-------
4.2.5 Smelt Dissolvtng Tank
The smelt dissolver is a large tank located below the recovery furnace
hearth. Molten smelt (sodium carbonate and sodium sulfide) that accumulates
on the floor of the furnace is dissolved in water to form green liquor in the
tank. The tank is equipped with an agitator to assist dissolution, and a steam
or liquid shatterjet system to break up the smelt stream before it enters the
solution. Contact of the molten smelt with the water causes the evolution of
large volumes of steam, which must be vented.
4.2.6 Lime Kiln
The lime kiln is an essential element of the closed-loop system that
converts green liquor (solution of sodium carbonate and sodium sulfide) to
white liquor. The kiln calcines the lime mud (calcium carbonate which precipitates
from the causticizer) to produce calcium oxide (quicklime, CaO). The quicklime
ts vetted (slaked! by the water tn the green liquor solution to form calcium
hydroxide, ;CaCOftI2» for the causticizihg reaction.
The kraft pulp industry typically uses large rotary kilns that are capable
of producting 36 to 360 megagrams (40 to 400 tons) of quicklime per day. Lime
mud is fed in at the elevated end as a 55 to 60 percent solid-water slurry. The
mud is contacted by hot gases produced by the combustion of natural gas or fuel
oil and proceeding through the kiln in the opposite direction. Large motors
turn the entire kiln at low speeds (1-2 rpm), causing the lime to proceed
downward through the kiln toward the high-terperature zone (980 to 1090°C;
1800 to 2000°F) to discharge at the lower end. As the mud moves along, it
dries in the upper section, which may be equipped with chains or baffles to
give the wet mud better contact with the gases. As the lime mud moves down
farther, it agglomerates into small pellets and finally is calcined to calcium
oxide in the high-temperature zone near the burner.
4-9
-------
Fluidized bed calciners are presently being used at four kraft pulp mills,
but the production rate of each kiln at this time is under 136 megagrams (150 tons)
of lime per day.
4.2.7 Black Liquor Oxidation Systejn
Black, liquor oxidation is the. practice of oxidizing the sodium sulfide in
either weak, or strong black, liquor to sodium th.ios.ulfate or possibly higher
oxidation states. Black, liquor oxidation is designed to decrease the emissions
from the direct contact evaporator by producing a negligible sodium sulfide
concentration in the black, liquor. In those mills which, oxidize black, liquor,
air is most often used. However, molecular oxygen has- been used instead of
air at two mills. Sparging reactors, packed towers, and bubble tray columns
have been used in single or multiple stages to provide intimate contact
between the liquor and air.
4.2.8 Condensate Stripping System
When digester and multiple-effect evaporator off-gases are condensed,
some TRS gases are partially dissolved in the condensate. To prevent the
release of kraft odor from the water treatment ponds, the TRS compounds can
be stripped from the digester and multiple-effect evaporator condensates
prior to being discharged to the ponds. The two principal ways of stripping
are air stripping and steam strippino. Stripping can be performed in multistage
(multiple tray) columns with a large countercurrent flow of air or steam.
4-10
-------
REFERENCES FOR CHAPTER 4
1. Atmospheric Emissions from the Pulp and Paper Manufacturing Industry,
EPA-450/1-73-002, September 1973, page 5. (Also published by NCASI as
Technical Bulletin No. 69, February 1974).
4-11
-------
5. EMISSIONS
5.1 NATURE OF EMISSIONS
The characteristic kraft mill odor is caused principally by a variable
mixture of hydrogen sulfide, methyl mercaptan, dimethyl sulfide, and dimethyl
disulfide. All of these qases contain sulfur, which is a necessary component
of the kraft cooking liquor.
Hydrogen sulfide emissions originate from breakdown of sodium sulfide,
which is a corr.^cnent of the kraft cooking liquor. Methyl mercaptan and
dimethyl sulfide are formed in reactions with the wood component lignin.
Dimethyl disulfide is formed through the oxidation of mercaptan groups derived
from the thiolignins.
5.1.1 Hydrogen Sulfide
Hydrogen sulfide (H2S) is a weak acidic qas which partially ionizes in
aqueous solution. The ionization proceeds in two staqes with the formation
of hydrosulfide and, with increasing pH, sulfide ions
H2S Z HS" + H+ -t S= + 2H+ (5-1)
increasing pH ->
Black liquor contains a hiah concentration of dissolved sodium sulfide
in strongly alkaline solution. If the pH were depressed, the sodium sulfide
would hydrolyze to sodium hydrosulfide. Below pH 8, appreciable unionized
hydrogen sulfide would form as the reaction equilibria in equation 5-1 moves
from right to left. It is reported that at a pH of about 8.0, most hydrogen
5-1
-------
sulfide forms hydrosulfide ions. Consequently, in normal black liquor conditions,
there is very little dissolved hydrogen sulfide in the liquor.
Due to the equilibrium between the hydrosulfide ion and water vapor,
hydrogen sulfide gas can be stripped from black liquor at steam vents. There
could be, therefore, a significant concentration of H2S in the evaporator
areas of the kraft mill.
Hydrogen sulfide is formed in the recovery furnace and lime kiln as the
sulfur-containing compounds from the black liauor or lime mud are volatilized
and reduced. Hydrogen sulfide generally represents the largest gaseous emission
from the kraft process.
5.1.2 Methyl Mercaptan
Methyl mercaptan (MeSH) is a reduced sulfur compound which ~;s formed
during the kraft cook by the reaction of hydrosuKide ion and thf methoxy-lignin
2
component of the wood:
Lignin - OCH3 + HS" * MeSH + Lignin - 0~ (5-2)
Methyl mercaptan will also dissociate in an aqueous solution to methyl mercaptide
ion. It is reported that this dissociation is essentially completed above a
pH of 12.0. Methyl mercaptan is, therefore, present in low concentrations as
a dissolved gas in the black liquor. As the pH decreases, MeSH gas is evolved.
Methyl mercaptan is primarily emitted from the digester relief and blow
where it is formed, and from the brown stock washers where the pH of the liquor
drops below the equilibrium point. Emissions decrease as the residual
concentration in the liquor diminishes.
5.1.3 Dimethyl Sulfide
Dircethyl Sulfide (MeSMe) is primarily formed through the reaction of methyl
mercaptide ion with the methoxy-lignin component of the wood. It does not,
however, dissociate as hydrogen sulfide and methyl mercaptan do:
5-2
-------
Lignin - OCH3 + MeS~ + Lignin - 0~ + MeSMe (5-3)
Dimethyl sulfide may also be formed by the disproportionate of methyl
mercaptan. At normal liquor temperature (150-200°F) it is highly volatile.
5.1.4 Dimethyl Disulfide
Dimethyl disulfide (MeSSMe) is formed by the oxidation of methyl mercaptan
throughout the recovery system, especially in oxidation towers:
4 MeSH + 02 2 2 MeSSMe + 2 H20 (5-4)
Dimethyl disulfide has a higher boiling point than any of tne other compounds
and its retention in the liquor is therefore greater.
5.2 UNCONTROLLED TOTAL REDUCED SULFUR EMISSIONS
Uncontrolled total reduced sulfur (TRS) emissions are listed in table 5-1
for each of the TRS sources under consideration. These emission rates are for
a 907 megagrams per day (1000 tons per day) kraft pulp mill. Table 5-1
also lists typical gas volume rates for each source.
5.2.1 Recovery Furnace System
TRS emissions are generated both in the furnace and in the direct-contact
evaporator. The furnace generated TRS concentration is as high as several
hundred parts per million (ppm) and as low as 1 ppm depending on the furnace
design and operation. Recovery furnace emissions are affected by the relative
quantity and distribution of combustion air, rate of solids (concentrated black
liquor) feed, spray pattern and droplet size of the liquor fed, turbulence in
the oxidation zone, smelt bed disturbance, and the sulfidity and heat content
value of the liquor fed. The impact of these variables on TRS emissions is
independent on tho. absence or presence of a direct-contact evaporator.
TRS emissions generated in the direct-contact evaporator depend largely
on the concentration of sodium sulfide in the black liquor. Acidic gases such
5-3
-------
as carbon dioxide in the flue gas react with sodium sulfide in the black liquor
to form hydrogen sulfide.
Uncontrolled TRS emissions from a conventional recovery furnace system
range from 0.75 to 31 grams per kilogram (1.5 to 62 pounds per ton) of air
dried pulp and average about 7.5 grams per kilogram (15 pounds per ton) of air
dried pulp (ADP). This is an averaae of about 550 ppm.
5.2.2 Digester System
The noncondensable gases from the relief system and the blow tank vent
contain TRS concentrations as high as 30,000 ppm. Both stream-- are so.iietimes
referred to as digester "noncondensables". TRS compounds formed in the digester
are mainly methyl mercaptan, dimethyl sulfide and dimethyl disuKide. Uncontrolled
TRS emissions from a digester system *-ange bet>'eep 0.24 and 5.25 g/k^ ADP
(0.47 and 10.5 Ib/ton ADP) and average about 0.75 g/kg ADP (1.5 Ib/ton ADP)
at a concentration of 9,500 ppm. Operating variables that affect diqester
TRS emissions include the black liquor recycle rate, cook duration, cooking
liquor sulfidity (percentage of sodium sulfide to total alkali, Na2S and NaOH,
in white liquor), and residual alkali level.
5.2.3 Multiple-Effect Evaporator System
The noricondensable gases from a multiple-effect evaporator (MEE) system
consist of air drawn in through system leaks and reduced sulfur compounds that
were either in the dilute black liquor or formed during the evaporation process.
o
TRS emissions from the MEE system are as high as 44,000 ppm. Uncontrolled
TRS emissions from a MEE system average about 0.5 g/ka ADP (1.0 Ib/T ADP) at
O
a concentration of 670C ppm.
The type of condenser used can influence the concentration of TRS emissions.
Certain types of condensers (e.g. direct-contact) allow the noncondensable qases
and the condensate to mix, which results in a limited quantity of hydrogen
5-4
-------
Table 5-1. TRS EMISSIONS FROM AN UNCONTROLLED 907 ft
(1000 TONS PER DAY) KRAFT PULP MILL*1 -
Source
Recovery Furnace
Digester System
Multiple-Effect
Evaporator System
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving Tank
Condensate Stripper
System
Typical
Exhaust Gas
Flow Rate
m3/s(acfm)
212(450,000)
3(6,200)
1(2,200)
37(79,200)
71(150,000)
14(30,000)
27(58,100)
2(4,000)
TRS Emission Range
ppm g/kq ADP (Ib/T ADP)
18-1303 0.75-31(1.5-62)
1525-30,000 0.24-5.3(0.47-10.5;
92-44,000 0.015-3.2(0.03-6.3)
3-613 0.01-2.1(0.02-4.2)
0.005-0.5(0.01-0.9)
3-335 0.005-0.37(0.01-0.73)
5-811 0.007-1.9(0.013-3.70)
-
[EGAGRAMS PER DAY
Average^ TRS
ppm g/s(lb/hr)
550 79(625)
9,500 8(63)
6,700 5(42)
170 4(33)
30 2(13)
35 0.5(4)
60 1(8)
5000 10(83)
Emission Rate
g/kg ADP(lb/T ADP)
7.5(15.0)
0.75(1.5)
0.5(1.0)
0.4(0.8)
0.15(0.3)
0.05(0.1)
0.1(0.2)
1.0(2.0)
^'Uncontrolled emission data for condensate strippers were obtained from Reference 13. Data for all other sources
were obtained from Reference 5.
^'Average values listed are calculated from data listed in Reference 5. Insufficient informt+ion was available in
Reference 5 to evaluate the operation of the units for which data were reported.
-------
sulfide and methyl mercaptan gases dissolved in the water. This reduces the
TRS concentration from the system, but increases the sulfide level in the
condensate. Sulfidity and pH of the weak black liquor also have an effect on
the TRS concentration from the multiple-effect evaporators. Higher sulfidity
levels result in higher TRS emissions. TRS levels increase with decreasing pH levels.
5.2.4 Lime Kiln
TRS emissions can be generated in the lime kiln proper and in the downstream
scrubber which is normally installed to control particulate emissions.
TRS emissions originating in the lime kiln are affected by the oxygen
content of the exhaust, the kiln length to diameter ratio, the lime mud sulfide
content, cold-end exit gas temperature, and simultaneous burning of sulfur
bearing materials contained in the lime mud (e.g. green liquor dregs; the
Q
impurities resulting from clarifying the green liquor).
If digester and evaporator condensates are used as lime kiln scrubber
water, reduced sulfur compounds can be stripped into the exit qas stream.
If the scrubbing liquor contains sodium sulfide, as it does in some
installations, H2S may be formed in the scrubber from the reaction of ^S,
C02, and water in the same manner as it is in the direct contact evaporator.
Uncontrolled TRS emissions from a lime kiln average about 0.4 g/kg ADP
(0.8 Ib/T ADP) at a concentration of 170 ppm. TRS emissions from lime kilns
range between 3 and 600 ppm (0.02 to 4.2 Ib/T ADP) depending on combustion
characteristics of the individual kilns.
5.2.5 Hrown Stock Masher System
TRS emissions from the brown stock washers arise primarily from the
vaporization of the volatile reduced sulfur compounds. TRS compounds emitted
are principally dimethyl sulfide and dimethyl disulfide.
5-6
-------
Uncontrolled TRS-emissions from the brown stock washer system average
about 0.14 g/kg ADP (0.27 Ib/T ADP) at a concentration of 30 ppm. About
0.05 g/kg ADP (5-37 ppm) are emitted from the hood vent and about 0.08 g/kg ADP
(240-600 ppm) are emitted from the filtrate tank (under) vent.
Brown stock washer TRS emissions are affected by the wash water source,
water temperature, degree of agitation and turbulence in filtrate tank, and
blow tank pulp consistency. TRS emissions will increase significantly if
contaminanted condensate from the digester and evaporator systems are used for
washing. Higher temperatures and agitation result in increased stripping of
the TRS during the washing.
5.2.6 Black Liquor Oxidation System
TRS emissions from the oxidation system are created by the stripping of
the reduced sulfur compounds from the black liquor by air passing through the
liquor. Uncontrolled TRS emissions (principally dimethyl sulfide and dimethyl
disulfide) are in the range of 0.005 to 0.37 g/kg ADP (about 3 to 335 ppm)
jo
and average 0.05 g/kg ADP (35 ppm). Oxidation systems that use only molecular
oxygen have the advantage of emitting virtually no off-gases because the
total gas stream rexacts in the sparge system.
Primary factors affecting TRS emissions from black liquor oxidation
systems are the inlet sulfide content, the temperature of the black liquor,
residence time, and the air flow rate per unit volume. TRS emissions tend to
increase for higher liquor temperatures and greater air flow rates because of
greater volatility of the gases and stripping action of the air, respectively.
TRS emissions also tend to increase with increasing sulfide concentrations in
the incoming black liquor and with increasing residence time.
5.2.7 Smelt Dissolving Tank
Because of the presence of a small percentage of reduced sulfur compounds
in the smelt, some of these odorous materials escape the tank with the flashed
5-7
-------
steam. Uncontrolled TRS emissions are as high as 2.0 g/kg ADP (811 ppm) and
13
as low as non-detectable. The average is about 0.1 g/kg ADP $60 ppm).
Several factors affect the TRS emissions. Among these are the water
used in the smelt tank, turbulence of the dissolving water, scrubbing liquor
used in the particulate control device, pH of scrubbing liquor, and sulfide
content of the particulate collected in the control device. The use of
contaminated condensate in the smelt tank or the scrubber can result in the
stripping of TRS compounds into the gas stream. Turbulence can increase the
stripping action. Increased h^S formation can occur with a increase in sulfide
content of the scrubbing liquid and a decrease in pH of the scrubbing liquor.
5.2.8 Condensate Stripping System
Presently there are onjy five condensate strippers in operation in the
U.S. kraft pulp industry. Actual TRS emission data are unavailable, but TRS
emissions from condensate strippers are expected to be high because the condensate
contains high concentrations of dissolved TRS compounds. The stripping
efficiency is greater than 95 percent. Uncontrolled TRS emissions are
estimated to be about 1 g/kg ADP (5000 ppm) from a condensate stripping system.
5.3 TYPICAL'TRS EMISSIONS
Typical controlled TRS emissions are listed in table 5.2 for each of the
sources under consideration. These values represent average TRS emissions
from existing facilities, based on the information listed in Appendix A.
Appendix A lists emission rates for five sources (recovery furnaces, lime kilns,
digesters, aultipie-effect evaporators, and brown stock washers) at each kraft
pulp mill in the United States. Information in Appendix A was obtained from
the literature, state pollution control agencies, and the kraft pulp mills.
Emission rates for uncontrolled sources are average uncontrolled values (See
section 5.2) for the industry, except where actual levels are known. Controlled
5-8
-------
Table 5-2. TRS EMISSIONS FROM THE EXISTING KRAFT PULP INDUSTRY
Average Uncontrolled Level
Source :
Recovery Furnace
Digester System
Multiple-Effect
Evaporator System
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving Tank
Condensate Stripper
System
ppm
550
9500
6800
170
30
35
60
5000
g/kg ADP
(lb/T ADP)
7.5
(15.0)
0.75
(1.5)
0.5
(l.o)
0.4
(0.8)
0.15
(0.3)
0.05
(0.1)
0.1
(0.2)
1.0
(2.0)
Percent Capacity
Control ledU)^
88.7
58.4
58.6
28.2
2.8
2.1
-
100
Typical Controlled Level
ppm
5-70
5
5
5-40
5
0-10
-
5
g/kg ADP
(lb/T ADP)
0.075-1.05
(0.15-2.1)
0.01
(0.02)
0.01
(0.02)
0.0125-0.1
(0.025-0.2)
0.01
(0.02)
0.0-0.01
CQ.O-Q.Q2)
-
0.01
(0.02)
Average National
ppm
92
4050
2920
130
30
35
60
500
g/kg ADP
(lb/T ADP!
1225
(2.5)
0.32
(0.64)
0.22
(0.43)
0.31
(0.62)
0.15
(0.3)
0.05
(0.1)
0.1
(0.2)
0.11
(0.22)
Emissions
megagrams
39,000
10,000
6,700
9,700
4,420
1,470
2,940
0.4
( 'Percentage based on mills controlled by existing state regulations, plus information collected
during previous surveys.
-------
levels listed are the actual levels where these were known; otherwise the
applicable state standard is listed.
Table 5.2 also gives an estimate of the percentage of facilities presently
controlled, and the TRS level to which they are most frequently controlled.
These estimates on the percent of facilities controlled are based on existing
or soon to be adopted state regulations. The estimates also include information
obtained from various surveys of the industry on controlled facilities which
are not presently covered by state regulations.
In most cases the typical emissions from existing facilities are equal
to or near the uncontrolled levels. A few mills presently control TRS
emissions from the brown stock washers, smelt dissolving tanks, lime kilns,
and black liquor oxidation system. The other TRS sources have been controlled
to some extent by a large percentage of the industry. The typical TRS
emissions from these sources are discussed in the following sections.
5.3.1 Recovery Furnace System
TRS emissions from direct contact systems depend on the design and operation
of the recovery furnace and,iff utilized, an oxidation system. A survey of
32 recovery furnace systems where black liquor oxidation is not used shows TRS
emissions ranging from 35 to 1300 ppm, representing 0.75 to 31 g/kg ADP (1.5 to
62 Ib/T ADP). The average is 7.7 g/kg ADP (15.4 Ib/T ADP). A survey of 17
units that utilize black liquor oxidation indicates a broad TRS emission ragge
of 0.1 to 13.0 g/kg ADP (0.2 to 25.9 Ib/T ADP), with an average value of 3.7
g/kg ADP (170 ppm). TRS emissions from noo-contact systems are usually
confined to a narrow range of about 0.015 to 0.15 g/kg ADP (1 to 11 ppm).
Based on Appendix A, it is estimated that TRS emissions from about 89
percent of the existing furnaces are either controlled by black liquor oxidation
5-10
-------
or have been replaced with or converted to a non-contact system.
It is also estimated that the average national emission level is 1.25 g/kg ADP
(92 ppmv).
5.3.2 Digester and Multiple-Effect Evaporator Systems
The digester and multiple-effect evaporators will be considered together
because their emissions are normally combined for treatment. Until recently,
the noncondensable gases were in most cases vented to the atmosphere uncontrolled.
However, several mills now incinerate the gases to control odors. Most comonly,
the gases are burned in the lime kiln. Based on EPA tests, incineration can
reduce TRS emissions to less than 5 ppm (Q.0075 g/kg ADP). It is estimated
that approximately 58 percent of the mills are incinerating these gases or are
installing systems to incinerate these noncondensables. White liquor (caustic)
scrubbers are used at a few mills. These scrubbers are only effective in
removing hydrogen sulfide and methyl mercaptan. TRS emissions from these
scrubbers are estimated to be aboit 0.5 g/kg ADP (1 Ib/T ADP).
Based on Appendix A, the average national emission rate from digester
systems is calculated to be 0.32 g/kg ADP (0.64 Ib/T ADP). The average
national emission rate from multiple-effect evaporators is calculated to be
0.22 g/kg ADP (0.43 Ib/T ADP). These values are based upon 58 percent being
controlled to 5 ppm and 42 percent being uncontrolled.
5.3.3 Lime Kiln
TRS emissions from a lime kiln installation are dependent on the operation
of the kiln, the mud washing efficiency, and the type of water used in the
scrubber. Only about 28 percent (See Appendix A) of the kilns are actually
operated to control TRS emissions. This percentage is mostly based on kilns
affected by existing state or local regulations. Based on this percentage of
5-11
-------
controlled kilns, the calculated average national emission rate for lime kilns
is 0.31 g/kg ADP (130 ppm).
5.3.4 Condensate Stripping System
All the condensate strippers in operation are controlled for TRS emissions.
Incineration is the control technique used at four mills. The TRS emissions
from these sources are estimated to be 5 ppm as mentioned in section 5.3.2.
A caustic scrubber is utilized at the remaining one mill, but no data is
available on the TRS emissions.
The average national emission rate is estimated to be 0.11 g/kg ADP
(0.22 Ib/T ADP). This is based on 5 ppm TRS being achieved at 4 mills and
50 percent control at the mill that uses a caustic scrubber.
5.3.5 Brown Stock Masher Systems, Black Liquor Oxidation Systems.aand
Smelt Dissolving Tanks
These three sources are generally not controlled for TRS emissions.
However, two U.S. mills incinerate the vent gases from the brown stock washer
systems. Two other U.S.mills use molecular oxygen in their black liquor
oxidation system, which results in no vent gases and no TRS emissions. One
mill is controlling TRS emissions from the brown stock washers by a chlorine
scrubber. One other mill is controlling TRS emissions from their brown stock
washers and black liquor oxidation system by a chlorine gas injection system.
5-12
-------
REFERENCES FOR CHAPTER 5
1. Shih, T. T., Hrutfiord, B. F., Sarkanen, K. V., and Johanson, L. N.,
Hydrogen Sulflde Vapor-Liquid Equilibrium In Aqueous Systems As a
Function of Temperature and pH. TAPPI, 50(12), 630-4, 1967.
2. McKean, W. R., Hrutfiord, B. F., Sarkanen, K. V., Kinetic Analysis of
Odor Formation 1n the Kraft Pulping Process, TAPPI, 48(12), 699-7UJ, 1965.
3. Shih, T.,T., Hrutfiord, B. F., Sarkanen, K. V., Johonson, L. N.,
Methyl Mercaptan Vapor-Liquid Equilibrium 4n Aqueous Systems As a
Function of Temperature and pH. TAPPI, 50(12), 634-8, 1967.
4. Control of Atmospheric Emissions in the Mood Pulping industry, Environ-
mental Engineering Inc., and J. E. Sirrine Company, Final Report, EPA
Contract No. CPA-22-69-18, March 15, 1970.
5. Atmospheric Emissions from the Pulp and Paper Manufacturing Industry,
EPA-450/1-73-002, September 1973. (Also published by NCASI as Technical
Bulletin No. 69, February 1974).
6. Reference 5, Table 14.
7. Reference 5, Table 3.
8. Reference 5, Table 7.
9. Suggested Procedures for the Conduct of Lime Kiln Studies to Define
Emissions of Reduced Sulfur Through Control of Kiln and Scrubber Operating
Variables. NCASI Special Report No. 70-71, January 1971.
