      106
                                            Btu
          MMBtu j
                                  20.9 - 7
         = 3,668
dscm @ 7%O2

  ton MSW
for non-RDF units
      V =
dscf @ 0% CL
Q S7fl ^ 2
MMBtu

7
-------
       Example Calculation

 For a mass bum/waterwall (non-RDF) unit, rated at 500 tons/day capacity (365,000 tons/yr),
 with a dioxin concentration of 10 ng/dscm (total mass) @7 percent C>2, annual dioxin
 emissions are calculated to be:
           -12-21 @ 7% 02    3,670 dscm@7%02  - L^^  tonl (o9])
            dscm            II              ton       II          yr I
       =  6.1 g dioxin/yr (total mass), or
       =  6.1/50 = 0.12 g dioxin/yr (TEQ)

Individual emission estimates were developed for all operational MWCs in the U.S., and the
individual estimates were summed to  provide the 1993 national CDD/CDF emissions estimate
for MWCs.

The 1995  CDD/CDF estimate is based on the same data and methodology as that used to
develop the  1993 estimate.  To develop the 1995 estimate, data for 11  facilities were adjusted
to reflect reduced emissions levels. During the 1993 inventory,  11 facilities were found to
ha\e elevated dioxin emission concentrations and corrective actions were initiated at these
MWCs to  reduce emissions.  The 1995 CDD/CDF emissions estimate incorporates the
expected emissions reductions at these facilities after corrective actions have been taken.  The
development of the  1995 CDD/CDF estimate is discussed in detail in Reference  1.

The national emissions estimate of 2,3,7,8-TCDD TEQ from MWCs is 1.61 Ib/yr.
                                         A-13
-------
MEDICAL WASTE INCINERATION

Basis for Calculation

There are approximately 3,400 medical facilities in the United States.2 Using facility
capacities, it was estimated that about 846,000 tons of medical waste were incinerated in
1995.5 The national dioxin/furan emissions estimates for medical waste incinerations (MWIs)
were obtained directly from work done by EPA's Emission Standards Division (ESD) to
support Maximum  Achievable Control Technology (MACT) standards for this source
category.6 A summary of the methods used to determine the estimates are provided here.
More detailed information on the derivation of the estimates can be obtained from
Reference 6.

The starting point for the national estimates is a  1995 inventory of existing MWIs, which
includes for each MW1 the location, type (batch  or non-batch), and the design capacity of the
unit.  The information used to develop the inventory was taken from a number of sources
including a listing of MWIs prepared by the American Hospital Association (AHA), state air
permits gathered by EPA,  and a survey of MWIs in California and New York conducted by
EPA in 1995.  The AHA inventor}' was itself taken from two sources; an EPA "Locating and
Estimating" document and a vendor listing. Once this information was compiled, the
inventory was reviewed and modified based on updates from state surveys, commercial
sources, and MWI  vendors.

The capacity of each MWI was provided in the  inventory.  The capacities for the continuous
and intermittent MWIs in the inventory' were  expressed in terms of an hourly charging rate in
pounds per hour (Ib/hr).  Batch MWI capacities were provided in pounds per batch (Ib/batch).
Therefore, batch MWIs were evaluated separately from the continuous and intermittent  MWIs
and the batch capacities were not converted to hourly burning rates.6
                                         A-14
-------
Also included in the inventory was the applicable state paniculate matter (PM) emission limit
for each MWI. The AHA listing of MWIs and the state air permits included PM limits for
each MWI.  Where PM limits were not listed, limits were applied based on State regulatory
requirements using the same methods described in the MACT floor analysis for the proposed
standards.  PM limits could be used to estimate the type of emission control at MWIs for
which the control was not identified for the MWI.

Nationwide MWI emissions were calculated by first calculating MWI emissions on a unit
specific basis using the MWI inventory and considering unit  specific parameters.  Actual
emission control data was used where available and was estimated when not available. In
calculating emissions and estimating operating parameters, there are three distinct types of
MWIs as follows:  batch, intermittent, and continuous.  The difference in these three MWI
types is in the methods of charging waste to the MWI and removing ash from the primary
chamber of the MWI.   Continuous MWIs, which are the largest of the three types, have
mechanical ram feeders and continuous ash removal systems.  These features allow the unit to
operate 24 hours per day  for many days at  a time.  Most intermittent MWIs also have
mechanical ram feeders that charge waste into the primary chamber. However, intermittent
MWIs do not have an  automatic ash removal  system, and can only be operated for a limited
number of hours before the unit must be shut down for ash removal.  In batch MWIs, all of
the waste to be burned  is  loaded into the primary  chamber and, once the burning  cycle begins,
no additional waste  is loaded.  After the burn cycle for a batch unit is complete and the unit
has cooled down, the ash  is removed manually. In the inventory of existing MWIs used to
estimate the nationwide dioxin  emissions, a differentiation was made between batch and
nonbatch (continuous and intermittent) MWIs. However, no  distinction was made between
continuous  and intermittent MWIs in the final inventory.6

The hours of operation were  estimated for each MWI type in the inventory in order to
determine the annual waste incinerated.  The hours of operation were defined as the hours
during which the MWI combusts waste.
                                         A-15
-------
For batch MWIs, it is estimated that a typical MWI charges waste 160 times per year (i.e.,
160 batches/yr; 3 batches per week).  The amount of waste burned each year in a batch unit
and the yearly emissions produced depend primarily on the unit capacity and the annual
number of batches. Because of this relationship, it was unnecessary to determine the actual
hours of operation for batch MWIs.

For continuous and intermittent MWIs, operating hours were estimated for three size
categories (<500, 501 to 1,000, and >1,000 Ib/hr).  All MWIs with capacities less than
500 Ib/hr were assumed to  be  intermittent MWIs.  The waste charging hours for intermittent
MWIs with capacities less than 500 Ib/hr were estimated  at 1,250 hours per year (hr/yt).
Since the inventory does not indicate whether an MWI is continuous or intermittent, a ratio of
about 3 to  1, intermittent to onsite continuous, was used to estimate the hours of operation for
onsite continuous and intermittent MWIs with capacities greater than 500 Ib/hr.  The average
operating hours for continuous MWIs in the 501 to 1,000 Ib/hr size category was 2,916 hr/yr
and the average charging hours for  intermittent  MWIs in  this size category was 1,500 hr/yr.
The weighted average of the charging hours for the combined continuous and intermittent
MWIs was determined as follows:
                  (1,500 hr/yr x 0.77) -t- (2,916 hr/yr x 0.23) = 1,826 hr/yr

Large MWIs with design capacities greater than 1,000 Ib/hr were  estimated to operate
2,174 hr/yr and all commercial MWIs were estimated to operate 7,776 hr/yr.  A summary of
the waste charging hours for the continuous and intermittent MWIs is presented below.0
MWI Capacity
<500 Ib/hr
50 1-1, 000 Ib/hr
> 1,000 Ib/hr
All commercial MWIs
Charging Hours
(hr/yr)
1,250
1.826
2,174
7,776
Capacity Factor
(%)
29
33
40
89
                                          A-16
-------
NOTE:       "Capacity factor" means ratio of tons of waste actually burned
              per year divided by the tons of waste that could be burned per
              year had the unit ben operating at full capacity.
A capacity factor represents the percentage of operational time a MWI has operated in 1 year.
Capacity factors were calculated for each MWI size category based on the ratio of the actual
annual charging hours to the maximum annual charging hours. For intermittent MWls, the
maximum annual charging hours were estimated to be 4,380 hr/yr. This is based on
12 charging hours per day and 365 days per year because intermittent MWIs must shut down
for daily ash removal.  The maximum annual charging hours for continuous MWIs were
estimated to be 8,760 hr/yr based on 24 charging hours per day and 365 days per year.  All
commercial MWIs were assumed  to be continuous units.   The maximum annual charging
hours for onsite MWIs with capacities greater than  500 Ib/hr were estimated to be 5,475 hr/yr,
based on the 3 to 1  ratio of intermittent to onsite continuous MWIs discussed previously.  The
calculated capacity factors for each MWI size category are shown in  the box on the previous
page.