10. Reference 5. Table A-5.
11. Factors Affecting Emission of Odorous Reduced Sulfur Compounds from
Miscellaneous Kraft Process Sources, NCASI Technical Bulletin No. 60,
March 1972.
12. Reference 5, Table 21.
13. Reference 5.
5-13
-------
14. Butryn, 6. L. and Ayers, K. C., Mead Experience in Steam Stripping
Kraft Mill Condensate, presented at TAPPI Environmental Conference,
May 14-16, 1975.
15. Air Emisslon Control Program For Hoerner Waldorf Corporation Mill
Expansion Missoula, Montana, submitted by Hoerner Waldorf Corporation
to Montana State, March 12, 1974.
16. Reference 5 , Table 15.
17. Malodorous Reduced Sulfur Emissions From Incineration of Non-condensable
Off-gases. EPA Test Report 73-KPM-1A.
5-14
-------
6. CONTROL TECHNIQUES FOR TRS FROM KRAFT PULP MLLLS
6.1 ALTERNATIVE CONTROL TECHNIQUES
The various control techniques that have been or can be applied
to the emission sources affected by NSPS is discussed in this section.
The affected sources: are the recovery furnace, digester system,
multiple-effect evaporator system, lime kiln, brown stock washer
system, black liquor oxidation system, smelt dissolving tank, and
condensate stripper system. The applicability and effectiveness of the
control techniques when retrofitted on existing facilities is also
discussed. Table 6-1 summarizes the control techniques and corresponding
TRS levels achievable for each source of TRS. Section 6.2 discusses
alternative control systems for entire kraft pulp mills. Retrofit
models are presented which permit estimates to be made of required
costs for retrofitting existing facilities with, the alternative
control systems.
6.1..1 Recovery Furnace System
TRS emissions from a recovery furnace system can originate in the
recovery furnace its,elf, or in the direct contact evaporator if this
type of evaporator is. us.ed. Most existing recovery furnace systems
have direct contact evaporators. About 75 percent of the new recovery
furnaces that have been installed in the last 5 years are, however, of
the non-contact desfgn. In these furnaces, the furnace flue gases
never directly contact the black liquor and TRS cannot be formed in the
6-1
-------
Table 6-1. TECHNIQUES FOR CONTROLLING TRS EMISSIONS
FROM SOURCES IN A KRAFT PULP MILL
Source
Control
Technique
Achievable TRS
Level
Recovery furnace
Digester system
Multiple-effect
evaporator system
Lime kiln
Brown stock
washer system '
Black liquor
oxidation system
Smelt dissolving
tank
Condensate stripping
system
1) Process controls +
black liquor oxidation
2) Process controls +
conversion to non-contact
evaporator
1) Caustic scrubbing
2) Incineration
1) Caustic scrubbing
2) Incineration
1) Process controls
2) Process controls + good
mud washing
3) Process controls, good mud
washing + caustic scrubbing
1) Incineration
1) Molecular oxygen
2) Incineration
1) Fresh water usage
1) Caustic scrubbing
2) Incineration
20 ppm (Old design
furnaces)
5 ppm (New design2
furnaces)
25 ppm (Cross recovery
furnaces)
20 ppm (Old design
furnaces)
5 ppm (New design
furnaces)
25 ppm (Cross recovery
furnaces)
7,000 ppm3
5 ppm
350 ppm3
5 ppm
40 ppm
20 ppm
8 ppm
5 ppm
0 ppm
5 ppm
0.0084 g/kg BLS
5 ppm
'Old design furnaces are defined as furnaces without welded wall or membrane wall
construction or emission-control designed air systems.
2New design furnaces are defined as furnaces with both welded wall or membrane wall
construction and emission-control designed air systems.
3Calculated based upon scrubber removing only hydrogen sulfide and methyl mercaptan
and using reference 5 to determine percent of hydrogen sulfide and methyl mercaptan
present in vent stream.
-------
evaporator. The non-contact furnace was first introduced in
1967.
Several operating and design variables that have some effect on,
or relationship to, the generation of TRS emissiona in a recovery
furnace have been identified. These include the quantity and manner
of introduction of combustion air, the rate of solids (concentrated
black liquor) feed, the degree of turbulence in the oxidation zone,
the oxygen content of the flue gas, the spray pattern and droplet
size of the liquor fed the furnace, and the degree of disturbance of
1 2
the smelt bed. ' The effect of these variables is independent of the
absence or presence of a direct contact evaporator. There is no
evidence that sulfide content of the liquor combusted in the furnace
o
bears any relationship to the TRS emissions from the recovery furnace.
This is not to be confused, however, with sulfur compounds generated
or stripped in a direct contact evaporator.
The age of existing furnaces has been reported to be a significant
indicator of the furnace's ability to control TRS emissions. (The
typical life of a recovery furnace is considered to be 25 years. )
Generally,the age reflects an absence or lack of refinement in
controls and instrumentation that assist the operator in maintaining
close control of the process. Also, older furnaces may not incorporate
recent manufacturers' improvements, such as new;means of introducing
air, flexibility in distributing air in the furnace and means to change
ai.r velocity at injection ports. Furthermore, a major design change
was made to recovery furnaces in late 1964. This change consisted of
6-3
-------
installing a jnembrane between the wall tubes, located in front of the
furnace's wall insulation. Thi.s design change made the furnace air-
tight. The wall insulation on furnaces, without this membrane wall
concept tends to deteriorate. This allows air to leak into the furnace.
Thi.s in turn affects; the combustion in the furnace and reduces
significantly the capability of the operator to control TRS emissions.
These older recovery furnaces could be modified to incorporate
these new design features, but the modifications would be extremely
Q
expensive. However, changes in operating procedures can more easily be
made.
There are two control techniques to reduce TRS emissions from the
direct contact evaporator: black liquor oxidation and conversion to
a non-contact evaporator. Black liquor oxidation inhibits the reactions
between the combustion gases and black liquor that normally generate
hydrogen sulfide. This is accomplished by oxidizing the Na2S to
Na^SpO- in the black liquor before it enters the direct contact
evaporator. In converting to a non-contact evaporator, the direct
contact between furnace gases, and black liquor is eliminated, and
hydrogen sulfide formation is prevented.
There are several modes, of operation of black liquor oxidation
systems.. The black liquor is sometimes oxidized before being
concentrated in the multiple-effect evaporators (weak black liquor
oxiddtion), sometimes following evaporation (strong black liquor
oxidation] and sometimes, both, before and after. Air is the normal
oxidizing agent, but molecular oxygen is also used when available on
site. Air sparging reactors are the most common units, but
6-4
-------
packed towers and bubble tray towers are also used.
In modifying an existing recovery furnace with, a direct contact
evaporator to a non-contact design, a black liquor evaporator
(concentrator) and a second feed water economizer is necessary. In
addition, elimination of the. existing direct contact evaporator will
result in an increased parti.culate concentration discharge from the
furnace system into the. particulate control device. To maintain
particulate emissions at the original level, it will be necessary
to replace the existing collector with a new higher efficiency
precipitator or install an additional secondary collector. This
conversion to a non-contact design has. been accomplished by at least
two pulp mills.
At one recovery furnace system, erected in 1966, where the
conversion was made, TRS emissions decreased from approximately 400
q
ppm to about 10 ppm. Modifications also included changes to the
operation of the furnace, such as, oxygen content and air distribution.
Therefore, a portion of the TRS reduction is attributable to decreased
emissions of TRS from the furnace system.
TRS emissions from direct contact systems depend on the design
and operation of the recovery furnace and the black, liquor oxidation
system. An analysis, of 2QQ stack gas samples showed the relation
between oxidation efficiency and TRS emissions, presented in Table 6-2.
Since these samples were taken at stacks on new recovery furnaces, the
furnace TRS contribution is assumed to be negligible. The data
show: a clear relationship between oxidation efficiency and TRS
6-5
-------
Table. 6-2
Relationship of Oxidation Efficiency and
Emissions
11
Oxidation
Efficiency
Percent
80-85
85-90
90-94
94-96
96-98
98-99
99-100
Number
of
Samples
8
15
29
18
15
19
96
H0S. Emissions
gO/Kg
Max Win Wean
4.1 0.75 2.3 8.1
3.0 0.05 1.6 6.0
3.3 0.25 1.2 6.6
2.2 0.05 0.9 4.3
1.4 0.05 0.65 2.8
1.1 0.0 0.35 2.1
1.6 0.0 0.2 3.2
Lb bLS/Ton Pulp
Win Mean
1.5
0.1
0.5
0.1
0.1
0.0
0.0
4.6
3.2
2.4
1.8
1.3
0.7
0.4
6-6
-------
emissions. Emissions from the direct contact evaporator will'.'be
eliminated by conversion to a non-contact type system. As. mentioned
previously, neither black liquor oxidation nor conversion to a non-
contact system are effective in reducing TRS emissions from the
furnace proper.
TRS emission tests conducted during the NSRS development program
indicate that TRS emissions from new recovery furnaces can be controlled
12
to at least 5 ppra. During the NSPS program, three recovery furnaces
Ctw.o direct contact systems and one non-contact system) were tested
by EPA. TRS emissions averaged from the individual furnaces are 1.4
ppm (_6 tests, each 4-hours) Q.6 ppm (6 tests, each 4-hours), and
3.9 ppm (_S tests, each 4-hours). Only one 4-hour test showed emissions
greater Cabout 7 ppra) than 5 ppm.
One furnace manufacturer indicates that many recovery furnaces built
since 1965 are basically the same design as new furnaces built today, and
that these should also be capable of achieving 5 ppm TRS with good process
control and either black liquor oxidation or conversion to a non-contact
13
evaporator. These existing recovery furnaces, defined as "new design"
furnaces, were designed for low TRS emissions (i.e., incorporates manufacturer's
improvements) and will have stated in their contracts that these furnaces
14
were constructed with air pollution control as an objective. Recovery
furnaces, mainly those built before 1965, that were not constructed with
air pollution control as an objective, have a somewhat different desiqn,
as mentioned previously, and are not capable of achieving 5 ppm TRS
(4-hour average basis). These furnaces, defined as "old design" furnaces,
6-7
-------
can generally achieve about 15 to 20 ppm TRS with good process control
15
and black liquor oxidation or conversion to a non-contact evaporator.
This TRS level (15 to 20 ppm) is presently being achieved by existing
recovery furnaces (see AppendixB), many built before 1965, in those
16
states which have TRS regulations of 17.5 ppm (daily average). Some
existing furnaces may have difficulty achieving even this level (20 ppm)
if they are operating at a much hiqher Airing rate than originally
designed or do not have the sufficient combustion control capability.
Cross recovery liguors are somewhat different than straight kraft
liquor, and, therefore, it is possible that the TRS emissions from a cross
recovery furnace are not controllable to the same degree as are those from
the straight kraft furnace. There are three reasons why TRS emissions
may be higher from cross recovery furnaces. The first relates to the
sulfur content of the liquor which is higher with this process than in
straight kraft processes. In cross recovery operations, the heat content
of the black liquor is lower than found in straight kraft mills. This is
because the NSSC process gives higher pulp yields than the kraft process
and, as a consequence, the spent liquor associated with the NSSC process
contains less organic content. Therefore, its Btu value is lower as compared
with kraft black liquor. The third reason pertains to the restriction on
excess oxygen available in cross recovery furnaces to oxidize the relatively
large quantities of volatile sulfur compounds given off as a consequence
of the heavy sulfur loading and lower furnace operating temperatures. If
enough excess oxygen is supplied to completely oxidize all volatile sulfur
compounds, a sticky dust problem will develop which can plug up the
precipitator and render furnace operation impossible.
6-8
-------
18
Based on a study conducted on one cross-recovery furnace, cross-
recovery furnaces which experience areen liquor sulfidities in excess of
28 percent and liquor mixtures of more than 7 percent NSSC on an air dry
ton basis can not achieve the same TRS levels as straight kraft recovery
furnaces. Emission data reported in the study indicate that TRS emission
levels of 25 ppm, corrected to 8 percent oxygen, can be achieved from
well-controlled cross-recovery furnaces.
A recently developed control technique for recovery furnaces is alkaline
adsorption with carbon activated oxidation of the scrubbing solution.
Pilot'plant studies indicate that this technique can reduce TRS emissions
19
from 20 to 2500 ppm to between 1 and 10 ppm. Reduction in particulate and
S0_ emissions are also reportedly achieved. This technique could be used
to control TRS emissions on those older existing furnaces or cross recovery
furnaces which do not have the combustion control capability for low TRS
emissions. This technique could prevent the need to replace or reduce the
load on older existing furnaces that are not capable of achieving the
necessary TRS regulations.
6.1.2 Digester and Multiple-Effect Evaporator Systems
The digesters and multiple-effect evaporators will be considered
together because non-condensable gases discharged from these two
sources are nerraally combined for treatment. At least half the mills are
incinerating the gases, to destory odors. Most commonly, the gases are
burned in the lime kiln. However, a few special gas-fired incinerators
are also used, either as backup for the kiln when it is shutdown, or
as the full time control deyi.ce.
Retrofitting an existing mill to handle and incinerate these
non-condensable gases is apparently no significant problem. Generally,
it is simply a matter of ducting the gases to the kiln or incinerator
and installing necessary condensers and gas holding equipment. The
6-9
-------
non-condensable gaaes. are added to the primary air to the kiln.
This retrofit situation has now been performed at over sixty mills.
The blow gases from batch digesters are generated in strong
bursts, that normally exceed the capacity of the lime kiln. For this
reason, special gas handling equipment has been developed to make
?n
the gas. flows more uniform. Adjustable volume gasholders, with
movable diaphragms or floating tops, receive the gas surges, and a
small steady stream is bled to the kiln. Although the non-condensable
gases form explosive mixtures in air, possible explosion hazards
have been minimized by- the development of appropriate gasholding
systems, flame arresters and rupture disks in the gasholding ducts,
and flame-out controls, at the lime kiln. Incineration of these
gases in existing process equipment such as the lime kiln is particularly
attractive since no additional fuel is required to achieve effective
emission control.
Scrubbers are used at a few existing mills. White liquor, the
usual scrubbing medium, is effective for removing hydrogen sulffrde and
21
methyl mercaptan, but not dimethyl sulfide or dimethyl disulfide.
At least 3 mills scrub the noncondensable gases before incineration to:
0] recover sulfur, (2) condense steam, and (3) remove turpentine
vapors and mist, thereby reducing the explosion hazards.
Combustion of noncondensable gases in a lime ki.ln or gas-fired
incinerator provides nearly complete destruction of TRS compounds.
During an EPA test (conducted for NSPS) on a separate incinerator
burning noncondensables from a digester system and a multiple effect
6-10
-------
evaporator system, the residual unburned TRS was less than 5 ppm
22
CQ.OI g/kg ADP) . The TRS test results (4-hour averages) of the
four tests conducted ranged between 0.5 and 3.Q ppm, and averaged
1.5 ppm Cdry gas basis}. During the tests, the Incinerator was
operating at 1000°F (measured! with, a calculated retention time for
the gases of at least Q.5 seconds.
Scrubber efficiencies are much lower than properly operated
Incinerators because only hydrogen sulfide and methyl mercaptan
react with the alkaline medium. The composition of noncondensible
gases is highly variable, but on the average hydrogen sulfide and
23
methyl mercaptan comprise about half the TRS compounds. Uncontrolled
emissions are ( 9,500 ppm] from the digester system and ( 6700 ppm)
24
from the multiple-effect evaporator system. Since caustic scrubbing
is only effective in controlling hydrogen sulfide and methyl mercaptan,
alkaline scrubber efficiencies are, therefore, roughly 50 percent.
TRS emissions from a scrubber are calculated to be about 0.63 g/kg ADP
(0.59 g/kg ADP from digester system and 0.04 g/kg ADP from multiple-
effect evaporator system! or abouf'7500 ppm.
6.1.3 Lime ICi.ln
TRS emissions, principally hydrogen sulfide, can originate from
two areas in the lime kiln installation, the lime kiln proper and a
scrubber that serves, as the particulate control device. TRS emissions
from the lime kiln installation are controlled by maintaining proper
process conditions. The most important parameters that were identified
in an industry (National Council of the Pulp and Paper Industry for
or
Air and Stream Improvement) study include the temperature at the
6-11
-------
cold end (.point of exhuast discharge) of the kiln, the oxygen
content of the gases leaving the kiln, the sulfide content of the
lime mud fed to the kiln, and the pH and sulfide content of the
water used in a parti.culate scrubber. If contaminated condensate is
used as the scrubbing medium, the exhaust gases could strip out the
dissolved TRS and increase the TRS emissions from the lime kiln
installation. Scrubbing the exhaust gases with a caustic solution
26
can reduce the TRS emissions from a lime kiln.
The amount of retrofitting necessary to achieve proper process
conditions depends on the design of the existing kiln installation.
If the existing kiln does, not achieve sufficient oxygen levels,
increased fan capacity or changes, to the scrubber system may be
necessary to increase the air flow through the kiln. Molecular
oxygen can also be used to replace a portion of the combustion air
to increase oxygen levels. Additional lime mud washing capacity may
also be necessary to reduce the sulfide content of the mud and
thereby reduce TRS emissions. This may require replacement of
existing centrifuges, with, more efficient vacuum drum filters, and
the addition of another mud washing stage. Furthermore, a mill
presently using condensate that contains dissolved reduced sulfur
compounds for a scrubbing medium would have to either install a
condensate stripper to remove the dissolved TRS prior to the scrubber
or replace the condensate with fresh water.
TRS emissions from existing lime kilns range from about 0.01 to
2.0 g/kg ADR (4 to 840 ppm), depending on the degree of control, with
6-12
-------
27
an average of about 0.4 g/kg ADP (168 ppm). EPA tests (.conducted
for NSPS development} on two lime kilns indicate that lime kiln TRS
emissions can be reduced to below 20 ppm (4-hour average) using process controls.
Another lime kiln using caustic scrubbing in addition to process
control is eapable, based on EPA results, of TRS emissions below 8
ppm.(4-hour average). When tested by EPA, all three lime kilns were burning
non-condensable gases from the digester system and multiple-effect
evaporator system.
It appears that existing lime kilns can be retrofitted to also
achieve low TRS emissions. TRS emissions from two existing 'lime kiln
installations have reportedly been reduced from over 100 ppm to less
than 20 ppm by modifying the lime mud washing systems and making
28
adjustments in the process operation, However, the TRS levels
to which existing kilns can he retrofitted depends on the load at
which the kiln is normally operated. If the kiln is operated
sufficiently over design capacity, it may be very difficult to obtain
the oxygen levels necessary for low TRS emissions (a&out 20 ppm).
Discussions mth the kraft industry indicate that TRS emissions from
these lime kilns can be reduced, however, to about 40 ppm.
6.1.4 Brown Stock Washer System
Nearly all existing kraft mills vent the brown stock washing
system gases directly to the atmosphere without control. However, at
least three mills in the. United States and Canelda, and several
in Sweden, utilize the gases as combustion air in a recovery furnace.
The furnace systems handling these gases are newer furnace systems which
6-13
-------
which were designed to burn the washer gases. No existing recovery furnace
(not designed for burning these gases) has yet been used to incinerate the
washer gases.
As discussed in section 6.1.2 the residual TRS after incineration is
very low, less than 5 ppm (0.01 g/kg ADP). Since the qas volume from the
? 29
washer drums is large, about 112 nr/Mq (150 CFM/TPD; the most likely
equipment for combustion is a recovery furnace or power boiler. The gases,
due to their large volume, would have to supplement the recovery furnace's
combustion air requirements. Even if the washers were enclosed with tiaht
hoods, the gas volume would be too large to burn in a lime kiln. The
actual gas volume handle at one mill is 75 m3/Mg (100 CFM/TPD). The gas
volume that would need to be handled at other existing mills can be hiqher
or lower depending upon tightness of hooding and degree of condensing.
The vent gases from the filtrate tank are considerably smaller in
o o/-\
volume, about 4.5 m /Mg (6 CFM/TPD). This stream is sufficiently small
for combustion in a lime kiln, or blended with the hood vent gas and
burned in a recovery furnace.
Incineration of the washer gases is a recovery furnace will not
affect furnace operation provided the moisture content of the qases
31
is not too qreat. High moisture content can increase aaseous
sulfur emissions and produce unsafe operating conditions. Red
(furnace) temperature decreases almost linearly with increased content
of vaporized water in the combustion air because of sensible heat
losses. With decreased bed temperatures, S02 emissions increase at a
rapid rate and reduced sulfur compounds become increasinqly difficult
6-14
-------
32
to control. Water entrained in the combustion qases can create
extremely dangerous conditions such as smelt-water explosions.
One furnace manufacturer recommends that the washer gases be
incinerated only in the secondary or tertiary air zones of the furnace.
This would keep the moist washer gases away from the smelt bed. Burninq
the gases only in the secondary or tertiary zones may affect the flexibility
of the recovery furnace, however, since the operator would not have the
33
ability to vary the air flow rate to each zone.
High moisture content would result in an increase in gas flow and
reduce the capacity of the recovery furnace.
An alternative to incineration of brown stock washer gases is
chemical scrubbing. White liquor (caustic) scrubbing, as previously
mentioned, is only effective in controlling hydrogen sulfide and methyl
mercaptan. However, the TRS emissions from a brown stock washer
34
system are principally dimethyl sulfide and dimethyl disulfide. A
more effective system is reportedly a chlorination-caustic scrubbing
system. In this system, the chlorine absorbs and oxidizes the dimethyl
sulfide and dimethyl disulfide. This technique was installed at one
mill in February 1976 and tests conducted at that time demonstrated TRS
emissions of less than 5 ppm. Another technique is chlorine gas injection.
This technique is used at one mill and tests conducted demonstrated TRS
emissions of less than 5 ppm and a control efficiency of 80 percent.37
6.1.5 Black Liquor Oxidation System,
The vent gases from nearly all existing black liquor oxidation (BLO)
systems are emitted directly to the atmosphere without control.
6-15
-------
One control technique is incineration. Incineration has proved
highly effective in controlling similar streams in some mills, for
example, the vent gases from pulp washing systems, the noncondensable
gases from digesters and multiple-effect evaporators, and vent gases from
condensate strippers. Similar to the pulp washing system, incineration in
the recovery furnace or power boiler is most likely, since the BLO gas
volume is usually too large to be handled by an existing kiln. This
would result in no significant fuel penalty.
Because of the high moisture content of the BLO gases, it would be
necessary to use condensers to reduce the moisture content before
burning, especially if the moist washer gases are burned in the same
furnace. Incineration of these moist qases in the furnace would probably
cause increased corrosion problems in the forced-draft fan ductwork and
the forced-draft fan itself. This would probably necessitate the replace-
ment of this equipment with corrosion-resistant equipment. A larger
forced-draft fan may be necessary to handle the increased mass flow due
38
to the high moisture content of the gases, even after using condensers.
The recovery furnace operation should not be adversely affected by
burning the BLO gases, even in combination with the washer gases,
provided the moisture content is sufficiently reduced and the gases are
39
burned high in the furnace. Since the BLO qases are deficient in oxygen,
one furnace manufacturer suggests burning them in the secondary or
tertiary air zone but states that the gases should still contain
sufficient oxygen to preclude adversely affecting the furnace operation.
6-16
-------
As mentioned in Section 6.1.4, the operational flexibility of the furnace is
reduced because a portion (BLO gases and washer gases) of the total combustion
air must always be introduced into the secondary and tertiary air zones and
cannot be used in the primary air zone when air in this zone is needed to
adjust furnace operation.
Emissions will be reduced to low levels if oxidation vent gases are
burned. Since these gases contain the same TRS compounds present in the
digester and multiple-effect evaporator off-qases which EPA tested after
incineration, TRS combustion residuals of the BLO vent gas will be less than
5 ppm.
A second control technique is the use of molecular oxygen in oxidation
systems instead of air. At least two mills in the United States now oxidize
black liquor by pumping oxygen directly into the black liquor lines. There
are no vent gases from this closed system. The economic feasibility of such
a system depends largely on the price and availability of oxygen.
Another technique is chlorine gas injection. This technique is used at
one mill on the vent gases from primary oxidation system. Tests conducted
demonstrated TRS emissions of less than 5 ppm and a control efficiency of
95 percent.