Waste charging rates measured during emissions tests show the average hourly charging rates
to be about two-thirds of the MWI design rates specified by incinerator manufacturers.
Therefore, waste was assumed to be charged at two-thirds of the MWI design capacity.  Using
the operating hours per year (or number of batches  per year for batch units) and  the corrected
waste charge rate (two-thirds of the design rate), the amount of waste burned  annually was
determined for each MWI.

Actual emission control data was used where available and was estimated when necessary.
When emission control system type was unknown it was  estimated based on (1) the average
PM emission rates for the different types  of emission  controls  and (2) the PM limit to which
the MWI is subject.  For example, the average PM  emission rate for  intermittent and
continuous MWIs with 1/4-second combustion control was estimated from test data to be
0.30 grains per dry standard cubic feet (gr/dscf).  Thus, any MWI with a PM  emission limit
                                         A-17
-------
greater than 0.30 gr/dscf was assumed to have a 1/4-second combustion system.  The PM

emission limit ranges for all of the emission controls are shown below.
    PM Emission Limit (gr/dscf at 7% O2)
Assumed Level of Emission Control
 Intermittent and continuous MWIs

       >0.3
       0.16  0.079
       0.042 < x < 0.079
       0.026 < x O.042
       <0.026
    1/4-sec combustion control
     1-sec combustion control
     2-sec combustion control
          Wet scrubbers
An analysis of EPA-sponsored emission test data showed a direct relationship between the
CDD/CDF emissions on a "total" dioxin basis and a "TEQ" basis. For total CDD/CDF

emissions greater than 150 nanograms per dry standard cubic meter (ng/dscm), the ratio of
total  CDD/CDF emissions to the TEQ emissions was 48:1.  For total  CDD/CDF emissions
less than 150 ng'dscm, the ratio was 42:1.  These ratios were used with test data on total
CDD/CDF emissions to develop TEQ emission factors for each type of emission control.  The

resulting dioxin and TEQ emission factors are shown in Table A-3.


                                    Table A-3

                    Total Dioxin and TEQ Emission Factors
Type of Emission Control
1/4-sec combustion control
1-sec combustion control
2-sec combustion control
Wet scrubbers
TEQ Factors, Ib/TEQ
Dioxin/lb Waste
3.96 x 10'9
9.09 x lO'10
7.44 x 10-"
1.01 x 10'"
Dioxin Factors, Ib Total
Dioxin/lb Waste
1.94 x 10'"
4.45 x 10'*
3.65 x 10-';
4.26 x 10-'°
                                        A-18
-------
                                    Table A-3
             Total Dioxin and TEQ Emission Factors (Continued)
Type of Emission Control
Dry scrubbers no carbon
Dry scrubbers with carbon
Fabric filter/packed bed
TEQ Factors, Ib/TEQ
Dioxin/lb Waste
7.44 x 10'11
1.68 x 10"n
6.81 x 1CT10
Dioxin Factors, Ib Total
Dioxin/lb Waste
3.65 x 10-'°
7.04 x 10-"
3.34 x lO'8
In combination with the MWI parametric data, control technology data, and emission factors
the following equation calculates the annual dioxin emissions from each MWI in the MWI
inventory:
      Where:
             Emissions = (C x H x C,) x F x Q


Emissions    =     Annual dioxin emissions, g/yr
C            =     MWI capacity, Ib/hr
H            =     Charging hours, hr/yr
C,           =     Ratio of waste charging rate to design capacity, 2:3
F            =     Emission factor for the appropriate level of control (Ib
                   dioxin/lb waste charged), and
Q            =     Conversion factor for pounds to grams, 453.6 grams/lb.
The CDD/CDF emissions from the individual MWIs in the inventory were calculated by
multiplying the annual amount of waste burned by the appropriate emission factor from
Table A-3.  Next, the annual emissions from each MWI were summed to estimate the total
1995 CDD/CDF emissions from MWIs. CDD/CDF emissions from MWIs in 1995 are
estimated to be 16 pounds on a total mass basis and 0.332 pounds on a 2,3,7,8-TCDD TEQ
basis.0
                                       A-19
-------
SEWAGE SLUDGE INCINERATION

Basis for Calculation

In 1995, there were 143 operating sewage sludge facilities in the United States,2 and more
than 80 percent of the facilities were multiple hearth incinerators.7 In 1992, approximately
9.5x10^ tons of dry sewage sludge were incinerated.2 Emission factors for 2,3,7,8-TCDD and
2,3,7,8-TCDF from a multiple hearth incinerator with an impingement tray scrubber in place
were obtained from AP-42.7 The emission factors were multiplied by the tons of sludge
incinerated to estimate emissions.
Activity Level
9.5xl05
ton incinerated
2,3,7,8-TCDD
Emission Factor
l.OxlO'9
Ib/ton sludge
incinerated
2,3,7,8-TCDF
Emission Factor
3.6xlO'7
Ib/ton sludge
incinerated
2,3,7,8-TCDD TEQ
Emission Factor
5.57x10-"
Ib/ton sludge
incinerated
Example Calculation
2.3."\8-TCDD national emissions estimate = (9.5  x 10'  toa/yr incinerated)
                                           x (1.0 x 10~9  Ib/ton incinerated)
                                         = 9.5 x 10-4  Ib/yr

2.3.T.8-TCDF  national  emissions estimate  = (9.5 x  10? ton/yr incinerated)
                                           x  (3.6 x  10'"1 Ib/ton incinerated)
                                         = 3.42  x 10'! Ib/yr

2,3,7,8-TCDD TEQ national emissions estimate =  (9.5 x 105 ton/yr incinerated)
                                                 x  (5.57 x  10~8 Ib/ton incinerated)
                                              =  5.29 x 10'2 Ib/yr
                                         A-20
-------
HAZARDOUS WASTE INCINERATION

Basis for Calculation

The activity data for dioxins/furans were derived from total quantities of hazardous waste
generated.  In 1992, approximately 249 million metric tons (274 million tons) of hazardous
waste were generated.2 It is estimated that of the total amount of hazardous waste generated,
only 1.3 million metric tons (1.43 million tons) were burned in dedicated hazardous waste
facilities, and 1.2 million metric tons (1.32 millon tons) were burned in boilers and industrial
furnaces (BIFs).8

Emission factors reported in Section 4.1.4 of this document for hazardous waste incinerators
and the activity data reported in Reference 2 were used to estimate national emissions of
2.3,7,8-TCDD/TCDF. The factors were developed from testing performed at the EPA's
Incineration Research Facility designed to evaluate PCB destruction and removal efficiency.
The waste feed during testing was PCB-contaminated sediments.  The test incinerator was
equipped with a venturi scrubber followed by a packed column scrubber. The  activity level
and emission factor were  multiplied to calculate national 2,3,7,8-TCDD/TCDF  emissions from
hazardous waste incinerators.
                              2.3,7,S-TCDD Emission        2,3,7,8-TCDF Emission
       Activity Level                   Factor                        Factor
         1,43x10°                    1.68x10-'°                      1.91x10"'
       ton incinerated           Ib/ton waste incinerated        Ib/ton waste incinerated
Calculations
2,3,7,8-TCDD national emissions estimate = (1.68 x 10"10 Ib/ton waste incinerated)
                                           x (1.43 x 106 ton/yr waste incinerated)
                                         = 2.40 x 10-4 Ib/yr
                                         A-21
-------
2,3,7,8-TCDF national emissions estimate = (1.91  x 10~8 Ib/ton waste incinerated)
                                          x (1.43 x 106 ton/yr waste  incinerated)
                                        = 0.0273 Ib/yr


The national emissions estimate for 2,3,7,8-TCDD TEQ from HWIs was developed as part of
recent EPA regulatory programs for hazardous waste combustors.  The TEQ estimate is 22
grams/yr (0.049 Ib/yr) and represents 1996 emissions.9'10
                                        A-22
-------
LIGHTWEIGHT AGGREGATE KILNS (LWAKs)
Basis for Calculation
The national emissions for LWAKs was obtained from an EPA document that presents
national emissions estimates for hazardous waste combustor systems.10'11  The document
presents an estimate of 6.92x10'3 Ib/yr of 2,3,7,8-TCDD TEQ emissions from LWAKs.
                                     A-23
-------
PORTLAND CEMENT

Basis for Calculation

National emissions estimates for the Portland cement source category were developed for kilns
burning non-hazardous waste (NHW) and for kilns burning hazardous waste (HW). The
estimate for NHW kilns was taken directly from estimates prepared by the EPA to support the
Portland Cement Manufacturing Industry NESHAP standards program.12'13  The estimate for
HW kilns was obtained from an EPA document that presents national emissions estimates for
hazardous waste combustor systems.10'11  The details of the estimation process and the data
used to  develop national estimates can be found in References  10 and 13.