6.1.6 Smelt Dissolving Tank
Smelt dissolving tank TRS emissions are governed by process condi-
tions; that is; the presence of reduced sulfur compounds either in the
smelt or the water. The principal control option available is the
choice of water in the smelt dissolving tank or the particulate control
device. Clean water, low in dissolved sulfides, is preferable, although
6-17
-------
low emissions have been reported with nearly all process streams.*
If TRS emissions are high and no particulate control device (scrubber)
is used, a wet scrubber (e.g., packed tower) can be used to control the
TRS emissions. This scrubber would also result in controlling particulate
emissions. One mill reportedly reduced TRS emissions over 95 percent
from a level of about 0.19 g/kg of black liquor solids (BLS) (0.56 Ib/T ADP)
42
when a packed scrubber tower was installed.
TRS emissions from smelt dissolving tanks are normally low and average
about 0.007 g/kg BLS (0.02 Ib/T ADP).43 EPA tests on two smelt dissolving
tanks indicate TRS emissions below 0.0084 g/kg BLS (8 ppm).
These levels can be achieved on both new and existing smelt tanks.
Both these smelt tanks have wet scrubbers for controlling particulates.
Weak wash liquor was used as the scrubbing medium in both scrubbers.
6.1.7 Condensate Stripping System
In at least four United States mills, dissolved sulftdes and other
volatile compounds are stripped from the digester and evaporator conden-
sates prior to discharge to treatment ponds. One mill, which uses steam
as the stripping medium, discharges the gases from the stripper column
to a lime kiln. Two mills use air as the stripping medium. One of these
incinerate the stripper gases in a separate incinerator, while the other
incinerates the gases in the recovery furnace. One mill, which uses steam,
is presently scrubbing the stripper gases with white liquor, but this
44
technique is not as effective as incineration.
As mentioned in Section 6.1.2, incineration has proven to reduce
TRS levels from digester and multiple-effect evaporator systems to less
than 5 ppm. Since the vent gas from condensate strippers contains the
-------
same TRS compounds present tn the digester and multiple-effect
evaporator gases, TRS emissions in the condensate stripper gases after
incineration can be reduced to 5 ppm (0.01 g/kg ADP).
6.2 SUMMARY OF RETROFIT MODELS
Section 6.1.1 through Section 6.1.7 have examined the various
control techniques that can be applied to each source of TRS emissions and
have quantified the emission levels that can be achieved by applying
these controls. The economic and environmental impact of applying these
alternative control techniques will be discussed in Chapters 8 and 9,
respectively. In order to assess the impacts of applying controls
simultaneously to the various TRS sources in the entire mill, various
alternative control systems (retrofit models) were developed. The
alternative systems chosen range from controlling each TRS source to the
best achievable level, to controlling only the major TRS sources with
techniques less effective than best available technology. The six
retrofit models that use alternative control systems are listed in
Table 6-2. These six control systems were selected because the
differences between systems reflect major differences in the types
and costs of retrofits that would be carried out at an existing kraft
mill. The economic and environmental impacts of these retrofit models
will be analyzed in conjunction with the present controls already
installed at each existing kraft pulp mill. The six control systems
are discussed below.
Retrofit Model No. 1: All eight TRS sources are controlled to
the level of best available control technology. This system will
6-19
-------
Table 6-2. POSSIBLE CONTROL SYSTEMS FOR
EXISTING KRAFT PULP MILLS
Source
Recovery furnace
Digester system
Multiple-effect
evaporator system
Lime kiln
Brown stock
washer system
Black liquor
oxidation system
Smelt dissolving
tank
Condensate
stripper system
No. 1
Process control +
BLO/5 ppm
Incineration/5 ppm
Incineration/5 ppm
Process controls
+ caustic scrub-
bing/8 ppm
Incineration/5ppm
Incineration/5 ppm
Fresh water/8 ppm
(.0084 g/kg BLS)
Incineration/5 ppm
No. 2
Process control +
BLO/5 ppm
Incineration/5 ppm
Incineration/5 ppm
Process controls
+ caustic scrub-
bing/8 ppm
No control
No control
Fresh water/8 ppm
Incineration/5 ppm
Control Systems/emission level (ppm)
No. 3 No. 4
Process control +
BLO/20 ppm*
Incineration/5 ppm
Incineration/5 ppm
Process controls
+ caustic scruh-
bing/8 pom
No control
No control
Fresh water/8 ppm
Incineration/5 ppm
Process control +
BLO/20 ppm*
Incineration/5 ppm
Incineration/5 ppm
Process controls/
20 ppm
No control
No control
Fresh wat^r/S ppm
Incineration/5 ppm
No. 5
Process control +
BLO/20 ppm*
Incineration/5 ppm
Incineration/5 ppm
Process controls/
40 ppm
No control
No control
Fresh water/8 ppm
Incineration/5 ppm
No. 6
Process control +
BLO/5 ppm
Incineration/5 ppm
Incineration/5 ppm
Process controls/
40 ppm
No control
No control
Fresh water/8 ppm
Incineration/5 ppm
*The 20 ppm levels applies to old design furnaces; new design furnaces can achieve 5 ppm with application of the same control technology
(two-stage BLO).
-------
result in the lowest TRS emissions from a kraft pulp mill and will
require installation of a new furnace if the existing furnace is
relatively old and cannot achieve 5 ppm TRS. In most cases, this
system will also require caustic addition to the existing scrubber
and the improvement or replacement of the lime mud washing facility in
order to achieve a TRS level of 8 ppm from the lime kiln. TRS emissions
from the smelt dissolving tank will be controlled by using fresh water
in the tank and the particulate control device. Incineration will be
used to control TRS emissions from the digester system, multiple-effect
evaporator system, brown stock washer system, black liquor oxidation
system and condensate stripping system. The gases from the digesters,
multiple-effect evaporators and condensate strippers will be incinerated
in the lime kilns. The gases from the washers and oxidation system
for purposes of impact analysis are assumed to be burned in a separate
incinerator since no existing recovery furnace has been modified to
handle these gases.
Retrofit Model No. 2: This control system is similar to
Retrofit Model No. 1 except that the vent gases from the washer system
and the BLO system are not incinerated. This system was chosen as
a model because, based on the economic analysis performed for NSPS
development, these two smaller TRS emission sources are less cost effective
to control than the other sources. It is assumed that these two sources
would be combined for treatment. The cost of controlling one source is
related to the cost of controlling the other source because one
incinerator would be installed to handle both gas streams. Therefore,
modifications made to an existing mill would generally be the same
whether both or only one of these sources is controlled.
6-21
-------
Retrofit Model No. 3: This control system is similar to
Retrofit Model No. 2 except that less effective control of a higher
TRS level from the recovery furnace is allowed. This system was
chosen because the higher TRS level (20 ppm) will allow mainly older
recovery furnaces (mostly those built before 1965) to remain in operation. The 5
ppm level as required in Retrofit Models No. 1 and No. 2 would probably
require these older furnaces to be replaced. The cost impact of
Models No. 1 and 2 will be substantially greater than that of Model
No. 3 if furnace replacement is necessary.
Retrofit Model No. 4: This control system is similar to retrofit
Model No. 3 except that the lime kiln TRS level has been relaxed from
« to 20 ppm. This system would permit many kilns to achieve this
level without using caustic scrubbing. The TRS emissions from the lime
kiln would be controlled by process controls and require the lime mud
washing facility to be improved or replaced. Caustic scrubbing would
be a major expense if the caustic cannot be used in the pulping process
or a wet scrubber is not already used for particulate control.
Retrofit Model No. 5: This control system is similar to
Retrofit Model No. 4 except that the lime kiln TRS level has been
relaxed from 20 to 40 ppm. This system would permit many kilns to
achieve the level without modifying the lime mud washing installation.
TRS emissions from the lime kilns would be controlled by using process
controls on the kiln itself. Modifications to the mud washing system
are a major expense (see Chapter 8) in controlling TRS emissions from an
existing kraft pulp mill.
6-22
-------
Retrofit Model No. 6: This system is similar to Retrofit Model
No. 5 except that all recovery furnaces are controlled to 5 ppm TRS
rather than 20 ppm. This system was chosen to determine the differences
on the impacts in relaxing controls on the lime kiln (Model No. 5) in
comparison to relaxing controls on the recovery furnace (Model No. 3).
6.3 INSTALLATION AND START-UP TIME
The amount of time necessary to retrofit an existing kraft mill
depends on what TRS sources are to be controlled and what technologies
are to be used. It should also be pointed out that actual time require-
ments to implement a given control technology can vary widely depending
upon such factors as space limitations, weather conditions, lack of
j
available utilities, delays in equipment delivery, and time required to
develop engineering data.
Table 6-3 presents estimates of the normal length of time required
to retrofit the various sources in order to bring them into compliance.
Table 6-3 shows that the time necessary for initial design and approval
can vary from 6 months to 3 years, depending on the source and the
complexity of retrofitting that source. This time period includes:
a. Engineering design of the overall project;
b. Project fund approval;
c. Control agency approval;
d. Order placement.
Table 6-3 also presents estimates of the amount of time required
for installation of the necessary equipment. This time is for the
6-23
-------
Table 6-3. DESIGN AND INSTALLATION TIMES*
. ... . - ^_- - - -
TRS source
Recovery furnace
Digester system
Multiple-effect
evaporator system
Lime kiln
Brown stock
Design and
approval
(months)
18-36
6
6
6-24
6-24
-- . -
Installation
(months)
12-36
18
18
24
12-15
washer system
Black liquor 6 12
oxidation system
Smelt dissolving 6 18
tank
Condensate stripping 6 18
system
*Based on discussions with various companies and manufacturers. The actual times
of aooroval and installation may over-lao to some extent.
6-24
-------
period from the order date to start-up. This installation time varies
from about one year for installing black liquor oxidation system to
three years for installing a new recovery furnace.
6-25
-------
REFERENCES FOR CHAPTER 6
1. Factors Affecting Reduced Sulfur Emissions from the Kraft Recovery
Furnace and Direct Contact Evaporator, NCASI Technical Bulletin No. 44,
December 1969.
2. JheEffect of Combustion Variables on the Release of Odorous Sulfur
Compounds from a Kraft Recovery Furnace. Thoen, G. N., De Haas, G. G.,
Tallent, R. G., and Davis, A. S., TAPPI, 51(8):329-333, August 1968.
3. Op. cit., Reference 1, page 28.
4. Op. cit., Reference 1, page 31.
5. Meeting Report - Babcock & Wilcox Company and EPA, Durham, NC,
May 1, 1975.
6. Op. cit., Reference 4.
7. Op. cit., Reference 5.
8. Op. cit., Reference 5.
9. Improved Air Pollution Control for a Kraft Recovery Boiler: Modified
Recovery Boiler No. 3. Henning, K., Anderson, W., and Ryan, J., EPA
Report No. 650/2-74-071-a, August 1974.
10. Survey of Current Black Liquor Oxidation Practices in the Kraft
Industry, NCASI Technical Bulletin No. 39, December 1968.
11. Op. cit., Reference 10, Table 6.
12. Standards Fupport and Environmental Impact Document, Volume 1:
Proposedjjandjrds_pf_Performance for Kraft Pulp Mills, Environmental
Protection Agency, September 1976.
13. Letter from J. W. Kisner of Babcock & Wilcox Company to James
Eddinger of EPA, dated May 27, 1975.
14. Presentation given by Julius Gommi of Combustion Engineering at the
NAPCTAC meeting in Raleigh, North Carolina, on March 3, 1'977.
6-26
-------
15. Op. cit., Reference 13.
16. Monthly Reports obtained from the Washington Department of
Ecology and the Oregon Department of Environmental Quality.
17- A Report on the Study of TRS Emissions from a NSSC-Kraft Recovery Boiler,
Container Corporation of America, March 9, 1977.
18. Op. cit., Reference 17, page 2.
Ecology and the Oregon Department of Environmental Quality.
19. Considerations in the Design for TRS and Parti oil ate Recovery from
Effluents of Kraft Recovery Furnaces, Teller, A. J., and Amberg, H. R.,
Preprint, TAPPI Environmental Conference, May 1975.
20. Current Practices in Thermal Oxidation of Noncpndensable Gases jn the
Kraft Industry. NCASI Technical Bulletin No. 34, November 1967.
21. Atmospheric Emissions from the Pulp and Paper Manufacturing Industry,
EPA-450/1-73-002, September 1973. (Also published by NCASI as Technical
Bulleton No. 69, February 1974.)
22. Malodorous Reduced Sulfur Emissions from Incineration of Noncondensable
Off-Gases. EPA Test Report No. 73-KPM-1A, 1973.
23. Op. cit., Reference 21, Tables 3, 5, and 7.
24. Op. cit., Reference 21, Tables 3, 5, and 7.
25. Suggested Procedures for the Conduct of Lime Kiln Studies to Define
Emissions of Reduced Sulfur Through Control of Kiln and Scrubber Operating
Variables, NCASI Special Report No. 70-71, January 1971.
26. Kraft Odor Control at Mead Papers, Ayers, K. C., Clutter, L. W., and
Adams, A. B., TAPPI, September 1974.
27. Op. cit., Reference 21, Table A-5.
28. Trip Report on Visit to the Champion International Mill in Pasadena,
Texas, on April 4, 1975.
6-27
-------
29. Factors Affecting Emission of Odorous Reduced Sulfur Compounds
from Mi'scellanepus Kraft Process Sources, NCASI Technical Bulletin
No. 60, March 1972.
30. Op. cit., Reference 29.
31. Op. cit., Reference 13.
.32. Letter from S. T. Potterton of Babcock & Mil cox Company to James
Eddinger of EPA, dated July 23, 1974.
33. Op. cit., Reference 5.
34. Op. cit., Reference 29, page 6.
35. Letter from H. M. Patterson of Oregon Department of Environmental
Quality to James Eddinger of EPA, dated April 4, 1975; and letter from
L. A. Broeren of Crown Zellerbach Corporation to Doug Ober of Oregon
Department of Environmental Quality, dated January 10. 1975.
36. Telephone conversation between Chuck Clinton of Oregon Department
of Environmental Quality and James Eddinger of EPA on December 6, 1976.
37. Monthly Reports obtained from the Humboldt County Air Pollution
Control District.
38. Letter from Russell Blosser of NCASI to FT*;! Boys of EPA dated
November 17, 1972.
39. Op. cit., Reference 19.
40. Op. cit., Reference 37.
41. Op. cit., Reference 3, Table 25.
42. Telephone conversation between Richard Labrecque of S. D. Warren
Company and James Eddinger of EPA on July 19, 1973.
43. Op. cit., Reference 41.
44. Private communication between Arthur Plummer of Chesapeake Corpor-
ation and James Eddinger of EPA on February 13, 1975.
6-28
-------
CHAPTER 7. EMISSION MONITORING AND COMPLIANCE
TESTING TECHNIQUES AND COSTS
This chapter discusses the various monitoring and compliance testing
methods that have or could be used in the kraft pulp industry, and also
discusses the rationale leading to the selection of the reference test
method used for the TRS source tests conducted during the SPNSS development
program.
7.1 EMISSION MEASUREMENT TECHNIQUES
7.1.1 Emission Monitoring
Performance specifications for oxygen continuous monitors have already
been published in 40 CFR, Part 60, Appendix B, Performance Specification Three,
but it has not been demonstrated that these monitors will perform in the same
manner when used at a kraft pulp mill. There is, however, no technical reason
to believe that they will not be able to meet these requirements. A number of
commercially available instruments are capable of meeting the performance
specifications. The cost of one of these instruments, installed, is in the
range of $9,000 to $11,000.
Equipment is also commercially available for temperature monitoring. This
can be accomplished using a thermocouple, electronic cold junction, and a
millivolt strip-chart recorder. A system such as this could be purchased for
less than $2,000. Instrumentation is also available for continuously monitoring
the pressure loss of the gas stream through the scrubber and for monitoring the
scrubbing liquid supply pressure to the scrubber.
7-1
-------
At present, there are several types of instruments that have been used
to successfully measure TRS on a short-term basis. However, some questions
remain about their reliability for data gathering on a continuous basis. The
GC technique described in Method 16 was not designed to be used as a continuous
monitor and its suitability for this purpose has not yet been evaluated.
There are other systems which have been used on a long-term basis but necessary
maintenance and quality control procedures to insure that these monitors are
operating properly are still being developed. Work is presently underway in
cooperation with the National Council for Air and Stream Improvement to evaluate
a number of different types of systems for their suitability as continuous
monitors. Ultimately, this should result in published performance specifications
for TRS monitors.
7.1.2 Compliance Testing
7.1.2.1 TRS Compounds - The need for an effective test method for measurement
of reduced sulfur emissions from stationary sources resulted from a new source
performance standard NSPS program to establish performance standards for a
variety of kraft mill unit processes with respect to malodorous emissions. As
with previous NSPS programs, test methodology was needed to gather: (a) accurate
data which would demonstrate emission limitations attainable through the use
of best available emission control systems; and (b) enough sampling and analytical
data such that a reference method for performance testing could be prescribed.
At the inception of the NSPS kraft mill program in January 1972, a survey
was made to evaluate existing test methods for potential use. This survey
included a review of the literature, contact with mill personnel, and review of
previous research and evaluation of analytical techniques by the Environmental
Protection Agency (EPA). Since the degree to which methods are available for
field use in odor measurements is directly related to the complexity of the
odorant mixture to be measured, it was fortunate that the nature of emissions
7-2
-------
from kraft pulping operations had been well-defined. Emissions consist
primarily of sulfur dioxide (S02) and four reduced sulfur compounds -
hydrogen sulfide (H2S), methyl mercaptan (CH-SH), dimethyl sulfide (DMS),
and dimethyl disulfide (DMDS). These compounds are highly reactive,
particularly the H2S-S02 mixture which may form elemental sulfur, and are
present in low concentrations in well-controlled sources. In addition, the
sources of these emissions (recovery furnaces, lime kilns, smelt dissolving
tanks, digesters, multiple-effect evaporators, washer systems, oxidation
systems, and condensate strippers) are characterized by high temperatures and
moist, parti oilate-laden effluent streams.
After careful consideration, it was determined that an additive total
reduced sulfur (TRS) standard, reflecting all sulfur compounds present minus
SO-j was desired. Considering this and the previously mentioned source condi-
tions, a field method which could measure reduced sulfur compounds, either indi-
vidually or collectively, was sought.
7.1.2.1.1 Methods surveyed - A review of the literature revealed that
analytical methods fell into four main categories: colorimetry, direct
spectrophotometry, coulometry, and gas chromatography. Although most of the
methods surveyed were developed for measurement of ambient concentrations, this
did not preclude their possible application to the measurement of stack emissions.
Colorimetry - A sample is bubbled through a solution which
selectively absorbs the component or components desired. The absorbed compound
is then reacted with specific reagent to form a characteristic color which is
measured spectrophotometrically.
An example of a colorimetric method is the methylene blue
method which involves the absorption of TRS compounds in an alkaline suspension
7-3
-------
of cadmium hydroxide to form a cadmium sulfide precipitate. The precipitate
is then reacted with a strongly acidic solution of N, N, dimethyl-P-phenylene-
diamine and ferric chloride to give methylene blue, which is measured spectro-
photometrically. Automated sampling and analytical trains using sequential
techniques are available for this procedure. Inherent deficiencies for stack
sampling applications include variable collection efficiency, range limitations,
and interferences from oxidants.
Another colorimetric method is the use of paper tape samplers
impregnated with either lead acetate or cadmium hydroxide. These compounds
react specifically with H?S and the resultant colored compound can be measured
directly with a densitometer. Tape samplers would not be appropriate for all
TRS compounds unless they were all reduced quantitatively to H^S. In addition,
the range is limited and the method suffers from light sensitivity, fading, the
necessity for precise humidity control, and variability in tape response.
Spectrophotometry - The use of infrared and mass spectro-
photometry and other sophisticated spectroscopic methods for analysis of
individual odorants is well established. However, these methods were considered
expensive, time consuming, and not suitable for routine field applications.
One promising method in this area was split-beam ultraviolet
spectrophotometry, which utilizes the strong absorption of ultraviolet radiation
at 282 nm by S02. In this method, the gas sample is mixed with air, filtered,
and split into two streams. One stream passes through a catalytic oxidation
furnace where sulfur constituents are oxidized to SO,, and then through an
optical cell where its absorbance is measured. The second stream passes through
a dummy furnace and then into a reference optical cell. The difference in
abosrbance values between the two cells is a measure of the non-SOp sulfur
7-4
-------
constituents in the sample stream. The system is capable of S02/TRS concentrations
in the range of 10 to 2500 ppm. Since well-controlled kraft mill sources fall
below the minimum range of 10 ppm, this method was considered not applicable.
Co ul pine try - Coulometric titration is based on the principle
of electrolytically generating a selected titrant in a titration cell. The
titrant may be a free halogen (bromine or iodine) in aqueous solution as an
oxidizing agent, or a metal ion (silver), as a reducing agent. The electrolytic
current required to generate the titrant, as it is consumed, is a linear
measure of the concentration of reactive compounds in the gas sample.
Gas Chromatography - This system is based on the ability of
the gas chromatographic columns to separate individual sulfur compounds, which
are then determined individually by various analytical techniques. The most
sensitive determination is the flame photometric detector (FPD). This technique
involves measurement of light emitted from the excited S02 species formed when
a sulfur compound is burned in a hydrogen-rich flame.
7.1.2.1.2 Methods used for data gathering -
Analytical Techni ques - Based on the survey, the 6C/FPD
technique was considered to be the most promising and was selected for field
evaluation. At several of the plants, the coulometric titrator was also tried
since this instrument was widely used by the industry at the time.
Sample Collection - Considering the sulfur compound reactivity,
high moisture, and presence of particulate matter, EPA developed a special
sample handling system. It utilizes a sampling probe enclosed in a stainless
steel sheath with inlet ports perpendicular to the stack wall. A deflector
7-5
-------
shield is fixed on the underside to deflect the heavier particles while the
proble is packed with glass wool to trap finer particles. Teflon tubing
heated to 250°F is used to carry the sample from the probe to a dilution
system where the sample is routinely diluted 1:9 with clean dry air. The
heated sample line prevents condensation and teflon does not react with sulfur
compounds. After the sample is diluted in a heated dilution box, its moisture
content is reduced so that the dew point is below ambient temperature, preventing
condensation and sample loss during analysis.
Calibration of Instruments - For delivery to and calibration
of analytical instruments, a special system containing premeation tubes with
appropriate concentrations of S02, H2$, DMS, DMDS, and CHgSH were installed
into the sampling and analytical system. These gas permeation tube standards
were developed by EPA personnel specifically for use with GC systems.
Field Evaluation - Since 1972, EPA has used the sample
delivery system, dilution system, calibration system, and the GC/FPD methods
at a number of kraft mills. Two separate GC/FPD systems were employed to
facilitate the rapid analysis of both high and low molecular weight sulfur
compounds. One system resolved HLS, S0_, CFLSH, and DMS, while the other
simultaneously resolved DMDS and other high molecular weight homologs. To
ensure reliability of the data, the GC/FPD systems were frequently calibrated
with standards of each of the sulfur compounds.
Field experience has shown that the GC/FPD method is the
most reliable, sensitive, and precise for determination of TRS. This has
also been substantiated via verbal communications with industry experts.
There, may, however, be some loss of precision in asing this method on
sources having high levels of S0? in comparison to the level of TRS.
Further developmental work is underway to eliminate this problem and the
necessary changes will be made in Method 16 as soon as the work is completed.
7-6
-------
Conversely, at six of these kraft mills,' two different
coulometric instruments have yielded poor results, possibly due to the low
concentrations encountered, and the operational problems mentioned earlier.
This instrument is unacceptable for compliance testing.
7.1.2.1.1 Compliance method - As the result of field experience of
testing TRS compounds at kraft mills, Method 16 was prepared for determining
compliance with new source performance standards. This method requires
use of a GC/FPD system using the same measuring principle as used for
the data gathering process. Design specifications for the
required dilution system, calibration techniques, and instrumentation that
was considered necessary to insure accuracy, precision, and reliability
are specified.
7-7
-------
8. COST ANALYSIS OF ALTERNATIVE EMISSION CONTROL SYSTEMS
8.1 Introduction
The purpose of this chapter is to develop estimates of retrofit costs
for alternative emission control systems for reduction of total reduced
sulfur (TRS) emissions at existing kraft pulp mills. Capital and annualized
costs will be developed for alternative controls on each affected facility
on three sizes of kraft pulp mills: 500, 1000, and 1500 ton per day mills.
Following this aggregate costs will be presented for six alternative emission
control systems on a 1000 TPD kraft pulp mill. Finally, a summary of an
analysis for estimating industry-wide costs will be presented.
The determination of incremental costs for the various control al-
ternatives over state regulations for kraft mill sources is a critical
element in this analysis. Some 28 states with kraft mills have regulations
which vary widely in their effectiveness of reducing TRS emissions. The
variability in regulations from state to state has been taken into account
in the determination of total industry costs.
Throughout this chapter the terms capital cost and annualized cost
are used; therefore, a brief definition is in order. The capital cost
includes all the items necessary to design, purchase, and retrofit either
a control device or process equipment necessary to achieve the emission
reduction. The capital cost includes the purchase of all factory assembled
equipment, such as a recovery furnace boiler, black liquor oxidation system
components, incinerator, and so forth; ancillary items, such as fans, instru-
mentation, pumps; equipment installation cost including demolition, site
clearance, piping, wiring, and the cost of engineering, construction
overhead, and contingencies. Capital costs are reported in third quarter
8-1
-------
1976 dollars. The annualized cost of a retrofit project, whether it be a
control device or replacement of process equipment, is a measure of what
it costs the company to own and operate that system. The annualized cost
includes direct operating costs such as labor, utilities and maintenance;
and capital related charges such as depreciation, interest, administrative
overhead, property taxes, and insurance. The actual costs experienced by
individual mills may vary considerably and often are difficult to collect.
Nevertheless, attempts must be made to determine reasonable estimates of
these costs. The following values were chosen as typical and should pro-
vide a reasonable estimate of the annualized costs of the retrofit control
requirements. Operating labor is charged at a rate of $6 per hour with
supervision at $8 per hour. Electricity is assumed to cost 2,5 cents per
kilowatt-hour. Fuel costs are assessed at $2.00 per million BTU., For
purposes of estimating annualized costs, 328.5 operating days per year
were assumed.
Recovery famaces and lime kilns that would constitute as replacements
necessary to achieve TRS reduction are not allocated any charges for main-
tenance and repair. The reason for this is that such maintenance costs
incurred for replacement would be offset by maintenance charges foregone
in scrapping the old equipment.
For gas collection and piping systems, including incinerators, main-
tenance costs are assessed at a 2 percent of capital investment. For
black liquor oxidation systems and oxygen plants, a charge of 4 percent of
capital investment is expensed.
8-2
-------
Capital charges have been calculated on the basis of 100 percent
debt financing and recovery of capital by uniform periodic payments
(capital recovery factor). Rate of interest for institutional lending
is assumed to be 10 percent. The economic life assumed for all equipment
is 15 years. Property taxes and insurance are assessed at a rate of 2
percent. Administrative overhead costs involving records keeping,
monitoring, etc. are also assessed at a rate of 2 percent.
8.2 Methodology
The affected facilities and control systems were previously discussed
in Chapter 6. The retrofit control techniques that can achieve the necessary
emission reductions, as outlined in Table 6-2, are summarized in Table 8-1
for each affected facility and control system. These control techniques
were based on certain assumptions, which reflect the content of information
available for estimating retrofit costs. These assumptions will be dis-
cussed in detail in the following subsections for each affected facility.
The cost information on retrofit controls for recovery furnaces presented
in this chapter applies to cross-recovery furnaces as well as straight kraft
recovery systems. However, there is one precaution that should be noted.
The application of the control techniques for straight kraft recovery furnaces
as discussed here and in Chapter 6 will not necessarily result in the same
achievable emission levels for cross-recovery furnaces as reported in this
chapter.
For cost estimating purposes, a list of mill characteristics was
compiled for each kraft pulp mill in the United States. The information
compiled includes Kraft production capacity; number, capacity, manufacturer,
basis design (direct or indirect contact) and year of manufacture of the
recovery furnace; the number and production rate for lime kilns; the number,
8-3
-------
Table 8-1. SUMMARY OF RETROFIT CONTROL TECHNIQUES FOR ALTERNATIVE CONTROL SYSTEMS ON EXISTING KRAFT MILLS
SOURCE
Recovery Furnace
Digester System
Multiple Effect Evaporators
Lime Kiln
Brown Stock Washer System
Black Liquor Oxidation
System Vents
Smelt Dissolving Tank
Condensate Stripping
System
1
a) Replace Furnaces
>10 yrs. of age
b) Add 2nd Stage Black
Liquor Oxidation
(Including furnaces
<10 yrs. of age). Also
improve furnace air
distribution
Incineration
Incineration
b)
Increase Lime Mud
Washing Capacity
Increase fan cap.
& Monitor Oxygen
and temp, (kiln)
c) Ada caustic to
kiln scrubber
Incineration
Molecular Oxygen
Substitute fresh
water for condensate
Incineration
CONTROL SYSTEMS
Same as a) Replace Furnaces
1 >20 yrs. of age
b) Add 2nd Stage Black
liquor oxidation for
all other furnaces.
Also improve furnace
air distribution.
Incineration Incineration
Incineration Incineration
Same as Same as
1 1
Same as
3
Same as
3
Incineration Incineration
Incineration
Incineration
Incineration
Incineration
a) Increase Lime
Mud Washing
Capacity
b) Increase Fan
Cap. & Monitor
oxygen & temp(kiln)
a) Increase Fan
Cap. & Monitor
Oxygen and
temp. (Kiln)
Same as
1
Incineration
Incineration
Same as
5
No Control
No Control
Same as
1
No Control
No Control
Same as
1
No Control
No Control
Same as
1
No Control
No Control
Same as
1
"No Control
No Control
Same as
1
Incineration
Incineration
Incineration
-------
type, and TRS controls for multiple effect evaporators and digesters;
and the number of brown stock washing systems and number of washing stages
per system. Mill capacity and other data were compiled from Posts' and
Lockwoods' 2 directories and updated wherever possible from contacts with
pulp and paper companies.
The approach used to estimate retrofit control costs is as follows.
The National Council for Air and Stream Improvement (NCASI) was called
upon to provide EPA both the contacts within individual paper companies
for sources of cost data and the technical parameters, or guidelines/3'
for estimating costs. These guidelines were also, made available to Industry
personnel for estimating costs. The following 8 companies were selected
and contacted to provide maximum coverage in terms of mills and capacity:
(a) International Paper
(b) Weyerhaeuser
(c) Georgia Pacific
(d) Westvaco
(e) Boise Cascade
(f) St. Regis
(g) Western Kraft (Willamette Industries)
(h) Mead Corporation
Actual corporate data covering both costs and mill characteristics were
provided for 42 mills by these companies. This coverage constitutes 35
percent of the number of total mills and about 41 percent of total U.S.
capacity. Two boiler manufacturers, who have built all the kraft recovery
furnaces in the U.S.,were contacted to provide information on ages of
existing furnaces, as well as the cost of new furnaces. In-house informa-
tion was used to generate cost estimates, which were then used as a forum
for discussion during visits with the corporate staff of the eight paper
companies. The time for the data gathering phase was the second quarter
of 1975.
8-5
-------
The data received from the paper companies were analyzed for regression
possibilities with mill characteristics. The next step was to develop model
plant costs for the following model mills: 500 tons per day, 1000 tons per
day, and 1500 tons per day production capacity. These model plant costs in
combination with the information on mill characteristics and furnace age
were used to estimate costs for the remaining 65 percent of the mills in
the U.S.
8.3 Costs for Affected Facilities
8,3.1 Recovery Furnaces
The major problem from the standpoint of TRS emission in existing
kraft mills is associated with the burning of black liquor in direct
contact recovery furnaces. The methods used to reduce these emissions are:
(1) close monitoring and control of the process variables in the recovery
furnace and, (2) oxidizing the black liquor to reduce the sulfides content
before evaporation of the black liquor in the direct contact evaporator.
Control of the process variables depends significantly upon the original
design and configuration of the recovery furnace. Hence, furnace age is a
critical factor in determining the extent of retrofit costs.
Recovery furnaces, including both direct contact and indirect contact,
of recent vintagethose built since 1965--are capable of achieving a 5 ppm
level of TRS without any additional costs for process controls, as indicated
in Chapter 6. Many of those process controls discussed in Chapter 6, such
as flexibility in air distribution and the membrane insulation between wall
tubes, are inherently designed in furnaces constructed since 1965. Mills
with a direct contact furnace of the design just mentioned are assumed to
incur costs only for a second stage of black liquor oxidation as the
8-6
-------
requirement for achieving a 5 ppm level of TRS from the total recovery
furnace affected facility. Mills with indirect contact furnaces are
assumed to achieve the 5 ppm without any additional costs. Note that the
indirect contact furnace design concept has been commercially available
only since 1967.
Recovery furnaces built prior to 1965 are assumed to be incapable of
achieving a 5 ppm level of TRS despite any attempts to achieve the most
efficient black liquor oxidation system. The key element in the assumption
is related to the earlier design of the furnace itself, which was not con-
ducive toward reduction of TRS to low levels. Therefore, the control
strategy for a 5 ppm level would require the replacement of the recovery
furnace, which constitutes as the retrofit cost. For furnaces built
between 1955 and 1965, the 20 ppm level is assumed to be achievable, according
to NCASI guidelines and discussion in Chapter 6. The only costs incurred
are those associated with adding a second stage of black liquor oxidation.
For furnaces built prior to 1955, the retrofit costs assumed include re-
placement of the furnace plus a complete 2-stage black liquor oxidation for
both the 5 ppm and the 20 ppm control levels. This assumption is in accord
with NCASI guidelines.
Another important factor besides age that would affect costs is the
need for additional black liquor burning capacity. In particular, mills
that have not added a recovery furnace since 1965 may be overloading fur-
naces with black liquor from a design standpoint. Such a practice leads
to high TRS emissions. Mills that have furnaces built between 1955 and
1965 and appear to have underrated furnace capability, according to
available data on mill characteristics, are assumed to add additional
8-7
-------
black liquor burning capacity in order to achieve the 20 ppm strategy.
Therefore, such mills would have to purchase a new furnace in order to
achieve the 20 ppm level.
Additional guidelines were developed by NCASI to provide these com-
panies technical parameters for black liquor oxidation as a complement to
process controls in the recovery furnace toward achieving the 5 ppm and
the 20 ppm levels.
The guidelines used for estimating costs of black liquor oxidation
requirements are stated as follows. The oxidation system representative
of best technology consists of: (a) two-stage oxidation with at least one
stage of oxidation, (b) retention time of five hours for both stages com-
bined, (c) stand-by blower capacity, and (d) monitoring of both oxidizing
air and liquor flow rates. The NCASI provided these guidelines to each
company, which in turn was to assess its requirements relative to their
existing oxidation system capabilities.
The costs presented in Table 8-2 for model plants were analytically
derived from the results received for some 42 mills. Cost data for recovery
furnaces in 17 mills, which management felt may be required, correlated
reasonably well with mill size as a parameter. Conversely, data received
from various companies for oxidation requirements as to the condition of
existing oxidation systems, sulfidity of black liquor, and level of ex-
perience acquired with development of highly efficient oxidation systems
did not correlate well with mill size. This result occurred despite the
specific guidelines set forth by NCASI to the managerial staff of the 8
companies. A probable explanation is that the nature of mill operations
and mill lay-out vary widely, which affect costs significantly.
-------
\
\
Table 8-2. RETROFIT CONTROL COSTS FOR RECOVERY FURNACES
A. Furnace Replacement
CD
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
13,400,000
2,270,000
13.82
1000
23,300,000
4,000,000
12,18
1500
32,200,000
5,470,000
11.09
00
VO
B. Second Stage Black Liquor Oxidation
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
600,000
153,000
0.93
1000
1,000,000
242,000
0.74
1500
1 ,500,000
360,000
0.73
(1) Control strategy required to achieve 5 ppm for mills whose furnaces were built prior to 1965. Costs
include secondary black liquor oxidation.
(2) Control strategy required to achieve 20 ppm for furnaces built since 1955. Pre-1955 furnaces are
assumed unable to achieve 20 ppm; require replacement.
-------
8.3.2 Digesters and Multiple Effect Evaporators
The control technique for achieving a 5 ppm level is incineration of
the noncondensible TRS compounds emitted from various vents. The nature
of the retrofit project involves the collection of the noncondensibles and
piping them to the incineration point, as well as necessary upgrading of
the blow heat recovery system associated with the digesters. Normally, the
incineration point is the lime kiln; however, in some situations, a separate
incinerator may be used as the back-up for the lime kiln or as the regular
control device.
Data gathered from the selected paper companies did not reveal any
significant correlation of cost with mill size. The probable explanation
of this are the number of digesters, their spatial arrangement, condition
?f blow heat recovery system, and type of gas holding system. Despite the
weak correlation of costs with capacity estimates were made for model plants
to represent three conditions: (1) batch digesters with extensive piping
requirements only (low retrofit penalty); C2) batch digesters with extensive
piping, refurbished blow heat recovery, and separate incineration; Chigh
retrofit penalty) and (3) continuous digesters. Multiple effect evaporator
vents are assumed to be combined with the digester vents for all three cases.
The costs for incineration are presented in Table 8-3. Utility requirements
for electricity and fuel were developed from data by an engineering con-
(c\
struction company. ' Fuel consumption assumed in Table 8-3 for destruction of
noncondensible TRS emissions are based on use of a separate incinerator as
a stand-by control device for 33 days, or ten percent of the time. This
represents an assumed duration of downtime for the lime kiln for maintenance
when the kiln becomes unavailable as a control device for incinerating TRS
emissions.
8-10
-------
Table 8-3. RETROFIT CONTROL COSTS FOR DIGESTERS AND MULTIPLE EFFECT EVAPORATORS
A. Batch Digesters - Low Retrofit Penalty'1'
Mill Size, TPD
Capital Costs ($)
Annuali zed Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
400,000
96,000
0.58
1000
900,000
210,000
0.64
1500
1,600,000
360,000
0.73
B. Batch Digesters - High Retrofit Penalty
(2)
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr. )
Annual ized Costs per Ton ($/T)
500
900,000
207,000
1.26
1000
2,000,000
452,000
1.38
1500
3,500,000
773,000
1.57
CO
C. Continuous Digesters
Mill Size
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
300,000
67,000
0.41
1000
400,000
96,000
0.29
1500
500,000
125,000
0.25
(1) Includes provision for extensive piping only.
(2) Includes blow heat recovery system, extensive piping, separate incinerator.
-------
8.3.3 Brown Stock Washers
The control technique for achieving the 5 ppm level of TRS emissions
from washer vents is incineration. The incineration may be carried out
either in the recovery furnace or a separate incinerator. As pointed out
in Chapter 6, incineration in an existing recovery furnace may restrict
the operation of the recovery furnace. The gas flows from the washer
system are too great for the lime kiln to incinerate. For some mills, a
separate incinerator may be the only available control device to treat the
washer vent gases.
Preliminary EPA costs were developed in-house on the basis of ventila-
tion requirements of 100 scfm/TPD, a parameter which is documented in
Chapter 6. The basis for estimating hooding and ducting requirements
was information provided from NCASI for actual retrofit situations. '
The capital costs for a separate incinerator were developed from data by
the Industrial Gas Cleaning Institute^ ' for an incinerator Cwith heat
recovery) application on an asphalt saturator plant. The preliminary EPA
costs were dissiminated to the selected group of paper companies for comment.
The responses from these companies was assimilated into revision of the
estimates, whieh are presented in Table 8-4.
It must be pointed out that some unusual retrofit problems can occur
at specific mills, where hooding brown stock washers may be constrained
severely by space limitations. In such mills, demolition of portions of
the washing building and reconstruction may be required. In other mills,
the washers may have to be replaced with semi-closed drum washers. For
these situations, retrofit costs may be much higher than those estimates
presented in Table 8-4.
8-12
-------
Table 8-4. RETROFIT CONTROL COSTS FOR BROWN STOCK WASHERS
A. Destruction 1n Separate Incinerator
Mill Size, TPD
Capital Costs ($)
Annual i zed Costs ($/Yr.)
Annual i zed Costs per Ton ($/T)
500
1 ,600,000
900,000
5.48
1000
2,500,000
1,670,000
5.08
1500
3,200,000
2,400,000
4.87
00
I
B. Destruction in Recovery Furnace
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs Per Ton ($/T)
500
900,000
192,000
1.17
1000
1 ,500,000
326,000
0.99
1500
1,900,000
423,000
0.86
-------
For separate incineration, a fuel penalty of 1.76 million BTU per
ton pulp was used. The basis for this is 22 million BTU per hour for a
(0)
30,000 scfm incinerator utilizing primary and secondary heat recovery. '
Responses from the industry indicated that this fuel penalty is just
slightly above the industry average. It should be pointed out that fuel
requirements would be sensitive to ventilation rates of the washers, the
degree of heat recovery, and use of catalyst to reduce ignition temperature.
The use of catalyst was not assumed In the above estimate.
8~.3_.4 Lime Kilns
Three levels of control are being considered for this affected
facility. The level of 40 ppm represents process controls of adding
more air and raising the cold-end temperature of the kiln by 100° F.
Adding a fan and instrumentation are assumed to be the required cost
items. In-house EPA estimates of costs were generated on the basis of
available costs for fans and instrumentation. These costs were then
reviewed with industry personnel for comparison with actual modification
co*ts for lime kilns.
The level of 20 ppm represents the combination of good lime mud
washing and control of the kiln process variables. Costs were provided
by companies on the basis of replacing centrifuges with vacuum drum
(9}
filters and adding another mud washing stage. NCASI guidelines ' for
2
cost estimating purposes were approximately 0.5 ft /ton pulp for filtra-
tion and 12 to 21 hours of retention in the clarifier stage.
The level of 8 ppm represents the addition of caustic to the lime
kiln scrubber as a complement to the aforementioned requirements for
achieving 8 ppm. The costs for caustic addition are assumed to be the
same as those reported in the Standards Support and Environmental Impact
Statement/10^
8-14
-------
The capital and annualized costs for the three control levels are
presented in Table 8-5. There are three points concerning Table 8-5 that
should be discussed. First, the fuel penalty assumed in the annualized
costs for process controls was based on use of 135 million BTU fuel for a
1000 ton per day mill. This is the enthalpy requirement to raise the
cold-end temperature of the kiln 100° F. Second, either process controls
or combined process controls with mud washing may not achieve the 40 ppm
level. In this circumstance, the mill may be short on lime burning capacity
and would have to add another lime kiln unit. Results from surveying the
industry indicate that a kiln addition in the range of 160-200 TPD (CaO
basis) seemed typical, regardless of mill size. Capital costs for the kiln
are $3 million; and annualized costs, $510,000. For this mill, the kiln
addition plus process controls on the existing kiln would be the requirement
to achieve 40 ppm.
Lastly, the costs in Table 8-5 are presented to demonstrate retrofit
problems with mud washing. Results from the industry survey varied widely.
To take account of this, costs were estimated for a low retrofit and a high
retrofit case. Some of the problems associated with high retrofit costs may
be related to space limitation, replacement of the entire washing system,
or changing condensate wash water to fresh water by addition of a condensate
stripper.
8.3.5 Black Liquor Oxidation Vents
The control techniques for achieving the 5 ppm level of TRS emissions
from black liquor oxidat.ion system vents are incineration and the use of
molecular oxygen for oxidation. The vent gases may be combined with the
brown stock washer vents and destroyed in the recovery furnace or separate
8-15
-------
Table 8-5. RETROFIT CONTROL COSTS FOR LIME KILNS
A. Process Controls^ '
CO
I
01
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr. }
Annual ized Costs per Ton ($/T)
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annualized Costs per Ton ($/T)
500
75,000
54,000
0.33
B. Mud Washing Capacity
500
480,000
91,200
0.56
C. Mud Washing Capacity
500
960,000 1
182,000
1.11
1000
100,000
107,000
0.33
- Low Retrofit
1000
730,000
139,000
0.42
- High Retrofit
1000
,460,000
278,000
0.85
1500
200,000
175,000
0.36
Penalty^
1500
950,000
181,000
0.37
Penalty^ 3^
1500
1,900,000
362,000
0.73
D. Caustic Addition
Hill Size, TPD
Capital Costs ($)
Annualized Costs ($/Yr.)
Annualized Costs per Ton ($/T)
500
-0-
3000
0.02
1000
-0-
6000
0.02
1500
-0-
9000
0.02
(1) In some situations, mills may be short on lime burning capacity. Typical capacity requirements
might be on the order of 160 to 200 TPD CaO, which was fouiind to be independent of mill size,
Capital costs would be $3 million; annualized costs of $510,000 per year.
(2) Includes only additional filtration and clarifier capacity.
(3) Includes condensate stripper for removal of TRS from water used for mud washing and/or charges
for space limitation.
-------
incinerator. However, incineration costs for the BLO vents will be con-
sidered separately from the brown stock washers.
Costs for incineration in recovery furnaces were taken from the costs
developed for new sources/11^ Very few responses were received from indus-
try, but one company did report similar costs,"z* which were found to be
comparable with EPA estimates. Costs for separate incinerators were
developed on the basis of a 26,000 scfm incinerator required for a 1000
ton per day mill. The source of the basic cost information was information
provided by the Industrial Gas Cleaning Institute.^^' Costs for a 500 ton
per day and 1500 ton per day were developed by using a scale factor of 0.3,
Fuel requirements were calculated to be 19 million BTU per hour for the
26,000 scfin incinerator on the basis of using primary and secondary heat
recovery. No costs for separate incineration were reported from industry.
Costs for molecular oxygen were developed on the basis of 45 tons
[\A\
oxygen per 800 ton pulp produced.vy Capital costs for skid-mounted low-
pressure oxygen plants were obtained from Airco, Inc. ^ ' The energy con-
sumption for use of molecular oxygen is assumed to be 20 kilowatt-hours per
ton of pulp, or 380 kilowatt-hours per ton of oxygen produced.^ *
A summary of capital and annualized costs for control of black liquor
oxidation vents is presented in Table 8-6. The costs for incineration in a
recovery furnace should reflect minimal retrofit problems. On the other'
hand, separate incineration and molecular oxygen should represent high
retrofit costs.
8.3.6 Smelt Dissolving Tank
The control strategy for achieving 8 ppm is the use of fresh water in
a scrubber. This scrubber should already be installed in most mills because
of state regulations for controlling particulates. The use of fresh water
8-17
-------
Table 8-6. RETROFIT CONTROL COSTS FOR BLACK LIQUOR OXIDATION VENTS
A. Destruction in Recovery Furnace
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr. )
Annual ized Costs per Ton ($/T)
500
220,000
60,000
0.37
1000
330,000
96,000
0.29
1500
432,000
133,000
0.27
B. Destruction in Separate Incinerator
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
400,000
233,000
1.42
1000
560,000
420,000
1.28
1500
700,000
605,000
1.23
00
I
CO
C. Molecular Oxygen
Mill Size, TPD
Capital Costs ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
500
1,080,000
309,000
1.88
1000
1,640,000
509 ,000
1.55
1500
2,100,000
645,000
1.31
-------
is the only additional requirement which may be needed to reduce the TRS
effectively.
It is expected that costs associated with use of fresh water will be
small, During the survey, one company provided cost data for installing
particulate scrubbers on this affected facility on one mill. Other than
this example, no other company reported any cost for substituting fresh
water for process water with high sulfides content.
8.3.7 Condensate Strippers
The control technique for achieving the 5 ppm level is incineration
in the lime kiln for steam stripping and incineration in the recovery
furnace for air stripping. Only five mills in the U.S. presently employ
condensate strippers. Four of these mills are presently incinerating con-
densate stripper vents.
Those mills that may decide to install condensate strippers in the
future are assumed to incur costs similar to reported costs for new sources, '
Those costs which are in 1976 dollars are summarized in Table 8-7. These
costs include a fan, duct, seal pot, and flame arrester with the incinera-
tion point being the lime kiln. No retrofit penalty has been assigned
because there are very few condensate strippers in existing mills. Future
controls for new strippers should incur minimal retrofit costs. However,
it still is possible that some retrofit penalty could be incurred if prior
provisions have not been made to tap into the mill's piping system for
venting other noncondensibles to an incineration point within the mill.
8-19
-------
Table 8-7. CONTROL COSTS FOR CONDENSATE STRIPPER
CO
Mill Size, TPD
500
1000
1500
Capital Cost ($)
Annual ized Costs ($/Yr.)