Non-Hazardous Waste (NHW)  Kilns

Emissions from NHW kilns were only estimated on the basis of a dioxin/furan TEQ.  Average
TEQ concentrations were determined from actual test data for existing facilities of varying
types, design, and control configurations.  An average TEQ concentration of 0.25 ng/dry
standard cubic meter (dscm) of flow was determined and used  to calculate national emissions.
The national kiln clinker production rate used in the calculation was 67.6 million tons of
clinker produced per year from NHW kilns. Additional information used in the calculation
\\ere 66.225 dscf of flow/ton dry feed material  and 1.65 ton dry feed/ton of clinker produced.
The values  for these variables are  included in the Technical Background Document for the
standard. The equation used to calculate the national emissions estimate from NHW kilns is
as follows:12
                  ng        m3        g        Ib      66,225 dscf
                 	g__ Y  	 Y    C    V ———— X - ---—  —      Y
                          35.3 ft3    109 ng   454 g    ton dry feed
                          1.65ton dry feed    ton clinker    Ib TEQ
                          ~"                A  "" '""                  ~
                            ton clinker           yr          yr
                                         A-24
-------
The national estimate of 2,3,7,8-TCDD TEQ emissions from NHW kilns is 0.12 Ib/yr.

Hazardous Waste Kilns

The national emissions for HW kilns was obtained from an EPA document that presents
national emissions estimates for hazardous waste combustor systems.10'"  This document
presents an estimate of 0.13 Ib/yr of 2,3,7,8-TCDD TEQ emissions from HW kilns.
                                       A-25
-------
WASTE TIRE INCINERATION


Basis for Calculation


Emission factors for 2,3,7,8-TCDD/TCDF emissions from waste tire incineration (tire-to-

energy facility using a spray dryer, flue gas desulfurization followed by a fabric filter) were

obtained from source  testing.14  A national estimate of 5.5 x  105 tons of waste tires incinerated

per year was obtained from EPA's Office of Solid Waste.15  The activity data and emission

factors were multiplied to calculate national 2,3,7,8-TCDD/TCDF emissions from waste tire

incinerators.


                       2,3,7,8-TCDD           2,3,7,8-TCDF        2,3,7,8-TCDD TEQ
  Activity Level       Emission Factor        Emission Factor        Emission Factor

     5.5 x  105           2.16 x lO'11             5.42x10-"             1.08 x 1(T9
     ton tires      Ib/ton tires incinerated  Ib/ton tires incinerated   Ib/ton tires incinerated
    incinerated


Calculation

2.3.7,8-TCDD national emissions estimate =  (5.5 x 105 toa/yr  tires  incinerated)

                                            x  (2.16 x  ICT11 Ib/ton tires incinerated)

                                         =  1.19 x  ID"5 Ib/yr


2.3,7.8-TCDF national emissions estimate = (5.5 x 10s  ton/yr tires incinerated)

                                           x (5.42 x 10""  Ib/ton tires  incinerated)

                                         = 2.98 x 10-5 Ib/yr


2,3,7,8-TCDD TEQ  national emissions estimate = (5.5 x 105 ton/yr tires incinerated)

                                                 x (1.08  x 10"9  Ib/ton  tires incinerated)

                                               = 5.94  x lO'4  Ib/yr
                                         A-26
-------
UTILITY COAL COMBUSTION

Basis for Calculation

The national 2,3,7,8-TCDD, 2,3,7,8-TCDF, and 2,3,7,8-TCDD TEQ emissions estimates and
factors for utility coal combustion were obtained from an EPA study on toxic pollutants from
utility boilers conducted over the past several years.16  The EPA is conducting this study in
response to a Clean Air Act mandate to prepare a Report to Congress on toxic emissions from
utility sources.  It is important to note that these data are preliminary and have not yet been
finalized by the EPA.  Also, the factors do not represent a specific source but are composites
of individual factors for various furnace configurations and control devices.  The factors and
estimates developed from the utility boiler study are presented below.


CDD/CDF
2,3,7,8-TCDD
2,3.7.8-TCDF
2.3.7.8-TCDD TEQ

Emission Factor
(Ib/trillion Btu)
1.6 x 1Q-4
3.9 x lO'0
Not reported
1990
Emission Estimate
(ton/yr)
1.4 x 10'5
3.4 x 10'5
1.5 x JO"4
1990
Emission Estimate
(lb/yr)
0.028
0.068
0.30
Note  These values are draft estimates and have not been finalized bv the EPA.
                                        A-27
-------
UTILITY RESIDUAL OIL COMBUSTION

Basis for Calculation

The national 2,3,7,8-TCDD, 2,3,7,8-TCDF, and 2,3,7,8-TCDD TEQ emissions estimates and
factors for utility residual oil combustion were obtained from an EPA study on toxic
pollutants from utility boilers conducted over the past several years.16  The EPA is conducting
this study in response to a Clean Air Act mandate to prepare a Report to Congress on toxic
emissions from utility sources.  It is important to note that these data are preliminary and have
not yet been finalized by the EPA.  Also, the factors do not represent a specific source but are
composites of individual factors for various furnace configurations and control devices.  The
factors and estimates developed from the utility boiler study are presented below.
CDD/CDF
2,3,7,8-TCDD
2,3.7,8-TCDF
2.3.7.8-TCDD TEQ
Emission Factor
(Ib/trillion Btu)
6.5 x ID'6
4.6 x 10'6
Not Reported
1990 Emission
Estimate (ton/yr)^
4.0 x ID'6
2.9 x 10'6
1.1 x 1(T5
1990 Emissions
Estimate (Ib/yr)
0.008
0.0058
0.022
Note.  These \alues are draft estimates and have not been finalized bv the EPA.
                                        A-28
-------
INDUSTRIAL WOOD COMBUSTION

Basis for Calculation

Emission factors for 2,3,7,8-TCDD and 2,3,7,8-TCDF emissions from industrial wood
combustion were obtained from source test data and used to estimate national emissions. The
emission factors represent the average of two processes, controls, and fuel types.2  Data from
the nine boilers tested by the National Council of the Paper Industry for Air and Stream
Improvements, Inc. (NCASI) were used to develop the 2,3,7,8-TCDD TEQ emission factor
that was used to estimate national emissions.17  A national estimate of the amount of wood
combusted in industrial boilers was obtained from the Department of Energy.18  The activity
data and the emission factor were multiplied to calculate national 2,3,7,8-TCDD/TCDF
emissions from industrial wood combustion.
  Activity Level
 2,3,7,8-TCDD
Emission Factor
 2,3,7,8-TCDF
Emission Factor
2,3,7,8-TCDD TEQ
  Emission Factor
9.06 x 10"
ton dry wood
burned
7.34 x 10-"
Ib/ton dry wood
burned
1.05 x 10'10
Ib/ton dry wood
burned
2.48 x lO'1'
Ib/ton dry wood
burned
Example Calculation

2.3.~,8-TCDD TEQ national emissions estimate = (9.06 x 107 ton/yr dry wood burned)
                                              x (2.48 x 10"9 Ib/ton dry wood burned)
                                            = 2.25 x 10-' Ib/yr