Annual ized Costs per Ton ($/T)
16,200
6',300
0.04
22,700
7,800
0.02
28,100
8,900
0.02
-------
8.4 Incremental Costs For Model Mills
The purpose of this section is to present incremental control costs
over requirements for state regulations, on a total mill basis. To take
into account the interrelationships involved with the many significant
factors in estimation of retrofit costs, three model mill situations will
be utilized to depict costs. These model situations are described as
follows:
(1) a post-1965 modern mill
(2) an old (pre-1965) mill with low retrofit penalty
(3) an old (pre-1965) mill with high retrofit penalty
The modern mill has a recovery furnace of modern design that can
achieve the 5 ppm level under good operating conditions. The black liquor
burning unit may be either a direct contact furnace or an indirect contact
furnace. The latter type of furnace has been installed in mills only since
1967. Only the direct contact furnace will incur any control costs, which
will be for secondary black liquor oxidation to assure achievement of the
5 ppm level. As for as the remaining affected facilities, only low retro-
fit costs are assumed to be incurred.
The old mill built before 1965, with low retrofit penalty, is assumed
to have a recovery furnace(s) capable of achieving the 20 ppm level.
Furnace replacement costs would be incurred only for a 5 ppm system. On
the remaining affected facilities, low retrofit costs would be associated
with addition of controls.
The old mill built before 1965, with a high retrofit penalty, is
assumed to have a recovery furnace(s) that cannot achieve the 20 ppm
level. This may be due to a very old furnace, greater than 20 years of age,
3-21
-------
or insufficient black liquor burning capacity. The result of this
condition is that a new furnace will be required for all control systems.
The remaining affected facilities will incur high retrofit costs with the
addition of control.
The next important factor influencing incremental costs is the
variability in state regulations. Some states with pulp mills have no
emission control regulations for existing sources. One set of costs will
be presented for such states. Many states with pulp mills have regulations
which call for controls on existing recovery furnaces and incineration of
noncondensible gases from miscellaneous sources such as digesters, multiple
effect evaporators, and condensate strippers. Another set of incremental
costs will be presented for such states. Very few states, perhaps one or
two, require controls on existing lime kilns, and no states require controls
on existing brown stock washers and black liquor oxidation system vents.
Tables 8-8 and 8-9 present incremental costs for a modern mill under
the two regulatory situations as described earlier. Capital and annua-
lized costs are presented for the six alternative control systems detailed
in Table 8-1. Costs are presented for the direct contact and indirect
contact recovery furnace designs. For brown stock washers and black
liquor oxidation system vents, costs are presented only for destruction in
a separate incineration. Destruction of TRS in the recovery furnace may
not be widely applicable due to attendant problems of inflexibility and
possible explosions, as pointed out in Chapter 6.
Where no regulations exist (Table 8-8), incremental annualized control
costs for the model mill with the direct contact recovery furnace range
from $1.73 per ton for control system 6 to $8.53 per ton for system 1. For
8-22
-------
Table 8-8. INCREMENTAL RETROFIT CONTROL COSTS FOR A 1000 TPD MODERN MILL (BUILT AFTER 1965)
Location: Statt with No Regulations
No. 1
No. 2
No. 3
00
ro
CO
A.
B.
Recovery Furnace
a) Direct Contact
b) Indirect Contact
Batch Digester and^'
Capital.
Costs
($1000)
1,000
0
Multiple Effect Evaporator goo
C.
D.
E.
F.
Brown Stock Washers*2'
Black Liquor Oxidation
System Ventsl2)
(Direct Contact Only)
Lime Kiln*3)
Condensate Stripper
TOTAL COSTS
a) Direct Contact
b) Indirect Contact
2,500
560
830
23
5,813
4,253
Annual 1 zed
Costs
($1000/yr)
242
0
210
1,670
420
252
8
2,802
2,140
Unit
Annual 1zed
Costs
($/T)
0.74
0
0.64
5.08
1.28
0.77
0.02
8.53
6.51
Capital
Costs
($1000)
1,000
0
goo
0
0
830
23
2,753
1,753
Annual 1 zed
Costs
(SlOOO/yr)
242
0
210
0
0
252
8
712
470
Unit
Annual 1 zed
Costs
(JAM
0.74
0
0.64
0
0
0.77
0.02
2.17
1.43
Capital
Costs
($1000)
1,000
0
900
0
0
830
23
2,753
1,753
Annual 1 zed
Costs
($1000/yr)
242
0
210
0
0
252
8
712
470
Unit
Annualized
Costs
($/T)
0.74
0
0.64
0
0
0.77
0.02
2.17
1.43
No. 4
No. 5
No. 6
A.
B.
C.
D.
E.
F.
Recovery Furnace
a) Direct Contact
b) Indirect Contact
Batch Digester and*1)
Multiple Effect Evaporator
Brown Stock Washers* '
Black Liquor Oxidation
System Vents*2'
(Direct Contact Only)
Lime Kiln' '
Condensate Stripper
TOTAL COSTS
a) Direct Contact
b) Indirect Contact
1,000
0
900
0
. 0
830
23
2,753
1,753
242
0
210
0
0
246
8
706
464
0.74
0
0.64
0
0
0.75
0.02
2.15
1.41
1,000
0
900
0
' 0
100
23
2,023
1,023
242
0
210
0
0
107
8
567
325
0
0
'''
0
0
1
0
.74
0
.64
0
0
.33
.02
.73
.99
1,000
0
900
0
0
100
23
\0'3
1 ,023
242
0
210
0
0
107
8
567
325
0.74
0
0.64
0
0
0.33
0.02
1.73
0.99
(1) Low retrofit penalty
(2) Destruction in separate incinerator
(3) Low retrofit penalty
-------
Table 8-9. INCREMENTAL RETROFIT CONTROL COSTS FOR A
Location:
No. 1
1000 TPD MODERN MILL (BUILT AFTER 1965)
State with Typical Regulations^1)
No. 2
No. 3
A.
B.
C.
D.
E.
F.
Recovery Furnace
a) Direct Contact
b) Indirect Contact
Batch Digesters and
Multiple Effect Evaporator
Brown Stock Washers'2'
Black Liquor Oxidation
System Vents (2)
(Direct Contact Only)
Lime Kiln(3)
Condensate Stripper
TOTAL COSTS
a) Direct Contact
b) Indirect Contact
Capital
Costs
($1000)
1,000
0
0
2,500
560
830
0
4,890
3,330
Annual ized
Costs
($1000/yr)
242
0
0
1,670
420
252
0
2,584
1,922
Unit
Annual ized
Costs
($/T)
0.74
0
0
5.08
1.28
0.77
0
7.87
5.85
Capital
Costs
($1000)
1,000
0
0
0
0
830
0
1,830
830
Annual ized
Costs
($1000/yr)
242
0
0
0
0
252
0
494
252
Unit
Annual ized
Costs
($/T)
0.74
0
0
0
0
0.77
0
1.50
0.77
Capital
Costs
($1000)
0
0
0
0
0
830
0
830
830
Annual ized
Costs
($1000/yr)
0
0
0
0
0
252
0
252
252
Unit
Annual ized
Costs
(J/T)
0
0
0
0
0
0.77
0
~Q~.77
0.77
co
r\>
A.
B.
C.
D.
E.
F.
Recovery Furnace
a) Direct Contact
b) Indirect Contact
Batch Digester and
Multiple Effect Evaporator
Brown Stock Mashers'2'
Black Liquor Oxidation
System Vents!2'
(Direct Contact Only)
Lime Kiln'3'
Condensate Stripper
TOTAL COSTS
a) Direct Contact
b) Indirect Contact
0
0
0
0
'o
830
0
830
830
No. 4
0
0
0
0
0
246
0
246
246
0
0
0
0
0
0.75
0
0.75
0.75
lio. 5
0
0
0
.0
0
100
0
100
100
C
0
0
0
0
107
0
107
107
0
0
0
0
0
0.33
0
0.33
0.33
No. 6
1,000
0
0
0
0
100
0
1,100
100
242
0
0
0
0
107
0
349
107
0.74
0
0
0
0
0.33
0
1.07
0.33
(1) A typical state is assumed to require 20 ppm for the recovery furnace and Incineration (5 ppm) of TRS emissions from digesters, multiple
effect evaporators, and condensate strippers.
(2) Destruction In separate Incinerator.
(3) Low retrofit penalty
-------
the indirect contact furnace, the model mill incurs incremental annualized
control costs ranging from $0.99 per ton for system 6 to $6.51 per ton
for system 1. In states with the typical composite of regulations (Table
8-9), incremental annualized control costs range from $1.07 per ton for
system 6 to $7.87 per ton for system 1 in a mill with a direct contact
furnace. For the mill with the indirect contact furnace, incremental
annualized costs for these respective systems range from $0.33 per ton to
$5.85 per ton.
For modern mills, the most significant cost is that associated with
incineration of brown stock washer gases. The control of this affected
facility alone is $5.08 per ton for system 1 (Table 8-8 and 8-9) of which
$3.50 per ton is a fuel penalty. The fuel penalty would not be incurred
in some mills where the washer gases can be Incinerated in a recovery furnace.
It should be noted that the rather significant $1.28 per ton cost for
controlling black, liquor oxidation system vents only occurs for the mill
with a direct contact furnace.
Tables 8-10 and 8-11 present control costs for an old mill with a low
retrofit penalty. Capital and annualized costs are presented for the
six alternative control systems detailed in Table 8-1. The costs for
control of brown stock washers and black liquor oxidation system vents is
based on separate incineration. The replacement of the recovery furnace is
assumed to be required for systems 1,2, and 6. Furthermore, the furnace
is assumed to have ample capacity to support the mill's normal production
needs. This assumption eliminates any consideration of additional furnace
investment for systems 3, 4, and 5.
Where no state regulations exist (Table 8-10), incremental annualized
8-25
-------
Table 8-10. INCREMENTAL RETROFIT CONTROL COSTS FOR A 1000 TPD OLD MILL (BUILT BEFORE 1955) - LOW RETROFIT PENALTY
' Location: State with No Regulations
No. 1
No. 2
No. 3
A.
B.
C.
0.
E.
F.
A.
B.
C.
D.
E.
F.
Recovery Furnace'^'
Batch Digester and
Multiple Effect Evaporators
Brown Stock Washers ^'
Black Liquor Oxidation
System Vents (2>
Lime Kiln
Condensate Stripper
TOTAL COSTS
Recovery Furnace^ '
Batch Digester and
Multiple Effect Evaporators
Brown Stock Washers (2)
Black Liquor Oxidation
System Vents (2)
Lime Kiln
Condensate Stripper
TOTAL COSTS
Capital
Costs
23,300
900-
2,500
560
830
23
28,113
1,000
900
0
0
830
23
2,753
Annual Ized
Costs
(tlOoW)
4,000
210
1,670
420
252
8
6,560
No. 4
242
210
0
0
246
8
706
Unit
Annual 1 zed
Costs
($/T)
12.18
0.64
5.08
1.28
0.77
0.02
19.97
0.74
0.64
0
0
0.75
0.02
Capital
Costs
C$1000)
23,300
900
0
0
830
23
25,053
Annual 1 zed
Costs
($1000/yr)
4,000
210
0
0
252
8
4,470
Unit
Annual 1 zed
Costs
(VT)
12.18
0.64
0
0
0.77
0.02
13.61
No. 5
1,000
900
0
0
100
23
2.15 2,023
242
210
0
0
107
8
567
0.74
0.64
0
0
0.33
0.02
Caoital
Costs
($1000)
1,000
900
0
0
830
23
2,7.53
Annual i zed
Costs
($1000/yr)
242
210
0
0
252
8
712
Unit
Annual i zed
Costs
($/T)
0.74
0.64
0
0
0.77
0.02
2.17
No. 6 !
23,300
900
0
0
100
23
1.73 24,323
4,000
210
0
0
107
8
4,325
12.18
i
0.64
0
0
0.33
0.02
13.17
co
i
ro
CTl
0)
Recovery furnace would be a direct contact furnace only
(2) Destruction in separate Incinerator
-------
Table 8-11. INCREMENTAL RETROFIT CONTROL COSTS'FOR A 1000 TPD OLD MILL
Location: State with Typical Regulations
BEFORE 1965) - LOW RETROFIT PENALTY
No. 1
No. 2
No. 3
A. Recovery Furnace
B. Batch Digesters and
Multiple Effect Evaporators
C. Brown Stock Washers'2'
0. Black Liquor Oxidation
System Vents(2)
E. Lime Kiln
F. Condensate Stripper
TOTAL COSTS
Capital
Costs
($1000)
23,300
0
2,500
560
830
0
27,190
Annual ized
Costs
($1000/yr)
4,000
0
1,670
420
252
0
6,342
Unit
Annual 1 zed
Costs
($/T)
12.18
0
5.08
1.28
0.77
0
19.31
Capital
Costs
(tiooo)
23.300
0
0
0
830.
0
24,130
Annual 1 zed
Costs
(SlOOO/yr)
4,000
0
0
0
252
0
4,252
Unit
Annual Ized
Costs
($/T)
12.18
0
0
0
0.77
0
12.95
Capital
Costs
($1000)
0
0
0
0
830
0
830
Annual ized
Costs
($1000/yr)
0
0
0
0
252
0
252
Unit
Annual 1 zed
Costs
($/T)
0
0
0
0
0.77
0
0.77
CO
ro
A.
B.
C.
D.
E.
F.
Recovery Furnace
Batch Digesters and
Multiple Effect Evaporators
Brown Stock Washers^
Black Liquor Oxidation
System Vents (2)
Lime Kiln
Condensate Stripper
TOTAL COSTS
Capital
Costs
($1000)
0
0
0
0
830
0
830
No. 4
Annual ized
Costs
($1000/yr)-
0
0
0
0
246
0
246
Unit
Annual ized
Costs
($/T)
0
0
0
0
0.75
0
0.75
No. 5
Capital
Costs
($1000)
0
0
0
0
100
0
100
Annual ized
Costs
($1000/yr)
0
0
0
0
107
0
107
Unit
Annual ized
Costs
($/T)
0
0
0
0
0.33
0
0.33
No. 6
Capital
Costs
($1000)
23,300
0
0
0
100
0
23,400
Annual ized
Costs
($1000/yr)
4,000
0
0
0
107
0
4,107
Unit
Annual ized
Costs
($/T)
12.18
0
0
0
0.33
0
12.51
(1) A typical state is assumed to require 20 ppni for the recovery furnace and Incineration (5 ppm) of TRS emissions turn digesters, multiple effect
evaporators, and condensate strippers
(2) Destruction In separate Incinerator.
-------
costs range from $1.73 per ton to $2.17 per ton'for systems 3, 4, and 5.
The incremental annualized costs for systems 1, 2, and 6 range from $13.17
to $19.97 per ton. The replacement of the recovery furnace is responsible
for $12.18 per ton of these costs.
In a state with typical regulations, incremental annualized costs
range from $0.33 to $0.77 per ton for systems 3, 4, and 5. For systems
1, 2, and 6, incremental annualized costs range from $12.51 per ton to
$19.31 per ton, the most expensive being system 1. Again, recovery furnace
replacement is responsible for $12.18 per ton.
Tables 8-12 and.8-13 present control costs for the old model mill with
the incidental high retrofit penalty. Capital and annualized costs are
presented for the six alternative control systems similar to the previous
models. Costs of control for the washer gases and the oxidation vents is
based on separate incineration. The construction of new furnace sized to
support the entire 1000 ton per day mill is assumed in the cost estimates
for: (1) all control systems in states with no regulations, and (2)
control systems 1, 2, and 6 in states with typical regulations. The aspects
of the high retrofit costs for this model mill involve the controlling of
the digestors/evaporators and the lime kiln. The two factors associated
with the lime kiln, namely additional requirements for lime mud washing
and additional lime burning capacity, have been taken into account for this
model mill. In summary, Table 8-12 would represent the worst situation - -
an old rr..il with incidental high retrofit penalties in a state with no
regulations.
Where no regulations exist (Table 8-12), incremental annualized costs
range from $15.46 per ton for system 6 to $22.68 per ton for system 1. New
furnace costs are responsible for $12.18 per ton of these costs. In
8-28
-------
Table 8-12. INCREMENTAL RETROFIT CONTROL COSTS FOR A 1000 TPD OLD HILL (BUILT BEFORE 1965) - HIGH RETROFIT PEliALTY
Location: State With No Regulations
No. 1
No. 2
No. 3
A.
B.
C.
D.
E.
F.
10 A.
B.
C.
D.
E.
F.
Recovery Furnace
"ate1" "iGester and
Multiple Effect Evaporator
Brown Stock Washers' '
Black Liquor Oxidation
System Vents(2)
Lime K1ln(3)
Condensate Stripper
TOTAL COSTS
Recovery Furnace
Batch Digester and
Multiple Effect Evaporator
Brown Stock Washers^
Black Liquor Oxidation
System Vents (2)
Lime Kiln*3*
Condensate Stripper
TOTAL COSTS
Capital
Costs
($1000)
23,300
2,000
2,500
560
4,560
23
32,943
23 ,30U
2,000
0
0
4,560
23
29.U83
Annual ized
Costs
($1000/yr)
4,000
452
1,670
420
901
8
7,451
No. 4
4,OOU
452
0
0
896
8
b,3b5
Unit
Annual Ized
Costs
($/T)
12.18
1.38
5.08
1.28
2.74
0.02
22.68
12. 18
1.38
0
0
2.73
0.02
16.30
Capital
Costs
($1000)
23,300
2,000
0
0
4,560
23
29,883
23,300
2,000
0
0
3,100
23
28.4U/
Annual 1 zed
Costs
($1000/yr)
4,000
452
0
0
' 901
8
5,361
No. 5
4,000
452
0
0
617
8
b.O//
Unit
Annual 1 zed
Costs
($/T)
12
1
2
0
16
12
1
1
0
15
.18
.38
0
0
.74
.02
.32
. IB
.38
0
0
.88
.02
.46
Capital
Costs
($1000)
23.300
2,000
0
0
4,560
23
29,883
23.3UO
2.000
0
0
3,100
23
2B.407
Annual ized
Costs
($1000/yr)
4,000
452
0
0
901
8
5,361
No. 6
4,000
452
0
0
617
8
5,077
Unit
Annual ized
Costs
($/T)
12.18
1.38
0
0
2.74
0.02
16.32
12.18
1.38
0
0
1.88
0.02
15.46
(1) Recovery furnace - direct contact only. Furnace with an assumed age exceeding twenty years would have to be replaced for each control strategy.
(2) Destruction in separate Incinerator.
(3) High retrofit expenditures for the following: addition of a new lime kiln for each control strategy and addition of Condensate stripper for
strategy numbers 1 through 4.
-------
Table 3-13. INCREMENTAL RETROFIT CONTROL COSTS FOR A 10UU TPL) OLD MILL (13U1LJ>UEFORE 1965) - HIGH RETROFIT PENALTY
Location: State with Typical Regulations1 '
No. 1.
No. 2
No. 3
A
B
C.
D
L.
r
Re-overy Furnace
Ba ;ch Digester and
'lultiole Effect Evaporators
Brr.-;n Stock Washers'2'
Bl ck Liquor Oxidation
S.-stem VentsU)
Li e Kiln'3'
Co'idensate Stripper
TOTAL COSTS
A
B.
C.
D.
E.
F.
TO
Recovery Furnace
Batch Digester and Multiple
Lffect Evaporators
Brv.vn Stock Washers'2'
Bla;k Liquor Oxidation
System Vents' '
Lir'e Kiln'3'
Cotidensate Stripper
TAL COSTS
Capital
Costs
($1000)
23,300
0
2,500
560
4,560
0
30,920
Capital
Costs
($1000)
0
0
0
0
4,560
0
4,560
Annual i zed
Costs
($1000/yr)
4,000
0
1,670
420
901
0
6,991
No. 4
Annual ized
Costs
($1000/yr)
0
0
0
0
896
0
896
Unit
Annual ized
Costs
(S/T)
12.18
0
5.08
1.28
2.74
0
21.28
Unit
Annual ized
Costs
($/T)
0
0
0
0
2.73
0
2.73
Capital
Costs
($1000)
23,300
0
0
0
4,560
0
27,860
Annual 1 zed
Costs
($1000/yr)
4,000
0
0
0
901
0
4,901
Unit
Annual ized
Costs
($/T)
12.18
0
0
0
2.74
0
14.92
No. 5
Capital
Costs
($1000)
0
0
0
0
3,100
0
3,100
Annual ized
Costs
($1000/yr)
0
0
0
0
617
0
617
Unit
Annual ized
Costs
($/T)
0
0
0
0
1.88
0
1.88
Capital
Costs
($1000)
0
0
0
0
4,560
0
4,560
Annual ized
Costs
($10QO/yr)
0
0
0
0
901
0
901
Unit
Annual ized
Costs
($/T)
0
0
0
0
2.74
0
2.74
No. 6
Capital
Costs
($1000)
23,300
0
0
0
3,100
0
26 ,400
Annual ized
Costs
($lCOO/yr)
4,000
0
0
0
617
0
4,617
Unit
Annual ized
Costs
($/T)
12.18
0
0
0
1.88
0
14.05
CO
I
OJ
o
(1) A typical state is assumed to require 20 ppm for the recovery furnace and incineration (5 ppm) of TRS emissions from digesters, multiple effect
e'. iporators, and condensate strippers.
(2)D.-,truction in separate incinerator
(3) ' igh retrofit expenditures for the following: addition of a new lime kiln for each control system and addition of
:>ndensate stripper for system numbers 1 through 4.
-------
states with typical regulations, the incremental annualized costs range
from $14.06 per ton to $21.28 per ton for control systems 1, 2, and 6,
with system 1 being most expensive. For control systems 3, 4, and 5, the
incremental annualized costs range from $1.88 per ton to $2.74 per ton.
From the previous discussion on old mills, the most significant
factor that frequently re-appears in the total mill costs has been require-
ments for new recovery furnace investment. Up to this point all capital
related charges associated with purchasing, installing, and ownership of
the recovery furnace have been presented as control costs. However, the
recovery furnace is a productive capital asset in the sense that it
contributes to the economics of pulp production with recovery of energy
and chemicals. Consequently, some credit for a productive asset should be
deducted from the control costs. However, it is very difficult to estimate
this credit on a source by source basis in terms of dollars per ton.
Therefore, no credit was deducted.
The extent of credit to be deductible is very source specific. The
amount of credit would depend on the remaining economic life of existing
furnace equipment. In a specific mill the recovery furnace could be very
old, like thirty years of age, and very inefficient. Such a mill would
probably be scheduling for the replacement of the old furnace in the near
future. Here, the replacement cost should be treated as a normal produc-
tive asset with no credit given for control costs. In another mill, a
recovery furnace may have a significant amount of residual economic life,
say 15 years. Suppose a state should require a 5 ppm level which would
force the scrapping and replacement of this recovery furnace. In this
situation, the capital value foregone in scrapping the furnace should be
8-31
-------
the approximate control cost.
In a similar vein, mills that tend to overload recovery furnaces may
be required to provide additional black liquor burning capacity to reduce
TRS emissions. The incremental capacity sufficient to reduce the emissions
to a satisfactory level should be the approximate control cost although a
mill would install a complete new recovery unit which would exceed the
necessary incremental capacity.
8.5 Aggregate Costs For Industry
In this section the estimated incremental control costs are reported
for the existing kraft pulp industry for the six alternative emission
control systems outlined in Table 8-1. The approach used was to estimate
these costs for each individual mill on the basis of the best technical
information available for each mill regarding production rates, furnace
capacity and age, type of controls used, status of state regulations, and
other technical parameters. Section 8.3, Costs For Affected Facilities,
which relates control costs as a function of mill size was used to make
the estimates. The model mill approach as outlined in Section 8.4 was not
considered suitable to estimate total industry costs because of the wide
variability in mill characteristics and state regulatory requirements. How-
ever, the two approaches should give consistent results. Verification of
the model mill approach with the results obtained by the individual mill
approach does support this claim.
Actual cost information received from 42 mills during the EPA industry
survey was used to derive the Section 8.3 costs. From these costs, estimates
of capital and annualized were made individually for 77 mills which were not
contacted in the industry survey. The costs for these mills were then
8-32
-------
combined with the actual costs received for the 42 surveyed mills to
derive industry totals.
The summary of industry incremental costs are reported in Table 8-14
for each system. Capital and annualized costs are presented for industry
totals and on a unit basis. In addition, incremental capital costs are
related to mill investment as a measured percentage. The investment for a
battery limits mill is $150 million is 1976 dollars, which was derived
from a study for EPA's Office of Solid Waste ManagementJ18^ Similarly,
incremental annualized costs are related to the market pulp price as a
measured percentage. The price used was $330, which is the currently
quoted contract price for domestic bleached kraft pulp/ ' This price
represents the average of pulps derived from hard and softwoods.
The industry-wide incremental annualized control costs range from
$1.99 per ton for system 5 to $12.72 per ton for system 1. The $1.99 per
ton figure is predicated on the basis of replacement of 18 recovery
furnaces and 3 lime kilns. The $12.72 per ton figure is predicated on the
basis of replacement of 63 recovery furnaces and 33 lime kilns. It should
be noted that systems 3 and 4 would require replacement of 18 recovery
furnaces and 33 lime kilns. The corresponding percentages in relation to
market pulp price are 0.6 percent for system 5 ($1.99 per ton) and 3.9 per
ton for system 1 ($12.72 per ton).
Capital, requirements for incremental controls range from $10.32 per
ton capacity for system 5 to $46.20 per ton for system 1. In relation to
requirements for new mill investment, these estimates amount to 1.8 percent
for system 5 and 8.1 per cent for system 1.
The approach used to develop industry-wide costs represents a composite
8-33
-------
Table 8-14. Sumary of Industry IncrwtwUl Control Costs for Stx Alternatives Control Strategies
Total Capital ($1000)
Capital , per ton
capacity ($/T)
Capital , as a percent
of new mill(l)
Total Annual i zed Costs
($1000/yr)
Unit Annual i zed Costs ($/T)
Annuali zed costs, as a
percent of market pulp
price (2)
No. 1
1,602,000
46.20
8.1
441 ,000
12.72
3.9
No. 2
1,275,000
36.78
6.5
237,000
6.84
2.1
No. 3
495,000
14.27
2.5
94,300
2.72
0.8
No. 4
495,000
14.27
2.5
93,600
2.70
0.8
No. 5
358,000
10.32
1.8
69 ,000
1.99
0.6
No. 6
1,148,000
33.11
5.8
200,000
5.76
1.7
00
I
CO
(^Capital Costs for New Battery Limits 800 TPD Mill is $150 million, or $570 per ton (1976 dollars) Source:
Arthur D. Little, Reference 18.
(2)Market pulp price is $330 per ton, based on fourth quarter 1976 prices for domestic kraft bleached pulp.
Source: Paper Trade Journal.
-------
of many different types of state regulations and the individual character
of 119 mills. Although the model mill approach in Section 8.4 was
considered inappropriate to estimate industry costs, there should be some
linkage between the industry-wide costs and the model mill costs on a unit
basis. A comparison of the two approaches revealed that the industry-wide
costs in Table 8-14 fall about midway between cost requirements for a
modern mill in a state with typical regulations (Table 8-9) and for an old
mill with low retrofit penalty in a state with no regulations (Table 8-10)
for systems 1, 2, and 6. Industry-wide costs in Table 8-14 are somewhat
higher than the costs reported for an old mill (Table 8-10) for systems
3, 4, and 5. A conclusion would be that there is some reasonable agreement
in the magnitude of the costs developed from the two separate approaches.
8.6 Cost-Effectiveness
An analysis was made to evaluate the cost-effectiveness of the six
alternative emission control systems in terms of their contribution to
reducing national TRS emissions. The cost-effectiveness technique is a
useful tool in selecting an appropriate control system as a recommended
guideline. In this selection, those control systems that have significantly
high control costs in terms of their pollutant removal are rejected as
viable control recommendations. It should be strongly emphasized that the
cost-effectiveness approach for recommending controls is only applicable
for welfare-related 111-d pollutants, such as TRS. For health-related 111-d
pollutants, an economic impact analysis is a requirement for determining
affordability of best controls.
The industry aggregate annualized control costs presented in Table
8-14 and the national emission reduction data reported in Table 9-2 were
8-35
-------
used to make the cost-effectiveness calculations. The results are
presented in Table 8-15. The control systems are ranked in ascending
order in terms of emission reduction and costs, starting with system 5 as
the least expensive. Two calculations of cost-effectiveness are presented
for each control system in columns (E) and (F). The calculation in
column (E) simply represents the costs per ton removed by a particular
control system. The calculation in column (F) represents the marginal costs
per ton removed by a particular control system relative to a system of
lower ranking. The marginal cost calculation is a more sensitive indicator
in revealing the more expensive control system. For example, in Table
8-15, system 6 costs $75,500 per ton marginally. This is much more
significant than the $1750 for system 3 or $11,180 for system 4. With
respect to actual cost per ton, system 6 costs $3000 per ton, which is
significantly higher, to a lesser degree, than the approximate $1400 per
ton for systems 4 and 3.
Based on the data in Table 8-15, it would seem reasonable to reject
control systems 6, 2, and 1 as being cost-effective. System 5 might be
considered a minimal strategy, costing $1060 per ton. Control systems 4
and 3 cost somewhat more, about $1400 per ton, which would not seem to be of such
a magnitude to preclude consideration of these control systems as a viable
control technology.
8-36
-------
Table 8-15. COST EFFECTIVENESS DATA FOR ALTERNATIVE CONTROL SYSTEMS
Control System
5
4
3
6
2
1
A
National
Emission
Reduction
tons/yr.
65,000
67,200
67,600
69,000
71 ,500
77,700
B
Marginal
Emission
Reduction
tons/yr.
65,000
2,200
400
1,400
2,500
6,200
C
Industry
Annual ized
Control
Costs
$1000/yr
69,000
93,600
94,300
200,000
237,000
441 ,000
' D
Marginal
Annual ized
Control
Costs
$1000/yr.
69,000
24,600
700
105,700
37,000
204,000
E
Annual 1 zed
Costs per
ton Removed
(C)r(A).
($/ton)
1,060
1,390
1,395
3,000
3,315
5,675
F
Marginal
Annual ized
Cost per ton
Removed
(D)-(B),$/ton
1,060
11,180
1,750
75,500
14,800
32,900
00
I
oo
-------
References for Chapter 8
1. Post's 1973 Pulp and Paper Directory,Miller Freeman Publications,
Inc., San Francisco.
2. Lockwood's Directory of the Paper and Allied Trades, Lockwood Publishing
Co., New York, 1974.
3. Correspondence from Mr. Russell Blosser, National Council for Air and
Stream Improvement, Inc., to Mr. Frank L. Bunyard, EPA, OAQPS,
April 22, 1975.
4. See Reference 3.
5. Correspondence from Mr. C. T. Tolar, Rust Engineering Co., Birmingham,
Ala., to Mr. Paul A. Boys, October 20, 1972.
6. Correspondence from Mr. Russell Blosser, National Council for Air and
Stream Improvement, Inc., to Mr. Paul A. Boys, November 17, 1972.
7. Air Pollution Control Technology and Costs: Seven Selected Emission
Sources. Industrial Gas Cleaning Institute, EPA-450/3-74-060,
National Technical Information Service, Springfield, Va., December, 1974.
8. Report of Fuel Requirements, Capital Cost and Operating Expense for
Catalytic and Thermal Afterburners, CE Air Preheater for Industrial
Gas Cleaning Institute, EPA-450/3-76-031, National Technical Informa-
tion Service, Springfield, Va., September 1976.
9. See Reference 3.
10. Standards Support and Environmental Impact Statement - Volume I:
Proposed Standards of Performance for Kraft Pulp Mills, EPA-450/2-76-014a
National Technical Information Service, Springfield, Va., September 1976.
11. See reference 10.
12. Correspondence from Mr. Joe Kolberg, Boise Cascade, to Mr. Frank L.
Bunyard, OAQPS, EPA, May 1975.
13. See reference 8.
14. See reference 5.
15. Telephone conversation from F. L. Bunyard, OAQPS, EPA, to George
Horvat, Airco, Inc., December 30, 1974.
8-38
-------
16. Investment and Operating Cost Data for Low Pressure Oxygen Plant
Applicability to Non-Ferrous Metallurgy, Volcan - Cincinnati, EPA
Contract No. 68-02-2099, Task No. 2, September 29, 1972.
17. See reference 10.
18. Analysis of Demand and Supply for Secondary Fiber in the U.S. Paper
and Paperboard Industry. Volume 2: Section IX - Process Economics.
Arthur D. Little Report for Contract #68-01-02220, Environmental
Protection Agency, Office of Solid Waste Management Programs, March, 1975.
19. Paper Trade Journal. January 1, 1977.
8-39
-------
9. ENVIRONMENTAL IMPACT OF TRS CONTROLS
The environmental impacts discussed are for each of the control tech-
niques and control systems mentioned in Chapter 6. This includes discussions
on the impacts on air, water, and solid waste pollution and energy consumption
for a relatively large kraft pulp mill (907 megagrams of pulp per day) and on
a national basis.
9.1 AIR POLLUTION IMPACT
9.1.1 Annual Air Emission Reductions
Installation of the various control techniques described in Section 6.1
are estimated to reduce TRS emissions from the existing kraft industry by
the amounts indicated in Table 9-1. Emission reductions range from 20.6
percent for digester systems to 96 to 97 percent for digester systems and lime
kiln systems. All values presented in Table 9-1 are based on information
presented in Chapters 5 and 6 and Appendix A of this study.
The following procedure was used to arrive at the numbers listed in
Table 9-1. The values listed in Column 2 (Current National Average Emission
Rate) were previously mentioned in Chapter 5 and are based on the information
listed in Appendix A. Information in Appendix A is based upon discussions with
various kraft pulp mills and state control agencies. Column 5 presents the
percentage of existing facilities presently using the control techniques
described in Column 2 as based on the information listed in Appendix A. The
values in Column 6 were developed by applying the emission level achievable by
9-1
-------
TABLE 9-1
ENVIRONMENTAL IMPACT OF CONTROLLING THE
VARIOUS TRS SOURCES IN A KRAFT MILL
Source
Recovery Furnace
Digester
Multiple-effect
Evaporator
Lime Kiln
Brown Stock Washer
System
Black Liquor Oxidation
System
Smelt Dissolving Tank
Condensate Strippers
Current Average
National Emission
g/Kg ADP (IT ADP)
1 .25
1.25
0.32
0.32
0.22
0.22
0.31
0.31
0.31
0.15
0.05
0.05
0.10
0.11
0.11
(2.5)
(2.5)
(0.64)
(0.64)
(0.43)
(0.43)
(0.62)
(0-62)
(0.62)
(0.3)
(0.1)
(0.1)
(0.2)
(0.22)
(0.22)
Control
Technique
BLO ( 20 ppm)
BLO ( 5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1 ) Process Controls
2) Process Control & High
Eff . Mud Washing
3) Process Controls, High
Eff. Mud Mashing &
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Incineration
Scrubber
Emission Level
Achievable With
Control Technique
g/Kg ADP (# T/ADP)
0.31
0.08
0.59
0.01
0.04
0.01
0.10
0.05
0.021
0.01
0.01
0
0.013
0.01
0.5
(0.6)
(0.15)
(1.17)
(0.02)
(0.08)
(0.02)
(0.2)
(0.1)
(0.04)
(0.02)
(0.02)
(0)
(0.025)
(0.02)
(1.0)
% of Capacity
Not Presently
Achieving This
Level
35.7
86.9
41.8
42.4
41.4
45.9
71.8
90.5
99.4
99.1
98.0
98.0
20
0
Estimated Average National
Emission If Control
Technique Is Required
g/Ks ADP (# T/ADP)
0.25
0.08
0.26
0.01
0.03
0.01
0.09
0.05
0.021
0.01
0.01
0
0.01
0.01
0.11
(0.5)
(0.15)
(0.51)
(0.02)
(0.05)
(0.02)
(0.18)
(0.1)
(0.04)
(0.02)
(0.02)
(0)
(0.02)
(0.02)
(0.22
% Emission
Reduction
Achieved
Nationally
80.0
94.0
20.6
96.9
89.1
95.3
71.0
84.0
92.0
93.3
80.0
100.0
87.5
90.9
0
National
Emission Reduction
Mq/year (tons/year)
31 ,200
36,690
2,040
9,300
5,940
6,400
6,900
8,900
9,300
4,350
1,270
1,540
2,800
3,130
0
(34,400)
(40,450)
( 2,250)
(10,700)
( 6,550)
( 7.050)
( 7.600)
( 9,800)
(10.250)
( 4,800)
( 1.400)
( 1.700)
( 3,100)
( 3.450)
(0)
-------
a particulate control device (Column 4) to those existing mills which are not
presently achieving that level, as listed in Appendix A, and calculating the
national average emission level that would result. Column 7 (Percent Emission
Reduction Achieved Nationally) is the percent difference between Columns 2 and 6.
The national emission reduction achieved by a specific control device (Column 8)
was calculated by multiplying the difference between Columns 2 and 6 by the annual
kraft pulp production rate (31,196,000 megagrams/year).
Table 9-1 shows that the greatest reduction of TRS emissions is achieved
by controlling the recovery furnace system with the digester system, lime kiln,
multiple-effect evaporator system, brown stock washer system, smelt dissolving
tank, black liquor oxidation system and condensate stripping system following
in decreasing impact.
Table 9-2 shows the impact of the various control systems mentioned in
Section 6.2. For example, if System No. 1 (best available technology as
defined for NSPS) was applied to each source, the TRS emissions from the kraft
industry would be reduced by about 70,500 megagrams per year (77,700 tons
per year) or 94.2 percent. System No. 5, if applied, would result in the least
impact but would still reduce TRS emissions by about 59,000 megagrams per
year (65,000 tons per year) or 78.8 percent. Control of four sources in a
kraft mill account for a major portion of the impact achieved by each of the
control systems. These four sources are the recovery furnace, digester system,
multiple-effect evaporator system, and the lime kiln.
9.1.2 Annual Air Emission Increase
The only control techniques mentioned in Chapter 6 that would apparently
result in increasing the emission rates of other pollutants is the incineration
of the vent gases from the brown stock washer systems and the black liquor
9-3
-------
TABLE 9-2
ENVIRONMENTAL IMPACT OF VARIOUS CONTROL SYSTEMS
FOR EXISTING KRAFT PULP MILLS
Estimate Average
TRS Emission With
National
Emission
Control
System
No.
No.
No.
No.
No.
No.
1
2
3
4
5
6
Control System
g/Kq ADP (#T ADP)
0
0
0
0
0
0
.14
.32
.44
.45
.51
.40
(0.
(o.
(o.
(o.
(1.
(0.
28)
64)
87)
89)
02)
79)
Emission
Reduction*
94.
86.
81.
81.
78.
83.
2
7
9
5
8
5
Reduction
megagrams/year (tons/year)
70,
64,
61,
61,
59,
62,
500
900
300
000
000
600
(77
(71
(67
(67
,700)
,500)
,600)
,200)
(65,000)
(69
,000)
* Based on a current control level of 2.4 g/Kg ADP (4.8 Ib/T ADP).
9-4
-------
oxidation systems if these gases are burned in a separate incinerator. The
emission rates of nitrogen oxides (NO ) and sulfur dioxide (S0?) from a mill
/V
would increase by the amounts emitted from this separate incinerator. If
natural gas was fired in the incinerator at a 907 metric tons per day (1000
tons per day) mill, the NO and S09 emissions resulting are estimated to be
X £
160 and 220 kilograms.per day (350 and 480 pounds per day), respectively.
These are in comparison with a TRS reduction of 180 kilograms per day (400
pounds per day). Furthermore, if fuel oil C\% sulfur content) was used instead
of natural gas, the NO and S09 emissions resulting are estimated to be about
/\ £
380 and 1040 kilograms per day (840 and 2280 pounds per day), respectively.
Using a gas-fired or oil-fired incinerator to burn these gases is a realistic
alternative since the industry feels that burning these gases safely in a
recovery furnace has not yet been demonstrated. However, if these gases were
burned in a recovery furnace or power boiler, no increase in the S0? or NO
^ A
emissions from these sources are expected.
No increase in other pollutants is expected from burning the noncondensable
gases from the digester systems, multiple-effect evaporator system or condensate
stripper system since these gases will normally be burned in a lime kiln as
part of the normal combustion air. S02 generated should be absorbed by the
lime dust (calcium oxide) present in the kiln. The scrubbers used on most
lime kiln systems also are effective gas removal system. Very little S02 is
emitted from the kiln system for this reason.
9.1.3 Atmospheric Dispersion of TRS Emissions
A dispersion analysis was made on model kraft pulp mills to evaluate the
impact of the various control techniques and retrofit systems on ground-level
TRS concentration downwind of a kraft pulp mill. The models chosen were of
average design and layout as shown i-n Figure 9-1, and included the eight
9-5
-------
FIGURE 9-1. Typical Plant Layout (lOOD ton per day kraft pulp mill)
I
CM
H-50
4 H
t
o
Cl
U1
T
TREATMENT
POND
60 ACRES
« 180
H-50
)00
=1
in
U
«
T 1
'
100 *
e
r
176 1
H-50 r
( C J
© "
n-n /T\ x-x
ni © ©._,._
A' f C M flfl M n i
^^(TVas-HV-EO-^-*^^'0 s.
70 | "^^ (?) (°) i 'f
k< " j»0fl - '-> f »Q" /"?\ ^ ^ ^"^ "
L I § T o J[ C^)* 12° O '^
jii i3a°>o75p^2o>^"35"x^
t 175 / T Y12/ E ^1
S H-60 \ J \ J
T V_y ^-^ Stacks:
3 i* - 25 >
t ^
g 1 . Recovery Furnnce
01 A 2. Smelt Oissolvinq
I Tank
_ . .. j T 3. Lime Ki ; i
Ja -^ i / ^ i
1^- X N 5" * ... _=f", " '!-, r,-'«-|pv-c-,
\_^y \ y H-50 ^ '.".Vupora tor.';
Q)*30*CD^- 4 © t 6. Brown Stock Hashers
= _ s~^ 30r7 J, ,,, S l*-2S-Ol6,v j, 7. DxirlaHnn Svcitnin
CD 1AJ "V" ^ To WT 8- Conclensate Strionsr
00 § T ' " V ^)i
1 °" u
' X~X <-± . -211 *i
W © JL
TANK0-> A - 40' F 50*
B 35' G 15'
C CO1 H 10'
0-35' J 12'
E 60' K 20'
-------
affected facilities being considered. Modeling was performed for mills of
500, 1000, and 1500 tons per day of air-dried pulp (ADP) produced, a range
within which the majority of kraft pulp mill capacities fall.
Maximum ground-level concentrations of TRS were determined for the emission
rates corresponding to each control technique and system. The concentrations
decreased predictably with decreases in the emission rates. It was possible
to adjust the meteorological conditions of the study to achieve the worst cases
that would be expected to occur at and near a kraft pulp mill.
Ambient concentrations of TRS due to the alternative control techniques
and systems were calculated using state-of-the-art modeling techniques. These
calculations are assumed to be reliable within about a factor of two. The
following assumptions were applied for the analytical approach:
1. There are no significant seasonal or hourly variations in emission
rates for these mills.
2. The mills are located in flat or gently rolling terrain.
3. The meteorological regime is unfavorable to the dispersion of effluents.
This assumption introduces an element of conservatism into the analysis.
Calculations were performed assuming the presence of aerodynamic downwash
effects on the emissions. Unfavorable design characteristics of the model
mill such as: (1) a 220-foot structure adjacent to a 250-foot recovery furnace
stack; (2) a 175-foot smelt dissolving tank stack next to a 175-foot building;
and (3) a two-foot stack for the black liquor oxidation tank atop a 50-foot
building will result in downwash in most situations. Maximum ground-level
concentrations were estimated by assuming worst meteorological conditions.
The correlation of those estimates with observed concentrations at any particular
kraft pulp mill would depend upon many factors, including the accuracy of the
emission data, the mill configuration, the distance from the mill at which
9-7
-------
samples are obtained, the sampling period and the climatology of the mill
location.
The estimated maximum ambient TRS concentration (10 second average)
in a vicinity of a 907 megagrams (1000 tons) per day pulp mill resulting from
the individual affected facilities v»ith and without controls are listed in
Table 9-3. The maximum concentrations occur at 300 meters from the source.
Table 9-3 shows that the sources (excluding the condensate strippers) resulting
in the greatest impact on ambient concentrations of TRS are, in decreasing
order, the digester systems, multiple-effect evaporator systems, recovery
furnace, and the lime kiln. An uncontrolled digester system can result in a
maximum ambient TRS concentration of 20,000 yg/m whereas an uncontrolled
brown stock washer system results in a maximum ambient concentration of 370
3
yg/m . Table 9-3 also shows the percent reductions of applying each control
technique on uncontrolled levels and the ambient levels at various distances
under the controlled case.
Tables similar to Table 9-3 showing the impact of applying controls to
the various TRS sources on ambient TRS concentrations in terms of one-hour and
24-hour averages and for 454 and 1350 megagrams (500 and 1500 tons) per day
kraft pulp mills are included in Appendix C. For the stacks of each mill,
all averaging period maximum concentrations are noted at extremely close-in
distances (300 meters). This is due to considerable aerodynamic effect
influencing the plume rise in each case. The distances given in the tables are
distances from the stack in question. Concentrations closer to the stack than
the 300 meters given may be even higher. These tables also give an estimate of
the frequency of occurrence for the maximum ambient concentration due to each
source. The TRS concentrations with low frequencies of occurrence are the
9-8
-------
TABLE 9-3
IMPACT OF CONTROLLING THE VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 907 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
Frequency -
.% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving
Tank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1) Process Controls
2) Process Controls +
High Eff. Mud Washing
3) Process Controls * High
Eff. Mud Washing +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Incineration
Uncontrolled Distance
Leyel ? 0.3 km 0.3 0.6
8.030
8,030
20,000
20,000
3,750
3,750
800
800
800
370
310
310
560
14,000
14,000
320 210
80 50
15,290 5580
Ob
310 200
Ob
200 70
50 20
25 10
25
60 20
DC
70
7,000
Ob
from Source (km) IGreater than 1/2
1.0 1.5 2.0 the Maximum
120 50
30 13
3550
120
55 45 40
15 11 10
76 5
9 6
95 3
26 16
975 525
28
28
3
.
4
1
1
1
34
25
3
25
Percent
Reduction
96.0
99.0
23.5
100.0
91.7
100.0
75.0
93.8
96.9
93.2
80.6
100.0
87.5
50.0
100.0
a Reduction from uncontrolled average level
b Gases are assumed burned in the lime kiln. The levels from the lime kiln
include unburned TRS portion of these oases.
oases.
-------
10-second and 1-hour concentrations from the smelt dissolving tank, the
digesters, and the multiple-effect evaporators, along with all three averaging
period concentrations from the lime kiln. In each case, less than 5 percent
of the averages during the year were above half the maximum value for the
respective averaging periods. These maxima, then, appear to be caused by
conditions of usually high wind speed which bring about aerodynamic downwash.
Table 9-4 shows the estimated maximum ambient TRS concentrations resulting
from the various control systems. If System No. 1 (best available control
technology) was applied to each source, the estimated maximum ambient TRS
concentration would be 97 micrograms per cubic meter (10-second average).
Control Systems No. 2 and No. 6 would reduce the average ambient TRS concentration
around a kraft mill to about 308 yg/m (10-second average). Application of
Control Systems No. 3 and No. 5 would result in a TRS ambient concentration of
o
about 487 ug/m (10-second average). These concentrations are mainly caused
by emissions from three facilities: the recovery furnace, the smelt dissolving
tank, and the brown stock washer system. Contribution due to emissions from the
lime kiln and black liquor oxidation system are negligible in all cases. No
values are reported for the digesters, multiple-effect evaporators and
condensate strippers since it is assumed that the gases from these systems
would be burned in the lime kiln.
Averaging times of 10 seconds, one hour, and twenty-four hours were
selected for the TRS calculations, representing short and long-term exposures.
The 10-second average would be considered a "whiff", and applicable to the
study of odorous emissions. The one-hour average gives an indication of the
level of exposure experienced through casual contact, while the 24-hour average
shows the level of exposure of a person living near the mill.