2.3.7,8-TCDD national emissions estimate = (9.06 x  107 ton/yr dry wood burned)
                                         x (7.34  x 10"" Ib/ton dry wood burned)
                                       = 6.65 x  10'3 Ib/yr
                                       A-29
-------
2,3,7,8 -TCDF national emissions estimate = (9.06 x 107 ton/yr dry wood burned)
                                          x (1.05 x 10'10 Ib/ton dry wood burned)
                                        = 9.51 x 1C'3 Ib/yr
                                         A-30
-------
RESIDENTIAL COAL COMBUSTION


Basis for Calculation


Emission factors based on dioxin/furan concentrations in soot samples collected from seven

coal furnaces and AP-42 particulate matter (PM) emission factors were obtained for both
bituminous and anthracite coal combustion.2  For the purposes of estimating emissions, it was
assumed that the concentrations of CDD/CDF in the PM emitted from residential coal

combustion are the same as those measured in the soot samples.   A 1990 national estimate of

the amount of residential coal combusted for the two coal types was obtained from the

Department of Energy report.19 The activity levels and  emission factors for the two coal types

were multiplied and then added together to estimate national 2,3,7,8-TCDD/TCDF emissions

from residential coal combustion.
  Activity Level
    2,3,7,8-TCDD
   Emission Factor
    2,3,7,8-TCDF
   Emission Factor
 2,3,7,8-TCDD TEQ
   Emission Factor
   1.93 x 106ton
  bituminous coal


   ".32 x 10' ton
   anthracite coal
  4.79 x ID'9 Ib/ton
   bituminous coal
       burned
  3.20 x 10-° Ib/ton
anthracite coal burned
  1.26x  10-7 Ib/ton
   bituminous coal
       burned
  8.39 x  ID'8 Ib/ton
anthracite coal burned
  1.97 x 10'7 Ib/ton
   bituminous coal
       burned
  1.20x 10-7 Ib/ton
anthracite coal burned
Note  These values are draft estimates and have not been finalized by the EPA.
                                        A-31
-------
Calculation




 2,3,7,8-TCDD national emissions   ,„ -n   in_9 ,,u    ,.^   .         , ,     ,.

     estimate,  bituminuous coal     = <4'79 X I0  Ib/t0n blt™ous coal burned)



                                    x (1.93 x 106  ton/yr bituminous coal burned)



                                  = 9.24 x 1C'3 Ib/yr
 2,3,7,8-TCDD national emissions   ,-, OA   m-9 iu/*      *u          * v    j%
      '  .    ,     4.   ..      ,      = (3.20 x 10 9 Ib/ton anthracite coal burned)
      estimate, anthracite coal        v                                       '



                                    x (7.32 x 105  ton/yr anthracite coal burned)



                                  = 2.34 x I0'3 Ib/yr



 2,3,7,8-TCDD national emissions =           3      +           ,

     estimate  (both coal  types)                     •*                  J



                                  = 1.16 x 10'2 Ib/yr
 2,3,7,8-TCDF  national emissions    .,, n_    in6  .   ,   ...            , ,     ,.
      '.,,.,.          .    = (1.93 x  10" ton/yr bituminous coal burned)
    estimate, bituminuous coal       v               J                        '



                                    x (1.26  x  10~7 Ib/ton bituminous  coal burned)



                                 = 2.43 x  10-' Ib/yr



 2,3,7,8-TCDF  national emissions    ,_ _0    1A5  ,   ,     ,,         . ,      ,,
        ...     ,,   ..      ,     = (7.32 x  103 ton/yr anthracite  coal burned)
      estimate, anthracite coal        v               J                       '



                                    x (8.39  x  10'8 Ib/ton anthracite coal burned)



                                 = 6.14 x  ID'2  Ib/yr



 2,3,7,8-TCDF  national emissions    - ._    .„.,  ., ,      , . .    .„ , ,, ,
       .•   .  f,  .,     , .     x    = 2.43 x  10 '  Ib/yr + 6.14  x 10 2 Ib/yr
     estimate (both coal types)                    J                  J



                                 = 3.05 x  lO'1  Ib/yr
                                          A-32
-------
2,3,7,8-TCDD TEQ national emissions   ,. n,   1A6 t   ,   ,.    .         .

      estimate, bituminuous coal       = (L93 x 10  ^ b'^ous '°al



                                       x (1.97 x 10"7  Ib/ton  bituminous  coal burned)



                                     = 3.80 x 10-' Ib/yr



2,3,7,8-TCDD TEQ national emissions   ,_-0   1n, .   ,     .,   .      ,,      ,,
         ±-   .     i,   •>      ,        = (7.32 x 10- ton/yr anthracite  coal burned)
       estimate, anthracite  coal          v              J                •



                                       x (1.20 x 10"7  Ib/ton  anthracite coal burned)



                                     = 8.78 x 10-2 Ib/yr



2,3,7,8-TCDD TEQ> national emissions =           ,      +           2

      estimate (both coal types)                       J                  y



                                     = 4.68 x 10-' Ib/yr
                                       A-33
-------
RESIDENTIAL DISTILLATE FUEL OIL COMBUSTION

Basis for Calculation

Emission factors based on dioxin/furan concentrations in soot samples collected from
21 distillate fuel oil-fired furnaces used in central heating and AP-42 paniculate emission
-------
2,3,7,8 -TCDD TEQ national emissions estimate = (1.44 x  108 barrels/yr burned)
                                                x (5.26  x  10'8 lb/103 barrels  burned)
                                              = 7.57 x ID'3 Ib/yr
                                           A-35
-------
RESIDENTIAL WOOD COMBUSTION

Basis for Calculation

29.1 million cords (33.8 million tons) of wood were combusted in residential wood
combustors in 1990.20'21  The nationwide percentage of wood consumption is 28 percent for
fireplaces and 72 percent for woodstoves.2 Of the 72 percent combusted in woodstoves, no
more than five percent is combusted in catalytic and noncatalytic stoves.22 For calculational
purposes, it is assumed the remaining 95 percent (of the 72 percent) is combusted in
conventional woodstoves.

The dioxin/furan factors used to estimate emissions from residential wood combustion are
weighted emission factors that represent fireplace and woodstove use.  Dioxin/furan emission
estimates attributed to residential wood combustion were based on a methodology developed
by EPA's Office of Health and Environmental Assessment (now named the National Center
for Environmental Assessment).23  Using two recent studies (conducted in Switzerland and
Denmark) that reported direct measurement of CDD/CDF emissions  from wood stoves, an
average emission factor of 2 x 10"9 Ib TEQ/ton (1 ng TEQ/kg)  was derived.
Activity Level
3.38 x 10"
ton dry wood
burned
2,3,7,8-TCDD
Emission Factor
2.55 x 10-"
Ib/ton dry wood
burned
2,3,7,8-TCDF
Emission Factor
8.90 x lO'10
Ib/ton dry wood
burned
2,3,7,8-TCDD TEQ
Emission Factor
2.00 x 10'9
Ib/ton dry wood
burned
Example Calculation

2,3,7,8-TCDD TEQ  national  emissions estimate = (3.38 x 107 ton/yr dry wood burned)
                                               x (2.0 x 10~9 Ib/ton dry wood burned)
                                             = 6.76 x 10~2 Ib/yr
                                        A-36
-------
2,3,7,8-TCDD national emissions estimate = (3.38 x 107 ton/yr dry wood burned)



                                          x (2.55 x 10'" Ib/ton dry wood burned)



                                        = 8.62 x 10-4 Ib/yr







2,3,7,8-TCDF national emissions estimate = (3.38 x  107 ton/yr dry wood burned)



                                          x (8.90 x 10'10 «lb/ton dry wood burned)