9-10
-------
TABLE 9-4. ESTIMATED IMPACT OF THE CONTROL SYSTEMS ON MAXIMUM AMBIENT TRS LEVELS
AROUND AN EXISTING (907 MEGAGRAMS PER DAY) KRAFT PULP MILL
Control
System
1
2
3
4
5
6
Averaging
Time
10 sec
1 hr
24 hr
10 sec
1 hr
24 hr
10 sec
1 hr
24 hr
10 sec
1 hr
24 hr
10 sec
1 hr
24 hr
10 sec
1 hr
24 hr
Maximum
Combined
Concentration
uq/m3
97
30
7
308
95
24
487
150
38
487
150
38
487
150
38
308
95
24
RF
81
25
6
260
25
6
260
80
20
260
80
20
260
80
20
81
25
6
CONTRIBUTION
SDT
16
5
1
16
5
1
16
5
1
16
5
1
16
5
1
16
5
1
OF EACH SOURCE
LK
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
(yg/m3)
BLO
._
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
Neg.
m:
~-
211
65
17
211
65
17
211
65
17
211
65
17
211
65
17
-------
9.1.4 Changes in Solid and Liquid Wastes
Increased control of gaseous TRS compounds will not change the amount of
solid waste generated by the kraft pulp industry since none of the control
techniques result in collecting solids that can not be recycled to the process.
Water effluent from a mill may increase, however, due to the various TRS
controls. Controls requiring use of fresh water instead of contaminated
condensate will result in an increase in the mill effluent of the amount of
the condensate. This increase could be eliminated by using a condensate
stripper and reusing the stripped water. A condensate stripper would also
prevent the TRS dissolved in the condensate from being emitted from the treat-
ment pond during aeration. Increasing the mud washing efficiency to control
TRS emissions from the lime kiln can also increase the mill's water effluent.
However, this additional effluent from the mud washer can probably be recycled
back to the process.
9.1.5 Energy Consumption
The energy (fuel or electricity) required for each of the control techniques
mentioned in Chapter 6 are listed in Table 9-5. The additional emissions
resulting from a coal-fired power plant supplying the necessary power (electricity)
for these control techniques are also listed in Table 9-5.
As indicated in Table 9-5, the additional particulate, SCL and NO emissions
£. A
that will occur at a coal-fired power plant due to producing the electricity
that will be required to control emissions is small compared to the TRS
redu .ion that will be achieved at the kraft mill.
As indicated by Table 9-5, the only control techniques requiring additional
fuel consumption at a kraft mill are incineration (in a separate incinerator) of
the vent gases from the brown stock washer system and the black liquor oxidation
system, and process controls used on the lime kiln. Incineration of the
9-12
-------
VO
I
to
TABLE 9-5
COMPARISON OF CONTROL TECHNIQUES ON ENERGY IMPACT
FOR A 907 MEGA6RAMS PER DAY KRAFT PULP MILL
Fuel
Source
Recovery Furnace
Digester System
.
Multiple-effect
Evaporator System
Lime Kiln
Brown Stock Washer
System
Slack Liquor Oxi-
dation System
Smelt Dissolving*1
Tank
Condensate Stripper
System
Control
Technique
BLOb
Non -con tact
evaporator
Scrubbing
Incineration
Scrubbing
Incineration
Process Controls
Process Controls &
High Eff. Mud
Washing
Process Controls,
High Eff. Mud
Washing & Caustic
Scrubbing
Incineration
Incineration
Oxygen
Fresh Water
Incineration
o Requirement
10* J/day (10° Btu/day)
0
1150
0
0
142
142
142
2340
0
0
0
(0)
(1090)
(0)
(0)
(135)
(135)
(135)
(2220)
(0)
(0)
(0)
Electrical
Total Energy
Requirement Q Requirement
Kwh/day 10 J/day (10° Btu/day)
14,400
0
2,880
2,880
INCLUDED
1 ,075*
1,075C
1 ,075C
5.376
INCLUDED
20,000
2,400
840
52
1150
11
11
IN DIGESTER
147
147
147
2360
IN WASHER
80
8
3
(49)
(1090)
(10)
(10)
SECTION
(139)
(139)
(139)
(2240)
SECTION
(68)
(8)
(3)
Emissions Ib/day
Additional Emissions from Coal
-Fired Power Plant Supolyina
The Electrical Energy
Part. NO.,
5.7 (12.6)
0 (0)
1.1 (2.5)
1.1 (2.5)
0.5 (1.0)
0.5 (1.0)
0.5 (1.0)
2.1 (4.7)
8.0 (17.5)
1.0 (2.1)
0.3 (0.7)
40 (88)
0 (0)
8 (18)
8 (18)
3 (7)
3 (7)
3 (7)
15 (33)
55 (122)
7 (15)
2 (5)
TRS
a Reduction from
SO, Control Technique
68 (150)
0 (0)
14 (30)
14 (30)
5 (12)
5 (12)
5 (12)
25 (56)
95 (210)
11 (25)
4 (8)
650
650
180
680
420
450
270
340
350
140
45
45
80
910
(1440)
(1440)
(400)
(1500)
(920)
(1000)
(600)
(750)
(775)
(300)
(100)
(100)
(175)
(2000)
a Data are based on the new source performance standards for coal-fired power plants (Part. - 0.1 lb/10 Btu; NO. - 0.7 lb/10 Btu;
S02 - 1.2 lb/106 Btu). *
Requirements are for two-stage oxidation.
c Electrical requirement is for operating a condensate stripper to TRS from scrubbing water, if contaminated condensate Is used.
d Requirements are for a scrubbing system, 1f a scrubber is not already used to control partlcuUte'emissions.
-------
noncondensable gases (digester, multiple-effect evaporator, and condensate
stripper) would not require additional fuel if they are burned in the lime kiln
as part of the primary air feed.
Incineration of the vent gases from the brown stock washers and black
liquor oxidation system would require an additional fuel consumption of 2340
X 109 joules/day (1,750 million Btu/day).
It is estimated that an additional 142 X 10 joules/day (100 million Btu/day)
of fuel will be required when process controls (higher cold end temperatures
and higher oxygen levels) are used to control TRS compounds from a lime kiln.
This is approximately five percent of the normal fuel consumption of lime kiln.
The additional electrical energy needed for each of these control tech-
niques is estimated to be between zero and 15,000 kilowatt-hours per day.
Control System 1 would reauire about 23,500 kilowatt-hours per day of additional
electrical energy. Control Systems 2 through 6 would require about 18,125
kilowatt-hours per day of additional electrical energy. An additional 350
kilowatt-hour per day would be required for each system if a condensate stripper
and a scrubber for the smelt dissolving tank are needed.
Each control system would result in an additional fuel requirement of 142
g
X 10 joules/day (for lime kiln controls) except for Control System No. 1,
q
which would result in an additional fuel requirement of 2482 X 10 joules/day
(incineration of BLO and washer gases). A pulp and paper mill requires an
electrical requirement in the order of 700 to 1400 kilowatt-hours per ton of
2
product. Therefore, these control systems will result in an increase of
between one to three percent of the total mill electrical usage.
9-14
-------
REFERENCES FOR CHAPTER 9
1. Incineration of Malodorous, Gases in Kraft Pulp Mills. Burgess, T. L.,
Cater, D. N., and Mcl^chern, D. E. Pulp and Paper Magazine of Canada.
Volume 75, Number 5. May 1974.
2. Energy and Air Emissions in the Pulp and Paper Industry. James E.
J. E. Sirrine Company. Greenville, South Carolina.
9-15
-------
10. EMISSION GUIDELINES FOR EXISTING
KRAFT PULP MILLS
Various alternative control systems can be applied to exir^'ng ' , ?ft pulp
mills as described in Chapter 6. This chapter will select a system which is
judged to be the best for existing plants when costs are taken into account,
and will specify emission limitations that reflect the application of such a
system. Time requirements to incorporate control techniques for each affected
facility are discussed in Section 6.3. Section 10.3 will briefly discuss why
the other control systems were not selected as best retrofit technology.
10.1 GENERAL RATIONALE
The best retrofit control technologies for the reduction of TRS emissions,
taking into account the cost of this control, correspond to alternative control
system No. 4, as indicated in Table 6-2. The recommended control technologies
for brown stock washers, lime kilns, and black liquor oxidation systems are
less restrictive than those that have been proposed by EPA for new kraft pulp
mills. The recommended control technologies for the recovery furnace, digesters,
multiple effect evaporators, smelt dissolving tank, and condensate stripper
are the same for both new and existing sources. The following factors were
considered in determining best retrofit control technology:
1. The degree of emission reduction achievable through the application
of various demonstrated control technologies.
2. The technical feasibility of applying the various demonstrated tech-
nologies to existing sources. In particular, more than one basic design of
existing recovery furnacs was evaluated.
10-1
-------
3. The impact of the various control technologies on national energy
consumption, water pollution, solid waste disposal, and ambient air concentrations
of TRS.
4. The cost of adopting the emission guidelines. Control costs were
estimated for each alternative control system for each retrofit model, taking
into account the level of existing controls.
Identification of the best demonstrated control technology for new mills
was accomplished during the development of NSPS for the kraft pulp industry.
A question that must be answered by this study is whether or not it is technically
and economically feasible to apply this technology to existing sources. Where
this is not feasible, best retrofit technology considering cost is identified.
Evaluation of the technical problems and costs associated with a retrofit
project is complicated by the lack of actual data for some sources. For example,
only recently has an existing brown stock washer system and black oxidation system
been retrofitted for control of TRS. Also, no new black liquor oxidation
units have been installed with control systems. Retrofit information on
control systems was available for the other process facilities in existing mills.
Retrofit models were developed (see Section 6.2) to evaluate the environmental and
cost impacts of installing TRS controls on existing recovery furnaces, digesters,
multiple-effect evaporators, lime kilns, brown stock washers, black liquor
oxidation systems, smelt dissolving tanks, and condensate strippers. The
retrofit model approach presents the impacts on an entire kraft pulp mill of
applying control technologies to individual sources of TRS. The major tech-
nical problem, aside from space limitations, foreseen for the average mill is
the ability of existing furnaces to maintain good combustion for TRS control
10-2
-------
while burning the vent gases from the pulp washer and the black liquor oxi-
dation system.
Table 10-1 indicates the impact on annual TRS emissions from the kraft
industry if best retrofit control technology, (i.e. alternative control system
No. 4) was used. Adoption of best retrofit control technology would result in
emission reductions ranging from 40 percent at typically controlled mills
to 95 percent at uncontrolled mills. Total emissions from the industry would
be reduced by about 81 percent, resulting in a national TRS reduction of about
60,900 megagrams per year (67,150 tons per year).
Adoption of best retrofit control technology will result in a maximum
reduction of 95 percent in ambient air concentrations at uncontrolled mills.
Emission reductions, and likewise control costs, will be less for mills which
have already installed some control systems.
10.2 SELECTION OF BEST RETROFIT TECHNOLOGY AND EMISSION GUIDELINE
10.2.1 Recovery Furnace System
Emission Guideline - "Old Design" furnaces (i.e., furnaces without membrane
wall or welded wall construction, or emission-control designed air systems):
20 ppm of TRS as hLS (0.3 g/Kg ADP) on a dry gas basis and as a 4-hour average,
corrected to 8 volume percent oxygen.
- "New Design" furnaces (i.e., furnaces with both membrane wall or welded
wall construction and emission-control designed air systems): 5 ppm of TRS as
H~S (0.075 g/Kg ADP) on a dry gas basis and as a 4-hour average, corrected to 8
volume percent oxygen. (A "New Design" furnace will have stated in its contract a
TRS performance guarantee or that it was designed with air pollution control as an
objective.)
- Cross recovery furnaces (i.e., furnaces with green liquor sulfiditi.es in
excess of 28 percent and liquor mixtures of more than 7 percent NSSC on an air dry
ton basis): 25 ppm of TRS as H2S (0.6 q/Kg ADP) on a dry gas basis and as a 4-hour
average, corrected to 8 volume percent oxygen.
10-3
-------
Table 10-1. BEST RETROFIT"CONTROL TECHNOLOGY AND IMPACT FOR INDIVIDUAL
SOURCES IN THE EXISTING KRAFT PULP INDUSTRY
o
i
Resultant TRS reduction
at an average kraft
Resultant national
Source
Recovery furnace
Digester system
Multiple-effect
Bes.t-.Dernonstrated
control technique
Process controls +
BLO
Incineration
Incineration
TRS level mill
achievable mill
20 ppm*(01d Design)
5 ppm*(New Design)
25 ppm* (Cross Recovery)
5 ppm
5 ppm
(% of total
emissions)
44.2
13,0
8.6
TRS reduction
(tons/year)
36,500
10,700
7,050
evaporator system
Lime kiln
Brown stock washer
system
Black liquor
oxidation system
Smelt dissolving tank
Condensate stripping
system
Process controls
(inc. high eff. mud
washing)
No control
20 ppm
**
No control
Fresh water
Incineration
0.0084 q/kg BLS
5 ppm
Total % reduction
11.9
0
3.7
neg.
81.4
9,800
3,100
n'eg.
67,150
*Three percent of all four-hour TRS averages above the specified level are not considered to be excess emissions.
**Two percent of all four-hour TRS averages above 20 ppm are not considered to be excess emissions.
-------
Discussion - The emission guidelines represent the levels that can be
achieved by using a two-stage black liquor oxidation system together with
good furnace operation. The two specified levels of TRS emissions for straight
kraft recovery furnaces reflect the dependence of TRS emissions on the design
of the furnace, which in turn depends on the age of the recovery furnace. While
the design of the furnace affects the TRS level that can be achieved, the
reduction of TRS emissions from the direct contact evaporator necessary to
reduce emissions to the level of the guidelines requires the use of high efficiency
black liquor oxidation systems regardless of the design of the furnace. Most
recovery furnaces constructed since 1965 are generally considered capable of
achieving 5 ppm TRS because the furnace design is basically similar to furnaces
presently being installed which can achieve 5 ppm TRS. Approximately 40 percent
of the existing recovery furnaces were constructed after 1965. Recovery furnaces
which were constructed before 1965 generally do not have the appropriate design (i.e.,
membrane or welded wall construction and flexibility of air distribution) or
instrumentation necessary for achieving 5 ppm. As confirmed by the two furnace
2 3
manufacturers, ' however, these older furnaces are generally capable of limiting
TRS emissions to 20 ppm if the furnace is properly operated, uses high efficiency
black liquor oxidation, and is not operated at an excessive production rate.
As mentioned in Chapter 6, cross recovery liquors are somewhat different
than straight kraft liquors. Consequently, TRS emissions from a cross recovery
furnace are not controllable to the same degree as are those from straight
kraft recovery furnaces. The reasons for this include higher sulfur-to-
soda ratios and lower BTU value of the liquor fired. Furthermore, the tech-
nique of using excess combustion air (high oxygen levels) to reduce TRS
emissions is of limited utility because it reportedly results in a sticky
10-5
-------
dust which will foul the precipitator and render furnace operation difficult
or impossible. Tests performed on a non-contact type cross-recovery furnace
indicate that TRS emission levels of 25 ppm (4-hour average) can be achieved
from well controlled cross recovery furnaces.4
Appendix B presents TRS emission data for straight kraft recovery furnaces
and a cross recovery furnace.
Retrofit annualized costs for installing a second stage of black liquor
oxidation are about $240,000 for a 907 megagrams/day (1,000 tons ADP/day) mill.
Retrofit costs would be double if a mill does not presently have a single
stage of oxidation. Annualized costs, including capital charges, are estimated
to be about $0.75 per ton ADP, or about 0.25 percent of the pulp price to install
a second stage of black liquor oxidation. These costs are not considered
excessive.
It appears that approximately 18 recovery furnaces may not be able to
achieve 20 ppm TRS because the furnace either does not have sufficient control
for proper combustion or is operated at an excessive production rate and cannot
supply sufficient oxygen to achieve good combustion. Studies have demonstrated
that minimum TRS emissions are not achieved unless residual oxygen content of
the flue gas is in the range of 2.5 to 4.5 percent. (Low oxygen levels due to
overloading of the furnace can exist regardless of the age of the furnace.) If
these furnaces are required to achieve the emission guideline, a new furnace
would have to be installed (at an annualized cost of about $2.3 million for a
500 tons per day furnace) to compensate for the cutback on production of an
existing furnace. Many of the recovery furnaces that would have to be replaced
are at least 20 years old. This age is near the normal life (25 years) of a
furnace, considering the compliance schedule under Section lll(d), and are
near replacement.
10-6
-------
An alternative to replacing an old furnace would be to install a scrubber
system, as mentioned in Section 6.11, which is capable of achieving less than
20 ppm TRS. A scrubber system has been installed at one mill.5 Installation
and operation of such a system is expected to be a much less expensive alternative
than replacement of the furnace.
The emission guidelines for recovery furnaces are comparable to the emission
levels which existing furnaces in Oregon and Washington are required to meet
as of July, 1975 (17.5 ppm).*5 The 17.5 ppm level represents the'level that can
be achieved by most existing recovery furnaces, and the 1983 Oregon and Washington
level of 5 ppm represents the level achievable with the newer design furnaces
and allows time for the replacement of older furnaces (non-membrane wall construction)
The estimated impact of adoption of the emission guideline on annual TRS
emissions from recovery furnaces is 33,470 megagrams per year, an 85 percent
reduction. The predicted maximum ambient air TRS concentration due to emissions
from an uncontrolled recovery furnace would decrease by 96 to 99 percent with
the recommended control technology.
10.2.2 Digester System
Emission Guideline - 5 parts per million of TRS as H2S on a dry gas basis
jindtrs^ar 4-hour average.
Discussion - This TRS level is the same as that included in the new source
performance standards for new digester systems. The 5 ppm level is achievable
by incineration of the noncondensable gases. Existing mills in Oregon, Washington,
and several other states are required to incinerate the noncondensable gases
from digester systems as of July, 1975.' It is estimated that adoption of this
control technology will result in a reduction of 99 percent in the uncontrolled
TRS emitted from a digester system.
10-7
-------
The TRS level achievable by incineration of noncondensable gases from
digester systems has been well-demonstrated as reported in Section 6.1.2. The
gases from the digester system can be handled in the lime kiln as part of the
combustion air without requiring extensive modification to the digester system
or lime kiln. Incineration of the gases in lime kilns or in power boilers
is presently being accomplished by at least 60 mills. Nearly all of these
incineration systems were retrofitted to the existing mills.
Incineration is the only control option capable of providing high efficiency
TRS reduction. A thousand-fold increase in emissions to approximately 7000 pom
would result from control by white liquor scrubbers (see Chapter 6). These
scrubbers are effective in controlling H?S and methyl mercaptan which comnrise
only approximately 20 percent of the TRS emissions from digester systems.
If the emission guidelines were increased moderately, incineration costs
would not vary greatly. The cost of collecting and burning the gases in the
lime kiln is essentially fixed regardless of the selected emission level. Most
existing kraft pulp mills incinerate these gases in the lime kiln and normal
kiln operation will oxidize the TRS compounds to less than 5 opm.
Retrofit annualized costs are estimated to range from about $65,000 to
about $210,000 for a 454 megagram mill, or about $0.40 to $1.25/T ADP. The
low value represents costs for piping only, while the high value represents
costs for piping, blow heat recovery system, and a separate incinerator. These
costs are not considered excessive.
The estimated impact of adoption of best retrofit control technology on
annual TRS emissions from digester systems is significant, 11,800 meaaarams per
year or a 97 percent reduction from uncontrolled levels.
10-8
-------
10.2.3 Multiple-Effect Evaporator System
Emission Guideline - 5 parts per million of TRS as HgS on a dry gas basis
and as a 4-hour average.
Discussion - This TRS level is also the same as that in the new source
performance standards for new multiple-effect evaporator systems. It is estimated
that achievement of this level will require a reduction of 98 percent of the TRS
emitted from an uncontrolled multiple-effect evaporator system. Incineration
is capable of achieving this level. Existing mills in Oregon, Washington,
and several other States are required to incinerate these gases as of July, 1975.8
The TRS level achievable by incineration has been well-demonstrated as reported
in the Standards Support and Environmental Impact Statement document for new
kraft pulp mills. The non-condensable gases from the multiple-effect evaporators
can easily be handled in the lime kiln as part of the combustion air without
requiring extensive modifications to be made to the multiple-effect evaporator
system or the lime kiln. Incineration of these gases in lime kilns or in power
boilers is presently being accomplished by at least 59 mills. The majority of
these incineration systems were retrofitted to existing multiple-effect evaporator systems,
Incineration is the only control option capable of providing high efficiency
TRS reduction. A sixty-fold increase in TRS emissions to approximately 300 ppm
(see Section 6.1.3) would be required to allow the use of white liquor scrubbers. These
scrubbers have only about a 90 percent TRS collection efficiency when used on the
noncondensable gases from a multiple-effect evaporator system.
If the emission guidelines were increased moderately, incineration costs
would not vary greatly. The control costs are mainly for collecting and transferring
the gases to the control device whether incineration or scrubbing is practiced.
10-9
-------
Most existing kraft pulp mills incinerate the gases in the lime kiln along with
the digester gases, and normal lime kiln operations oxidize the gases to less
than 5 ppm TRS.
Retrofit costs for incineration of the noncondensable gases from the
multiple-effect evaporators are included in the retrofit costs reported for
the digester system (Section 10.2.2).
Estimated impact of adoption of best retrofit technology on annual TRS
emissions from multiple-effect evaporator systems is significant, 6,120
megagrams (6,750 tons) per year or a 96 percent reduction.
10.2.4 Lime Kiln
Emission Guideline - 20 ppm of TRS as ^S on a dry gas basis and as a
4-hour average, corrected to 10 volume percent oxygen.
Discussion - The specified level reflects the dependence of TRS emissions
on the operation of the kiln. This requires maintaining the proper oxygen
level and cold-end temperature, and using water that does not contain dissolved
sulfides in the particulate control scrubber. Existing mills will probably
need to improve their lime mud washing efficiency (additional filtration and
clarifier capacity) to reduce the sulfide level of mud fed to the kiln. Additional
fan capacity may be necessary to obtain the required oxygen levels in existing
kilns and thereby provide appropriate control over the combustion. There are
no apparent reasons why these changes cannot be made to existing kilns. Furthermore,
installation of a condensate stripper may be required to remove sulfides from
the condensate if it is used in the particulate control scrubber. Appendix B
presents TRS emission data, which were obtained during the NSPS program, for
several lime kilns that achieve this level.
10-10
-------
Retrofit annualized costs for additional fan capacity (to achieve higher
oxygen levels) and instrumentation are about $55,000 for a 458 megagrams/day
(500 ton ADP/day) mill. Retrofit annualized costs for additional mud washing
capacity are about $90,000. An additional $90,000 in retrofit annualized costs would be
incurred if a condensate stripper is needed to remove the sulfides from the
scrubbing water. These annualized costs, including capital charges, are estimated
to be about $0.90 per ton ADP if a condensate stripper is not needed and about
$1.50 per ton ADP if one is needed. These costs are not considered excessive.
The impact of adoption of best demonstrated retrofit control technology
on TRS emissions from kraft lime kilns is significant, an 84 percent reduction
(9,800 tons/year) from existing levels. Maximum ambient TRS concentration due
to an uncontrolled lime kiln would be reduced by 83 percent.
Lower TRS levels than the emission guideline are achievable as stated
in Section 6.1.4 and as reflected in the proposed standard for new lime kilns
(8 ppm TRS). The lower TRS level is achievable with the addition of caustic
scrubbing.
Many existing lime kilns are operating in excess of design capacity,
and some of these kilns, even with improved.mud washing efficiency, may not
be able to achieve TRS levels significantly lower than 40 ppm because of the
inability to supply sufficient oxygen for good combustion. It appears that
between 20 and 33 lime kilns (corresponding to about 20 percent of the existing
kilns) would have to be replaced or added in order to achieve 20 ppm TRS by applying
the best retrofit control technology discussed above (see_Chapter 8). Capital
costs for a new lime kiln are $3 million, and annualized costs are $510,000.
The lower TRS emission level of 8 ppm is not recommended as an emission guideline
so that the number of kiln replacements is minimized. The higher level also allows some
of those mills which cannot achieve 20 ppm from the lime kiln by applying process
10-11
-------
controls and improved mud washing to apply caustic scrubbing to achieve the guideline
rather than replacing or adding a new lime kiln. The results of trials conducted at one
pulp mill showed that levels of 40 to 50 ppm TRS could be reduced to a level
less than 20 ppm by using caustic addition.9 Nevertheless, the lower TRS level
is technically achievable at existing mills and can be imposed if the location of
the mill or lime kiln warrants additional controls.
10.2.5 Brown Stock Washer System
Emission Guideline - No emission guideline is recommended for existing
brown stock washer systems.
Discussion - No TRS control is recommended due to the high costs associated
with hooding and collecting the gases and the possible effect the gases may have
on existing recovery furnace operation.