                                       = 3.01  x lO'2 Ib/yr
                                        A-37
-------
IRON AND STEEL FOUNDRIES

Basis for Calculation

The national activity level for iron and steel foundries for 1990 is 10,199,820 ton of iron/steel
product produced.2  The national activity level estimate  for ferrous foundries includes
9.15xl06 tons of iron castings and  LlOxlO6 tons of steel castings produced by approximately
1100  foundries nationally in 1990.2 The emission factors used to estimate CDD/CDF
emissions from iron and steel foundries were derived from one facility test report.2"1 The test
report quantified emissions from a batch-operated cupola furnace charged with pig iron, scrap
iron, steel scrap, coke, and limestone.  Emission control devices  in operation during the test
were an oil-fired afterburner and a baghouse.  Fully  speciated dioxin/furan profiles were
available to calculate 2,3,7,8-TCDD TEQs.  The emission factors and activity level were
multiplied to calculate 2,3,7,8-TCDD/TCDF emissions from iron and steel foundries.
Activity Level
1.02x 107
ton iron/steel
product
2,3,7,8-TCDD
Emission Factor
2.47 x 10-'°
Ib/ton product
2,3,7,8-TCDF
Emission Factor
7.92 x 10'9
Ib/ton product
2,3,7,8-TCDD TEQ
Emission Factor
3.68 x 10-"
Ib/ton product
Example Calculation

2.3,7.8-TCDD national emissions estimate = (1.02 x 107 ton/yr product)
                                           x (2.47 x lO'10 Ib/ton product)
                                         = 2.52 x 10'3 Ib/yr

2,3,7,8-TCDF  national emissions estimate = (1.02 x 107 ton/yr product)
                                           x (7.92 x  10'9 Ib/ton product)
                                         = 8.08 x  10-2 Ib/yr
                                         A-38
-------
2,3,7,8-TCDD TEQ national emissions estimate =  (1.02 x 107 ton/yr product)



                                                x  (3.68 x 10'9 Ib/ton product)



                                             =  3.75 x  1(T2 Ib/yr
                                         A-39
-------
SECONDARY COPPER SMELTERS

Basis for Calculation

The activity data for secondary copper smelters for 1990 is available, but there is no available
emission factor that can be used with the activity data to estimate emissions.25  Test data are
available for 2,3,7,8-TCDD from one U.S.  facility, but it is not .possible to develop an
emission factor from the data.25  The test data were used to estimate annual 2,3,7,8-TCDD
emissions from the single facility, and that estimate is reported as the national emissions
estimate for the secondary copper smelting source category in this document.  Also, the
2,3,7,8-TCDD emissions data were used to calculate an annual 2,3,7,8-TCDD TEQ emissions
estimate and the estimate is presented in this document.  Thus, it should be noted that the
2,3,7,8,-TCDD TEQ emissions estimate is based on the 2,3,7,8,-TCDD congener only.
Emissions of other congeners are not accounted for in the 2,3,7,8-TCDD TEQ estimate
because data were not available.20 Also, it  should be noted that the 2,3,7,8-TCDD and
2,3,7,8-TCDD TEQ national emissions estimate presented in this document represent only one
facility.

The national emissions estimates  for 2,3,7,8-TCDD and 2.3,7,8-TCDD TEQ from secondary
copper smelters are  1.36xlO"2 Ib/yr and 1.36xlO~2  Ib/yr, respectively.  An estimate for
2,3,7,8-TCDF  emissions is not available.
                                         A-40
-------
SECONDARY LEAD SMELTERS
Basis for Calculation
The national activity level for secondary lead smelters for 1990 is 948,000 tons lead
produced.27 National emission estimates were developed using emission factors, control
technology, and production data compiled under the Secondary Lead Smelting NESHAP
program.
        28
There are three principal furnace types in operation at secondary lead smelting facilities in the
U.S.:  the blast furnace, the rotary furnace and the reverberatory furnace.  Emission control
technologies used include baghouses or a baghouse with a scrubber.

Table A-4 lists the emission factors that were used to develop the national CDD/CDF
emissions estimate for secondary lead smelters.  The dioxin/furan emission factors were
derived from industry test reports of three facilities representing the three principal furnace
types in use.29"31  Controlled (baghouse and scrubber) and uncontrolled (baghouse only)
emission factors for each furnace type were input into the NESHAP industry database to
estimate State  level emissions.  Fully speciated dioxin/furan profiles were available to
calculate 2,3.7.8-TCDD toxic equivalency.28
                Table A-4.  Secondary Lead Smelting Emission
                         Factors (Ib/ton Lead Produced)
Pollutant
Bagbouse Outlet
Scrubber Outlet
Rotary Furnace
2.3,7,8-TCDD
2,3,7,8-TCDF
2,3.7,8-TCDD TEQ
Total CDD
3.16 x lO'10
2.00 x 10'9
1.42x 10'9
1.49x 10'8
3.96 x lO'10
2.00 x lO'9
1.21 x lO'10
1.85 x 10'9
                                         A-41
-------
               Table A-4.  Secondary Lead Smelting Emission
                 Factors (Ib/ton Lead Produced) (Continued)
Pollutant
Total CDF
Baghouse Outlet
5.16 x lO'8
Scrubber Outlet
5.16 x lO'8
Blast Furnace
2,3,7,8-TCDD
2,3,7,8-TCDF
2,3,7,8-TCDD TEQ
Total CDD
Total CDF
4.46 x 10'9
1.85 x 10'8
1.76 x 10'8
2.94 x 10'7
5.10 x 10'7
5.38 x 10'10
1.97 x ID'9
1.68x ID'9
2.26 x ID'*
4.74 x 10-*
Blast/Reverb Furnace
2,3,7,8-TCDD
2,3,7,8-TCDF
2,3,7,8-TCDD TEQ
Total CDD
Total CDF
1.48x 10'10
8.34 x ID'9
2.68 x 10'9
1.12x ID'8
7.66 x lO'8
1.75 x ID'10
2.88 x 10-9
8.14 x 10-'°
1.42x 10-*
3.16 x 10's
The NESHAP estimates for dioxins/furans emissions are as follows:


      •     2,3,7,8-TCDD - 1.95 x 10'3 Ib/yr

      •     2,3,7,8-TCDF - 1.20 x 10'2 Ib/yr

      •     2,3,7,8-TCDD TEQ - 8.49 x 10° Ib/yr

      •     Total CDD - 1.27 x 10"' Ib/yr

            Total CDF - 2.50 x 10'1 Ib/yr
                                     A-42
-------
SECONDARY ALUMINUM SMELTERS

Basis for Calculation

A national 2,3,7,8-TCDD TEQ emissions estimate for secondary aluminum production was
developed from data provided by The Aluminum Association to the U.S.,EPA.32'33 Data that
could be used to develop mass emissions estimates of dioxins/furans were not available.

The emissions estimate is based on model processes that represent typical processes and
emission controls used by the secondary aluminum industry. An annual 2,3,7,8-TCDD TEQ
emission rate was developed for each process/control configuration, based on 8,760 hours of
operation per year.  In addition, a utilization factor (the percent of time that the process is
actually in operation) was developed for each configuration and the number of process units
for each configuration were identified.   To estimate actual annual TEQ emissions from a
process/control, the TEQ emission rate (Ib/yr) was multiplied by the utilization factor (percent
or fraction) and the number of process units in operation.  The data used to  develop the
emissions estimates are presented in Table A-5.
                                        A-43
-------
Table A-5
Process
Scrap Dyers
Scrap Dyers
Delacquering Units
Delacquering Units
Foundry Side-wells
Foundry Side-wells
Nonfoundry Side-wells
Other Reverberatory Furnaces
Controls
Afterburner
Afterburner/
Baghouse
Afterburner
Afterburner/
Baghouse
Baghouse
Uncontrolled
Uncontrolled
Uncontrolled
TEQ Emission
Rate (lb/yr)
1.3xlO'2
5.9xlO'3
1.5x10"
Z.OxlO'5
5.1x10"
4.25xlO'3
5.6xlO'5
5.6xlO's
Utilization
Factor
0.8
0.8
0.8
0.8
0.7
0.7
0.8
0.8
Number
of Units
19
5
14
7
41
39
8
564
Total Emissions
TEQ
Emissions
(Ibs)
2.0x10"'
2.4x1 0-
1.7xlO'3
1.1x10"
1.5x10-=
1.2x10-'
3.6x10-"
2.5x10'-
3.8x10-'
   A-44
-------
DRUM AND BARREL RECLAMATION/INCINERATION

Basis for Calculation

Approximately 2.8 to 6.4 million 55-gallon drums are reconditioned annually in the United
States.2 For purposes of this report, the average national activity for 1990 is 4,600,000 drums
reclaimed.2 National emission estimates were made using emission factors developed from
one facility test report and the reported total number of drums thermally reclaimed.
2,3,7,8-TCDD/TCDF isomer specific emission factors and homologue totals were used in
calculating 2,3,7,8-TCDD TEQs.
                        2,3,7,8-TCDD         2,3,7,8-TCDF       2,3,7,8-TCDD TEQ
    Activity Level       Emission Factor       Emission Factor       Emission Factor
  4.60xl06 55-gallon     4.61xlO'9 lb/103       8.05xlO'8 lb/103        1.09xlO'7 lb/103
      drums/yr         drums reclaimed       drums reclaimed       drums reclaimed
      reclaimed
Example Calculation

2.3,7,8-TCDD national emissions  estimate = (4.60 x 106 drum/yr reclaimed)
                                          x  (4.61 x lO'9  lb/103 drum)
                                       = 2.12 x 10-5 Ib/yr

2,3,7,8-TCDF national emissions estimate = (4.60 x 106 drum/yr  reclaimed)
                                         x (8.05 x  10-8 lb/103 drum)
                                       = 3.70 x 10~4 Ib/yr

2,3,7,8-TCDD TEQ national emissions estimate = (4.60 x 106 drum/yr reclaimed)
                                               x (1.09 x 10-7 lb/103 drum)
                                            = 5.01 x 10'4 Ib/yr
                                       A-45
-------
ON-ROAD MOBILE SOURCES

Basis for Calculation

The Federal Highway Administration's (FHWA) estimated 1992 national activity level for on-
road mobile sources is 2.2398xl012 vehicle miles of travel (VMT).34 This national activity
level estimate was developed from the 1992 annual Highway Performance Monitoring System
(HPMS) reports from each State in the nation; the HPMS reports are the standardized format
for reporting vehicle activity levels expressed as VMT to the FHWA.  The VMT estimates
account for travel by passenger cars, trucks, and motorcycles on all urban and rural roadways
within each  State.

The emission factors developed for this category reflect the  level of pollution control and the
fuel type for the vehicles from which the emissions were originally sampled.  Using EPA's
MOBILES model, separate dioxin/furan emission factors were derived for unleaded gasoline
powered vehicles (0.36 pg TEQ/km, for a national annual emission range of 0.4 to 4.1 g
TEQ/yr), leaded gasoline powered vehicles (range of 1.1 to  108 pg TEQ/km, for a national
annual emission range of 0.2 to 19 g TEQ), and diesel powered vehicles (0.5 ng TEQ/km, for
a national annual emission range of 27 to  270 g TEQ/yr).35

The VMT mix distribution in MOBILES a represents the national average distribution of. VMT
among eight gasoline and diesel vehicle classes.  The combined fraction for gasoline  vehicles
in the MOBILESa distribution is 94 percent; for diesel vehicles it is 6 percent.

                         2,3,7,8-TCDD         2,3,7,8-TCDF       2,3,7,8-TCDD TEQ
    Activity Level        Emission Factor       Emission Factor       Emission Factor
  2.2398x10'- VMT      3.60xlO-15 Ib/VMT     S.GSxlO'14 Ib/VMT     S.SSxlQ-'4 Ib/VMT
                                         A-46
-------
Example Calculation







2,3,7,8-TCDD TEQ national emissions estimate = (2.2398 x 1012 VMT/yr)



                                                x (8.85 x 10'14 Ib/VMT)



                                              = 1.98 x 10-'  Ib/yr





2,3,7,8-TCDD national emissions estimate = (2.2398 x 1012 VMT/yr)



                                           x (3.60 x 10-'5 Ib/VMT)



                                         = 8.06  x  ID"3 Ib/yr





2,3,7,8-TCDF national emissions estimate = (2.2398 x  1012 VMT/yr)



                                          x (5.65 x lO'14 Ib/VMT)



                                        = 1.27 x 10-' Ib/yr
                                        A-47
-------
PULP AND PAPER-KRAFT RECOVERY FURNACES

Basis for Calculation

The national activity level for pulp and paper industry kraft recovery furnaces for 1990 is
31,080,000 tons of black liquor solids burned.36  The emission factor for 2,3,7,8-TCDD TEQ,
as calculated from data collected by the National Council of the Paper Industry for Air and
Stream Improvement (NCASI),  is presented below for kraft recovery furnaces.17 Emission
factors and the national activity data from kraft recovery furnaces were used to estimate
2,3,7,8-TCDD TEQ emissions.

             Activity Level                   2,3,7,8-TCDD TEQ Emission Factor
  3.11xl07 ton black liquor solids burned        2.20x10"" Ib/ton black liquor solids

Example Calculation

2,3,7,8-TCDD TEQ national emissions  estimate =  (3.11 x 107 ton/yr black  liquor solids burned)
                                               x (2.2 x 10-" Ib/ton black liquor solids)
                                             =  6.84 x 10-4 Ib/yr
                                       A-48
-------
-------
WOOD TREATMENT

Basis for Calculation

The most current national activity data acquired for PCP wood treatment is for 1988 and is
10,800 tons of PCP used in wood treatment operations.2  The dioxin/furan emission factors
were derived using reported average emissions of five pressure treatment facilities in
California and their average associated PCP consumption.37 The emission data used in factor
development were derived using known concentrations of dioxin/furan species in PCP and
calculated fugitive emission rates.  Homologue totals were used in calculating 2,3,7,8-TCDD
TEQs. It was assumed that because no 2,3,7,8-tetra congener contamination was detected in
commercial PCP after dilution and mixture with co-solvents, 2,3,7,8-TCDD would not be
emitted to the atmosphere from the PCP wood treatment process.37

	Activity Level	2,3,7,8-TCDD TEQ Emission Factor
  l.OSxlO4  ton PCP used for wood treatment              7.06xlO'6 Ib/ton PCP
Example Calculation

2,3,7,8-TCDD TEQ  national emissions estimate = (7.06 x 10'6  Ib/ton  PCP)
                                               x (1.08 x 104 ton/yr PCP)
                                             = 7.62 x 1C'2 Ib/yr
                                        A-49
-------
CARBON REGENERATION/REACTIVATION

Basis for Calculation

The only data available for the amount of activated carbon consumed in a year is for water
and wastewater treatment operations.  The national activity level for activated carbon
consumption in water and wastewater treatment operations for 1990 is 71,900 tons of
activated carbon consumed.38  For calculational purposes, it is assumed that all activated
carbon used in water and wastewater treatment is regenerated. The dioxin/furan  emission
factors were derived by a weighted average of emission factors.2  The weighted emission
factors reflect  the following assumptions:  50 percent of the total  amount of activated carbon
thermally reactivated is from industrial uses and occurs in large multiple-hearth or similar
furnace types;  50 percent of the total is used for municipal wastewater/potable water treatment
applications.  2,3,7,8-TCDD/TCDF isomer specific emission factors and homologue totals
were used in calculating 2,3,7,8-TCDD toxic equivalency.
    Activity Level
  2,3,7,8-TCDD
 Emission Factor
  2,3,7,8-TCDF
 Emission Factor
2,3,7,8-TCDD TEQ
  Emission Factor
  . 19xl04 ton carbon
     reactivated
 2.10xKr10 Ib/tbn
carbon reactivated
 1.36xlO'9 Ib/ton
carbon reactivated
  3.46x10-' Ib/ton
 carbon reactivated
Example Calculation

2,3,7,8-TCDD  national  emissions estimate = (7.19 x 104 ton/yr carbon)
                                           x (2.10 x 10-'°  Ib/ton carbon)
                                        = 1.51 x  JO'5 Ib/yr
                                        A-50
-------
2,3,7,8 -TCDF national emissions estimate = (7.19 x 104 ton/yr carbon)



                                          x (1.36 x 1(T9 Ib/ton carbon)



                                        = 9.78 x ID'5 Ib/yr





2,3,7,8 -TCDD TEQ national emissions estimate  = (7.19 x  104 ton/yr carbon)



                                                x (3.46  x 1CT9 Ib/ton carbon)



                                              = 2.49 x lO'4 Ib/yr
                                         A-51
-------
FOREST FIRES

Basis for Calculation

Dioxin/furan emission estimates attributed to forest fires were based on a methodology
developed by EPA's Office of Health and Environmental Assessment (this office is now
named the National Center for Environmental Assessment).39  An average of 5.1 million acres
of biomass are burned in wildfires each year in the U.S., based on 40 years  of USDA Forest
Service data.  In 1989, 5.1 million acres were burned as a result of prescribed burning.
Biomass consumption  rates were estimated at 10.4 ton/acre for wildfires, and 8.2 ton/acre for
prescribed fires.  From these estimates, the national activity level for wildfires was estimated
at 53 million tons of biomass consumed and was estimated for prescribed fires at 42 million
tons, for a total of 95  million tons.2

Applying the emission factor developed for combustion in a woodstove [which is 0.19 Ib
TEQ/ton (1 ng TEQ/kg) biomass burned], annual  TEQ emissions from forest fires were
estimated at 0.19 Ib (86 g), with projected range from 0.06 Ib (27 g) to 0.6 Ib (270 g)
TEQ/yr.40

            Activity  Level                    2,3,7,8-TCDD  TEQ Emission  Factor
      9.50xlO~ ton biomass burned               2.00xlO"9 Ib/ton biomass burned


Example Calculation

2,3,7,8-TCDD  TEQ national emissions estimate  = (9.50 x  107  ton/yr biomass burned)
                                                x (2.00  x 10"9 Ib/ton biomass burned)
                                              = 1.90 x 10-'  Ib/yr
                                         A-52
-------
CREMATORIES

Basis for Calculation

Emission estimates attributed to crematories were based on emission factors from a CARB
source test report41 and 1991  activity data regarding the number of cremations per year.42 The
test report included emission  factor data for 2,3,7,8-TCDD and 2,3,7,8-TCDF but not for
2,3,7,8-TCDD TEQ.  The 2,3,7,8-TCDD and 2,3,7,8-TCDF emission factors were multiplied
by the activity level to calculate national 2,3,7,8-TCDD/TCDF emissions from crematories.
An emission factor for 2,3,7,8-TCDD TEQ was not available and, therefore, a national
emissions estimate for 2,3,7,8-TCDD TEQ from crematories was  not developed.
     Activity Level
2,3,7,8-TCDD Emission Factor    2,3,7,8-TCDF Emission Factor
   400,500 bodies/yr
      4.58x10'14 Ib/body
3.31x10'13 Ib/body
Example Calculation
2,3,7,8-TCDD national emissions estimate
2,3,7,8-TCDF national emissions estimate
                   =  (4.58x1O'14 Ib/body) x (400,500 bodies/yr)
                   =  1.83xlO-8 Ib/yr
                   =  (3.31xlO'13 Ib/body) x (400,500 bodies/yr)
                   =  1.33xlO-7 Ib/yr
                                        A-53
-------
REMAINING SOURCE CATEGORIES

National dioxin/furan emissions from the following source categories could not be calculated
because of lack of additional information (e.g., activity data):

       •      Industrial waste incineration;
       •      Scrap metal incineration;
       •      PCB fires;
       •      Municipal solid waste landfills; and
       •      Organic chemical manufacturing.
                                       A-54
-------
                                    REFERENCES
1.     National Dioxin Emission Estimates from Municipal Waste Combustors.
       U.S. Environmental Protection Agency, Emission Standards Division, Office of Air
       Quality Planning and Standards, Research Triangle Park, North Carolina.  June 1996.
       Docket A-90-45, Item VII-A-001.

2.     U.S. Environmental Protection Agency. Emissions inventory of Section 112(c)(6)
       Pollutants:  Polycyclic Organic Matter (POM),
       2,3,7,8-Tetrachlorodibenzo-P-Dioxin (TCDD)/2,3,7,8,-Tetrachlorodibenzofuran
       (TCDF), Poly chlorinated Biphenyl Compounds (PCBs), Hexachlorobenzene, Mercury,
       and Alkylated Lead.  Draft Report. Office of Air Quality Planning and Standards,
       Research Triangle Park, North Carolina, 1996.

3.     U.S. EPA.  Municipal Waste Combustors  - Background Information for Proposed
       Guidelines for Existing Facilities.  EPA-450/3-89-27e.  August 1989.

4.     U.S. EPA.  Compilation of Air Pollutant Emission Factors, 5th ed. , Vol. I:
       Stationary Point and Area Sources. Research Triangle Park, North Carolina:  U.S.
       Environmental Protection Agency, 1995.

5.     Federal Register, Vol. 61, No.  120, June 20, 1996. Standards of Performance for New
       Stationary Sources and Emission Guidelines for Existing Sources: Medical Waste
       Incinerators. Proposed Rule.

6.     National Dioxin Emissions From Medical Waste Incinerators.  U.S. Environmental
       Protection Agency, Emission Standards Division, Office of Air Quality Planning and
       Standards, Research Triangle Park, North  Carolina.  June 1996.  Docket #A-91-61,
       ITCM, IV-A-007.

7.     U.S. EPA.  Emission Factor Documentation for AP-42, Chapter 2.5 Sewage Sludge
       Incineration. Research Triangle Park, North Carolina:  U.S. Environmental Protection
       Agency, 1993.  pp.  2.5-13 to 2.5-17.

8.     Oppelt, E.T. Incineration of Hazardous Waste—A Critical Review. Journal of Air
       Pollution Control Association.  37(5):558-586, 1987.

9.     Memorandum from G. Rizeq, EER Corporation, to F. Behan, U.S. Environmental
       Protection Agency.  "TEQ and Mercury National Emissions Estimate for Hazardous
       Waste Incinerators."  April 3, 1997.

10.    Draft Technical  Support Document: HAPs National Emissions Estimate for Hazardous
       Waste Combustor Systems. Prepared by EER Corporation for the Environmental
       Protection Agency's Office of Solid Waste (OSW). April, 1997.
                                         A-55
-------
11.    Memorandum from G. Rizeq, HER Corporation, to F. Bekan, U.S. Environmental
       Protection Agency.  "National Emissions Estimates for TEQ and Mercury."
       April 7, 1997.

12.    Memorandum from E. Heath, Research Triangle Institute to J. Wood,
       U.S. Environmental Protection Agency.  March 18, 1996.  "Dioxin/Furan Toxic
       Equivalent Emissions from Cement Kilns that do not Burn Hazardous Waste."

13.    Draft Technical Support Document for Hazardous Waste Combustion (HWC) MACT
       Standards, Volume II: HWC Emissions Database, Appendix A: Cement Kiln Detailed
       Data Listing, U.S. Environmental Protection Agency, Office of Solid Waste and
       Emergency Response.  September, 1995.

14.    Radian Corporation. Modesto Energy Company Waste Tire-to-Energy Facility,
       Westley, California, Final Emission Test Report. Report No. 243-047-20. Research
       Triangle Park, North Carolina:  Radian Corporation, 1988.

15.    U.S. EPA.  Summary and Markets for Scrap Tires. EPA/530-SW-90-074B.
       Washington, D.C.:  Office of Solid Waste and Emergency Response,
       U.S. Environmental Protection Agency, 1992.

16.    Cole, J. (Research Triangle Institute).  Memorandum  to W. Maxwell,
       (U.S. Environmental Protection Agency), Research Triangle Park, North Carolina.
       January 19, 1995.

17.    Someshwar, A.V. Summary of Mill Test Reports on Wood Residue Combustion, Black
       Liquor Combustion, and Kraft Mill Sludge Burning.  NCASI Files.  Gainesville,
       Florida: National Council of the Paper Industry for Air  and Stream Improvement,
       Inc., 1995.

18.    Energy Information Administration.  Estimate of U.S. Biofuels Consumption 1990.
       DOE/EIA-0548(90). Washington, D.C.: Office of Coal, Nuclear, Electric, and
       Alternative Fuels, U.S. Department of Energy, 1991.  p.  9.

19.    Energy Information Administration.  State Energy Data Report.  Washington, D.C.:
       Office of Energy Markets and End Use, U.S. Department of Energy, 1992.

20.    Energy Information Administration.  February, 1993.   Household Energy Consumption
       and Expenditures 1990. DOE/EIA-0321(90).

21.    Energy Information Administration.  February 1993.  Household Energy Consumption
       and Expenditures 1990, Supplement:  Regional.  Office of Energy Markets and End
       Use.  DOE/EIA-0321(90/S).  pp. 30,  115, 205, and 297.
                                       A-56
-------
22.   Letter and attachments from David Menotti, Shaw, Potts, and Trowbridge, to
      Anne Pope, U.S. EPA.  Comments to the draft 112(c)(6) emissions inventory report.
      November 27, 1996.

23.   U.S. EPA.  June 1994.  Estimating Exposures to Dioxin-Like Compounds,  Volume II:
      Properties, Sources, Occurrence, and Background Exposures.  External Review Draft.
      EPA-600/6-88-005Cb. Office of Health and Environmental Assessment, Washington,
      DC.  pp. 3-143-3-146.

24.   California Air Resources Board. Emissions Measurement of Toxic Compounds from a
      Cupola Baghouse at a Steel Foundry.  Report No. ERC-61. Sacramento, California:
      California Air Resources Board, 1993.

25.   Research Triangle Institute.  1994. Preliminary Source Assessment for the Secondary
      Copper Smelting Industry.  Prepared for the U.S. Environmental Protection Agency,
      Research Triangle Park, North Carolina.

26.   U.S. EPA.  1987. National Dioxin Study Tier 4-Combustion Sources.  Final Test
      Report-Site 10 Secondary Copper Recovery Cupola Furnace MET-A.  EPA-450/4-87-
      014s.  Research Triangle Park, North Carolina.

27.   Larrabee, D.A. Lead. In: U.S. Industrial Outlook 1991.  U.S. Department of
      Commerce, International Trade Administration, Washington, DC.  1991.

28.   U.S. EPA.  Secondary Lead Smelting Background Information Document for
      Promulgated Standards.  EPA-453/R-95-008B.  Research Triangle Park, North
      Carolina.

29.   U.S. EPA.  Draft Emission Test Report.  HAP Emission Testing a Selected Sources at
      a Secondary Lead Smelter.  East Penn Manufacturing Company.  Prepared by Roy  F.
      Weston, Inc.   1993.

30.   U.S. EPA.  Draft Emission Test Report.  HAP Emission Testing on Selected Sources
      at a Secondary Lead  Smelter.  Tejas Resources, Inc.  Prepared by Roy F. Weston, Inc.
       1992.

31.   U.S. EPA.  Draft Emission Test Report.  HAP Emission Testing in Selected Sources at
      a Secondary Lead Smelter.  Schulkill Metals Corporation.  Prepared by Roy F.
      Weston, Inc.   1993.

32.   Memorandum from Bob Stricter, The Aluminum Association,  to Juan Santiago,
      U.S. Environmental Protection Agency.  Annual Dioxin Emissions from Secondary
      Aluminum Production.  May 17, 1996.
                                        A-57
-------
33.    Memorandum from Juan Santiago, U.S. Environmental Protection Agency, to Jack
       Johnson, Eastern Research Group, Inc.  New Dioxin/Furan Emissions Data Received
       from the Aluminum Association, Research Triangle Park, North Carolina.  January  22,
       1997.

34.    U.S.  Federal Highway Administration's Highway Statistics. Washington, D.C.: U.S.
       Department of Transportation, 1992.

35.    U.S.  EPA.  MOBILESa  Emission Factor Model. Ann Arbor, Michigan:  Office of
       Mobile Sources, U.S. Environmental Protection Agency, 1993.

36.    U.S.  EPA.  1990 Census of Pulp, Paper and Paperboard Manufacturing Facilities.
       Response to 308 Questionnaire,  Part A:  Technical Information. Washington, D.C.:
       Office of Water, U.S. Environmental Protection Agency, 1992.

37.    California Air Resources Board.  Inventory of Chlorophenol Use in the Forest
       Products Industry and Investigation of Related Emissions of Chlorinated
       Dibenzodioxins and Dibenzofurans, Final Report.  Sacramento,  CA:  California Air
       Resources Board, 1987.

38.    Humer, C.  Activated Carbon Plant Starts for American Novit.   Chemical Marketing
       Reporter. June 1992.

39.    U.S.  EPA.  Estimating Exposure to Dioxin-like Compounds, Volume II: Properties,
       Sources, Occurrence, and Background Exposures, External Review Draft.
       EPA-600/6-88-005 Cb.  Washington, D.C.: Exposure Assessment Group, Office of
       Health and Environmental Assessment, U.S. Environmental Protection Agency, 1994.

40.    Peterson, J. (Mt. Baker—Snoqualmie National Forest) and D. Ward (Forest Service
       Fire Lab). An Inventory of Particulate Matter and Air Toxic Emissions from
       Prescribed Fires in the United States for 1989.  Proceedings of the Air and Waste
       Management Association 1993 Annual Meeting and Exhibition. Denver, Colorado.
       June  14 to 18,  1993.

41.    California Air Resources Board.  Emissions Testing of a Propane Fired Incinerator at
       a Crematorium. Report No. ERC-39.  Sacramento, California:  California Air
       Resources Board, 1992.

42.    Cremation Association of North America.  Cremation Statistics. Cremationist.
       Chicago, Illinois:  Cremation Association of North America, 1992.
                                        A-58
-------
                                 TECHNICAL REPORT DATA
                 (PLEASE READ INSTRUCTIONS ON THE REVERSE BEFORE COMPLETING)
 REPORT NO.
 EPA-454/R-97-003
3. REGIMENTS ACCESSION NO.
 TITLE AND SUBTITLE
 LOCATING AND ESTIMATING AIR EMISSION FROM SOURCES OF
 DIOXJNS AND FURANS
6. REPORT DATE
   5/1/97
6. PERFORMING ORGANIZATION CODE
 AUTHORS)
 T. MOODY, J. JOHNSON
t. PERFORMING ORGANIZATION REPORT NO.
 PERFORMING ORGANIZATION NAME AND ADDRESS
 EASTERN RESEARCH GROUP, INC
 1600 PERIMETER PARK
 MORRISV1LLE, NC 27560
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
  68-D2-0160
2. SPONSORING AGENCY NAME AND ADDRESS
 U. S. ENVIRONMENTAL PROTECTION AGENCY
 OFFICE OF AIR QUALITY PLANNING AND STANDARDS (MD-14)
 RESEARCH TRIANGLE PARK, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
6. SUPPLEMENTARY NOTES
 EPA PROJECT LEADER: DALLAS W. SAFRIET
6. ABSTRACT
 TO ASSIST GROUPS INVENTORYING AIR EMISSIONS OF POTENTIALLY TOXIC SUBSTANCES, EPA PRODUCES
 A SERIES OF DOCUMENTS THAT COMPILE AVAILABLE INFORMATION ON SOURCES AND EMISSIONS OF TOXIC
 SUBSTANCES. THE INTENDED AUDIENCE OF THIS DOCUMENT INCLUDES FEDERAL, STATE AND LOCAL AIR
 POLLUTION  PERSONNEL AND OTHERS INTERESTED IN TRACKING ESTIMATES OF DIOXINS AND FURANS AIR
 EMISSIONS FROM MOSTLY COMBUSTION TYPE SOURCES.
17 . KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
DIOXINS/FURANS
ESTIMATING AIR EMISSIONS
AIR TOXIC
COMBUSTION SOURCES
EMISSION FACTORS
18. DISTRIBUTION STATEMENT
UNLIMITED
I
b. IDENTIFIERS/OPEN ENDED TERMS
UNCLASSIFIED
UNCLASSIFIED
C. COSATI FIELD/GROUP
21. NO. OF PAGES
321
22. PRICE
-------