Incineration of the vent gases is the emission control technique that could
be used to reduce TRS emissions from brown stock washer systems. Burning
these gases in an existing recovery furnace is considered by furnace manufacturers
to be technologically feasible. This control technique, however, has not yet
been demonstrated on an existing furnace, and the TRS level that can be achieved
from an existing furnace under these conditions has not been demonstrated. The
control costs for incineration, therefore, have been based on the use of a separate
incinerator. Incineration of the gases would require that the washer be hooded,
possibly with enclosed hoods, and ductwork would be necessary to transfer the
gases to the incinerator. (These gases would have to be ducted over 1500 feet
at some mills if the recovery furnace was used.) Incineration of gases in a
separate incinerator would require retrofit annualized costs of about $900,000
for a 454 meqaqram mill or about $5.50/T ADP. These costs are much more
10-12
-------
severe than retrofit costs for the other TRS sources and are considered to be
excessive in comparison with control of the other sources of TRS and with the
amount of TRS reduction achieved (about 1 percent of total mill TRS emissions).
10.2.6 Black Liquor Oxidation System
Emission Guideline - No emission guideline is recommended for existing
black liquor oxidation systems.
Discussion - No TRS control is recommended due to the expected cost impact
on the industry if existing sources were required to meet TRS levels achievable
for new systems. There is no less stringent control method possible (except for the
uncontrolled level) than that considered demonstrated for new sources.
Achievable control technology involves incineration of the vent gases or
the use of molecular oxygen instead of air to eliminate the vent gases. The cost
of controlling the low concentration/high volume gases from black liquor oxidation
systems is considered more severe and excessive in comparison with controlling
the largest sources of TRS at kraft mills (see Section 10.3). The control costs
for incineration have been based on the use of a separate incinerator, since the
effect of these oxygen-deficient gases on furnace combustion and thus TRS
emissions from existing furnaces has not been determined. Retrofit annualized
costs are estimated to be $230,000 for a 500 TPD kraft mill, or $1.50/T ADP.
These costs are considered excessive in view of the amount of TRS reduction that
would be achieved by incineration (about 0.4 percent of total mill TRS emissions).
10.2.7 Condensate Stripping System
Emission Guideline - 5 parts per million of TRS as I^S on a dry gas basis
and as a 4-hour average.
10-13
-------
Discussion - This emission guideline is the same as that included in the
new source performance standards for new condensate stripping systems. Only
five existing mills have condensate strippers, and only one is not presently
incinerating the off-gases. Incineration of the off-gases is necessary to
achieve this TRS level.
Retrofit annualized costs based on combining the stripper off-gases with
the noncondensable gas from the digesters and evaporators are estimated to be
about $6,500 for a 500-ton-per-day mill or about $0.05/T ADP. The cost impact
on the industry due to control of this facility is expected to be negligible.
Use of a white liquor scrubber, the only other control technique used,
would permit TRS emissions which are 100-fold higher than with incineration.
These TRS levels from scrubbers could be highly odorous.
10.2.8 Smelt Dissolving Tank
Emission Guideline - 0.0084 g/kg BLS of TRS as H^S (approximately 8 ppm),
on a 4-hour average.
Discussion - This emission guideline is also the same as that included in
the new source performance standards for new smelt dissolving tanks. Achievement
of this level would require the use of fresh water, or possibly weak wash liquor,
in the particulate control device (scrubber) to ensure compliance.
The control costs for achieving this level are not considered excessive.
Adoption of this level is expected to result in an emission reduction of about
2720 megagrams (3000 tons) per year of TRS.
10.2.9 Excess Emissions
Excess emissions are defined in terms of the applicable guidelines. For
example, if the emission guideline for a particular facility is 5 ppm of TRS,
4-hour average, then excess emissions would usually be defined as all occurrences
10-14
-------
during the reporting period for which 5 ppm TRS, 4-hour average, was exceeded.
In some special cases where emissions in excess of the nominal guideline can
be predicted to normally occur at a well operated facility for a small percentage
of the time, this is reflected in the definition of excess emissions. The
definitions of excess emissions for recovery furnaces and lime kilns are
discussed below.
Recovery Furnace Systems - A pulp manufacturer submitted six months of
TRS emission data from one of their recovery furnaces and requested that EPA
consider the data in defining excess TRS emissions fro recovery furnace facilities.
The furnace was tested by EPA in developing the data upon which the new source
performance standard is based. The submitted data, recorded by a continuous
monitor, show that over the 6-month period, the percent of time that the TRS
concentration exceeded 5 ppm during each month ranged from 0.7 to 7.7 percent
and averaged about 3 percent.
EPA has investigated the furnace operation and the monitoring system at
this mill and believes that the data are a true indication of normal, well
controlled operation for this furnace. Therefore, based on the information,
an allowance of 3 percent of the 4-hour averages has been given for excess
TRS emissions above the guideline.
Lime Kiln - Test data (see Appendix B) were supplied by a mill (Lime Kiln P)
that had retrofitted the lime kiln system with additional fan capacity and mud
washing capacity. These data give an indication of the variations in the emission
concentrations over a large number of four-hour periods. The data show that
for the period when the mill was maintaining good process controls (high cold
end temperatures, high oxygen levels, and high mud solids contents) on the kiln,
the four-hour average TRS concentrations exceeded 20 ppm for approximately
11 percent of the time.11 However, during this same period the mud filter
10-15
-------
(belt filter) was inoperative for 10 percent of the time. Process and
emission monitoring data obtained on Lime Kiln E (see Appendix B) show
excess TRS emissions of 2 percent with down time on the mud filter (vacuum
drum) of only 1 percent. Therefore, it is felt that with a reliable mud
filtering system and maintaining good process controls on the kiln, the
4-hour average TRS concentrations will exceed 20 ppm for approximately
2 percent of the time. Hence, an allowance of a maximum 2 percent of the
4-hour averages has been provided for excess TRS emissions above the guideline.
10.3 SUMMARY OF THE RATIONALE FOR SELECTING THE BEST RETROFIT CONTROL SYSTEM
The proposed TRS emission limits for new kraft pulp mills are technologically
achievable at existing kraft pulp mills when the best control techniques discussed
above are applied to each of the eight component process operations. However,
the costs of applying the best control techniques are considered excessive for
some existing mills, in part because some techniques involve replacement of
recovery furnaces or lime kilns. Further, alternative control techniques
which are effective but less costly are available for some process operations.
Therefore, the costs of applying the various control techniques had a considerable
influence on the selection of the recommended best retrofit control technology
(alternative control system No. 4 for an entire kraft mill).
Control of the brown stock washer system and black liquor oxidation system
(alternative control system No. 1) are not recommended because incineration of
these vent gases in a separate incinerator would result in excessive operating
costs and fuel requirements in comparison to the TRS reduction achieved by the
control technique. Incineration of these gases in an existing recovery furnace
is not presently considered to be demonstrated retrofit technology. No existing
recovery furnace not designed to handle these gases has demonstrated the ability
to burn these gases and still maintain proper combustion for controlling TRS
emissions from the furnace itself.
10-16
-------
The emission guideline recommended for existing recovery furnaces is
20 ppm for "old design" furnaces, 5 ppm for "new design" furnaces, and 25 ppm
for cross recovery furnaces. The older furnaces are not capable of achieving
5 ppm and a large number of existing furnaces would most likely have to be
replaced if such a level was required. The control technique required for each
type of furnace to meet the recommended levels is two-stage black liquor oxidation
and process controls.
Incineration of the noncondensable gases from the digesters, multiple-
effect evaporators, or condensate strippers in the lime kiln has been demonstrated
at many existing mills. Therefore, since the control costs are not excessive,
the emission guideline recommended is the same as the new source performance
standard (5 ppm TRS) for new kraft pulp mills.
An emission guideline of 20 ppm TRS is recommended for existing lime kilns.
Emission data obtained during the NSPS program show that 2Q ppm can be Achieved
with proper kiln operation and sufficient mud washing efficiency. Larger fans
and additional mud washing capacity will be necessary for most existing kilns.
Lower TRS levels are achievable, but several additional lime kilns would have
to be replaced or added in order to achieve a level of 8 ppm TRS.
The emission guideline recommended for smelt dissolving tanks will probably
prevent the use of contaminated condensate in the tank and the particulate
control device, if one is used. If a scrubber is not used already for controlling
particulates, one may have to be installed to reduce TRS emissions from an
existing smelt dissolving tank to the recommended guideline.
The best retrofit technologies (alternative control system No. 4) will
produce a large reduction in national TRS emissions (67,150 tons/year) and
in ambient TRS concentrations around existing mills.
10-17
-------
10.4 SELECTION OF THE FORMAT OF THE EMISSION GUIDELINES
Standards for kraft pulp mills could be expressed in terms of either mass
emissions per unit of production or a concentration of pollutant in the effluent
gases. The most common format now used by the industry and state control
agencies is pounds of pollutant per ton of air-dried unbleached pulp produced
(Ib/T ADP). This format offers the advantage of preventing circumvention
of the standards by the addition of dilution air or the use of excessive
quantities of air in process operations. The principal disadvantage is that
a control agency cannot readily or accurately measure the pulp production
over the short term. Due to storage capacity of the mill, the recovery
furnace, smelt dissolving tank, lime kiln, condensate strippers, black liquor
oxidation tanks, and multiple-effect evaporators can be operating on
accumulated inventories when the digesters are off-stream (no pulp production).
Similarly, the above facilities can be operating below capacity even though
the pulp production may be at design rates.
Concentration units are used as the format for the emission guidelines
for the digesters, the multiple-effect evaporators, the recovery furnace, the
lime kiln, and the condensate stripping system. The reasons for the selection
of this format are outlined below:
a. Concentration units can be corrected for excess oxygen in the lime
kiln and recovery furnace exhaust streams, precluding circumvention of the
standards by dilution.
b. The reference test method for TRS produces data in concentration units.
No conversion factors are therefore required in determining compliance for the
affected facilities.
c. Average concentrations rather than instantaneous concentrations are
proposed to allow for fluctuations in emissions which occur even during periods
of normal operation.
10-18
-------
d. Commercially available continuous monitors that may be used to measure
emissions from these facilities indicate concentration directly. A direct
indication of performance of the control systems would be available, and therefore
the operator would be aware of excess emissions that require corrective action.
The emission guideline for smelt dissolving tanks is expressed in grams
per kilogram BLS (g/kg BLS) to prevent circumvention by dilution. EPA tests
show that gas volumes from existing smelt tanks vary in exhaust concentrations
by a factor of as much as 2.5, even when the smelt dissolving tanks have the
same mass emission rate (g/kg BLS). Dilution cannot be prevented by correcting
for excess oxygen because the exhaust stream discharged from the smelt
dissolving tank is mostly ambient air.
10.5 RECOMMENDED MONITORING REQUIREMENTS
Monitoring requirements are necessary to ensure proper operation and
maintenance of the affected facility and its associated control system. The
volume concentration of TRS emissions can be monitored by use of measurement
systems (see Chapter 7). Since there are no process or control device
parameters that are appropriate indicators of concentration of TRS emissions
from recovery furnace systems and lime kilns, it is recommended that TRS
continuous monitors be required for recovery furnaces and lime kilns.
TRS concentrations in the effluent gases from an incinerator that controls
TRS emissions (from the digesters, multiple-effect evaporators, and/or condensate
strippers) can be measured by a continuous monitoring system. An effective
alternative method of monitoring TRS emissions from an incinerator is continuous
measuring and recording of the fire box temperature of 540°C (1000°F) and operation
at a residence time of at least one-half second in the fire box. Incinerators
are designed for a particular residence time that will not be reduced if the
incinerator is not operated above its design capacity. The fire box temperature
10-19
-------
can ba readily measured and recorded. IT non con Jensub] 2 yascis r,"om facilities
that JT2 covered by the guidelines are incinerated in the recovery furnace
or the lime kiln, the TRS monitoring system on the fnrntics or the lime
kiln v/ill serve to monitor the sources that are being inr.iccrate'.l.
Since the yuideline for smalt JJssol vimj I'-vnk-s is axprrj;;-^! in a format
of pollutant mass per unit of feed to the furnjce, the rjas flow rate and the
feed rate to the furnace would have to be measured simultaneously to reduce
the TRS concentrations measured by the monitor to units of the recommended
guideline. The inaccuracies involved in continuously measuring emissions from
the smelt dissolving tank are felt to be sufficiently large that no direct
monitoring of TRS emissions from the smelt dissolving tank is recommended.
10-20
-------
APPENDIX A. SUMMARY OF KRAFT MILLS
IN THE UNITED STATES
M111
Size
Ave. Kraft
Production
i.
Alabama
*
:>
Ari zona
Arkansas
i Corppanv
Allied Paper
American Can
.Champion
Container Corp.
.Gerogia Kraft
Gulf States
Gulf States
Hammer-mi 1 1
I-P-
'Kimberly-Clark
I
"MacMillan
Sloedel
: Scott
Union Camp
Southwest Forest
^Georgia-Pacific
Great Northern
. Green Bay
' I. P.
I. P.
Weyerhaeuser
I ,..-.-_
j i
_, Location
Jackson
Bulter
Courtland
Brewton
Mahrt .
Demo polis
Tuscaloosa
. Selma
, Mobile
Coosa Pines
j
Pine Hi|l
i
Mobile [
Montgomery
Snowflake
i -
Crossett
(
Ashdov;n
1
Morrilton
Camden
Pine Bluff
Pine Bluff
(Capacity
-.: ..tpd.. ".
500
930
(900)
500
(500)
900
(850)
1000
(975)
360
(400)
500
(475)
500
(500)
1300
(1200)
585
(600)
1000
(975)
1400
(1400)
870
(930)
600
(600)
1250
400
(400)
360
1
750
(750)
1220
(1300)
200
' (200)
No. of f
1
2
3
1
2
1
1
2
1
2
2
1
4
1
2
3
1
2
3
2.
1 .
Manuf
CE
B&W
B&W
B&W
B&U
B&W
B&W
B&W
B&W
CE
CE
B&W
CE
B&W
CE
CE
CE
B&W
B&W
B&W
CE
i Rating
; tpd
566
Year
Jn'.t/Olc-:
post-1965
350 post-1%5
390 (each) 1965
600
390
600
900
330
175
250
450
700
900
932
450;300
300
700
250
500
^00&500
850
540
665
250
500
2-275
1100
390
165
1959,1956
1968
1962
1969
1965
1955
1947
1941
1970
post-1965
post-1965
pre 1965
pre 1965
1960
1969
pre 1965
pre 1965
post-1965
post-1965
1965
1966
1946
1966
1959
pre 1965
Control
BLO
BLO
BLO
BLO
Low Odor
BLO
BLO
(oxyoen)
BLO
BLO
BLO
BLO
BLO'
BLO
BLO
TKS Leve
VT yn?
0.5
0.5
0.15
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0_c
0.5
0.5
15.0
0.6
15.0
15.0
15.0
15.0
15.0
L'nc incineration
i,C Continuous
BLO Black Liquor Oxidation
A-l
-------
Li.::;% Kiln
Size TR
No. of Tons ;CaO) Lev
L'n";s Per.L'ay ,v/T_
1 121 0.
t. 1 51 0.
each
i 181 0
2 120 0
t
1 0
. 1 120 o
2 130 0
75
1 125 o
0
2
1 225
4 1400
(total)
1 1 74
2
2 400
1 117
1
1 150
ADP
8
8
05
8
8
8
8
2
8
TDC ^~
Disaster 14'
I Type1" Control \ Level
.No.- -_L(.Size_L JechniqueJ/T. ADP
4 Batch Inc. Q.02
6 B Inc. o.02
1 C Inc. o.02
(500 tpd)
8 B Inc. 0.02
1 C Inc. 0.02
1 C Inc. 0-02
(400 tpd)
8 B Inc. 0.02
(495)
4 B Inc. o 02
(600)
Inc. o.02
6 B Inc. o.02
1 _ C
1 C Inc. 0.02
(925)
13 B (910) Inc. 0.02
C (500)
1 C Inc. Q.02
(870)
68 1.5
11 B Inc. 0.02
(1276)
4 B 1.5
(400)
2 B 1-5
2 C Inc. \02
1.5
3 B '1.5
Itiple-efr'.'ct ,,.
/apordlor ' ° «
Control Level
"P.- .Technique ?/T AD:1
1 Inc. 0.02
2 inc. 0.02
1 inc. 0.02
(
2 inc. 0.02 '.
1 inc. 0.02
1 inc. 0.02
2 scrub. 0.08
1 inc. o . 02
Inc. 0.02
inc. 0.02
1 inc. 0,02
4 inc. 0.02
1 inc. , 0.02
2 ' 1.0
1
4 Inc. 0.02
1 1.0
1 1.0
1.0
KO
' 1.0
urovii Slock V/as!i£r T
Capacity l-'asher Ic
Mo. ADTPD Jtagss ,v/r
1 3 0
2 3
4
1 2
1 3
2 300 3
475 4
1 600 3
2
,
5 i ;
|
2 |
1
2
1 3
1
1 ! 3
RS
/'.:!'
A OP
.3
J
I'l'Lonti oiled fiauros yivjn are avoraije uncontrolled figures fur the industry unli'ss actual level known, cont'.rcllad levels
o:':-'n art.1 actual ti.j'.ircs when known ,.!;,.er.rise state standard is given, furnaces controlled vJitli BLO but for which no
;'.. it« st.indard r.'.iply or actual level known, ,1 level of ?..'{ ;'/T (70 DPII') is assumed. A conversion table showing the
; jvesi.'ond.inf] eiiiiiSioii rate in tsrms of p^i,! and g/Kg ADP is presented at the end of the appendix.
A-2
-------
..--.._-.. -..._ 7--.-,~-r- . ... ..-..--^-T--
\t
1 ...... Co.i:pcny ...
Cali- Crown Siripson
fornia
Fibreboercl
Louisiana-Psc.
Simpson Lee
Florida Alton Box
Container Corp.
Hudson P & P
I. P.
Proctor & Gamble
St. Joe
St. Regis
St. Regis
Sebroia Continental Can
Continental Can
Brunswick
Georgia Kraft
Georgia Krcft
Gilman
Great northern
Interstate
Itt Rayon i or
'
- - . _ .
\ i
Lccetion
Fairhcvpn
Anlie > ii
Samoa
Anderson
Jacksonville
rornandine Beach
Palatda
Panana City
Foley
Part St. Joe
Jacksonville
Pensacola
Augusta
Port Hentworth
Brunswick
Krannert
Hacon
St. Itery
Cedar Springs
Picek.r'i
Jesup
Mill
Sire
Avo. Kraft
Production
( Kt.ft )
(Capacity) :ju. of
.... tPd Units ,
550
(GW)
600
(7( j)
150
(160)
f,75
(C50)
1
2
2
2
1
1500 2
(1700) (
950' ! 3
(950) !
1400 2
(1400)
900 [ 3
(900)
1300
(1300)
1350
(1400)
(920)
800
(800)
625
(600)
1550
(1550)
1550
(1550)
900
(900)
1100
(1000)
1 7P.r/
(170ri)
525
1200
(1250)
3
3
2
2
2
2
3
2
3
2
1
3
Recovery Furnace ,,
i Ralimj Year CnntrolJTRS Leve
l-bnuf.
P&W
CM'.'
cr
BM
CF
Bf,u
B)M'r
CF
CE
B&H
CF
CE
CF
PS.V
B&W
CE
B&W
CF
CF.
CE
BM'1
CF
PF
f.M-i
Ct
i-P.'3 . ./nstnl.le;' -.::!, ni'i'jc-.
800 1964 BLO
400 1959 BLO
350{eoc!-i)prc- 1965 BLO
150 1962 BLO
300 1973 Low Odor
750 post-1965 BLO
1000 1967 Low Odor
300 1955 BL(l(Oxyae
250(each)1950ri954
1200 post-1965
900(each)post-1965 RLO
413R550 1952&1956
500 ore 1965
233*300 ore 1965
1060 post-1965
300&383 pre -1965
SOO(each) 1973 Low Orior
400(each)1959?,1964 BLO
350(each)
1100 1970 Low Odor
450 ere 1965 BLO
300R550 pre 1965
500 oost-1965
300 pre 1965
500 1968 BLO
2-275 ore 1%5
66B ere 1965 Low Odor
1000 1972 Low Odor
4bO 1965
1100 1S70 Low Odor
465&350 pre 1965 RLO
,7/T ADP
0.5
0.6
0.5
6.5
0.5
0.15
n)
0.5
0.5
0.5
0.5
0.5
0.15
0.6
15.0
0.5
15.0
15.0
15.0
0.5
15.0
0.6
A-3
-------
Li mo Kiln ' ,pc !'.;ilti;>lc- -c ffoct ., .
-. . .. Sizn TRS ^ Digester , 14 ^'-poiMlor '"*f rt-rc.M .SU>cl: ',V-:.hcr TRS
;;o. of Tons (CaO) Level < Type" Control , Level' r:onu\,l Lcvc'l ' Cfp-"-i(-v IV-ci or i ,M,nv/l
.JJni.ts |. Per .Day..! »/T AD?.
1 0.15
2 0.05
1 700 0.12
1 50 0.05
(Fluo- i
solid) i
2 80 0.8
(each)
2 . 0.8
1
3 !
3
3 240
3 280 \
3 0.
f
2
1 110 tlpd 0.8
1 70 tlpd
(Fluo-
solid)
1 100
3 440
3 113
113
113
2 80
80
1 275
2 210
210
1 111
3 144
114
212
..No. . ^ISize). Tcchnique^/T ADP !!-'. .Toc!::iic]uc ,;/] /,P!> (;0. ADTiV; SLayos :'/T A;;f
2 C inc. 0.02 \ i illc. 0.02 _>, 0.27
L.K. .! I..).. I '
4 B inc. 0-02
L.K.
1 C inc. 0.02
(700)
1 C scruj 0.6
(170)
6 B 1.5
(700)
7 B 1.5
1 C
13 B
(1000)
19 B
10 B (1300)
1 .C (5CO)
12 B
/
18 B inc. 0.02
2 C 1.5
9 B
17 B
(1550)
14 B
8 B
13 B
10 8 (1900)
1 C (340)
4 B
6 B
26 B
iii;;. 0.02 ! 2 4 0.11
L.I;. c
inc. 0.02 l 1 2 0.19
P
I L.K. jj
1 scruN 0.08
2 1.0
4 1.0
3
3
3
3 , v
/
4 inc. 0.02
2 1.0
2 :
4
4
2
\
/
2 0.08
1 . 1.0
1 0
0.12
1 3 0.3
3 0.2
4
I
4 (
4 4
2 4
f
I
A-4
-------
Georgia
Idaho
Kentucky
Louisiana
Maine
*'' 1
Company
Owens-Illinois
Union Camp
Potlatch
Western Kraft
Westvaco
Boise Cascade
Boise Cascade
Continental Can
Crown Zellerbach
Crown Zellerbach
Georgia-Pacific
I. P.
I
I. P.
,01in
Pineville
Western Kraft
. Diamond Int.
Georgia-Pacific
I. P. ;-;--
Lincoln
1 !
_'! _... Location ._
Valdosta
Savannah
Lewis ton
Hawesville
Wickliffe
DeRidder
Elizabeth
Hodge
Bogalvsa
St. Francisville
Port Hudson
Bustrop
.Springhill
West. Monroe
Pineville
Campti
Old Town
Woodland
Jay
Lincoln
Mill
Si 2'.-
Ave. Kt\ift
Production
( Kruft )
(Capacity) No. of i
. tpd Units ;
9*0
2600
(ziibo;
850
(900)
300
(320)
600
(600)
1030
(1050)
300
(325)
1400
(1400)
1340
(1350)
500
(500)
530
(640)
1100
1000
1650
1125
(1150)
3
6
4
2
1
1
1
2
2
1
2
2
4
2
800 1
(750)
450
350
(550)
800 '
(800)
600
(600)
340
(400)
1
1
2
2
1
Recovery Furnace
Mdnuf
CE
IE
CL
B&K
B&W
CE
B&W
B&W
B&W
CE
B&W
CE
B&W
B&W
CE
CE
B&H
B&W
CE
B&W
B&W
CE
B&W
B&W
B&W
C&W
CE
B&W
i Rating
: tpd
350&250
1350
150S300
300
400
225
300
833
1000
300
1233
800
350
600
690
1000
300
1100
Year
Installed
pre-1965
post-1965
pre-1965
pre-1965
1954
;:ciii,-ci3TPs
Leva
vi:niq.u i/T ADr>
BLO
BLO
BLO
BLO
2.
2.
0.
1
1
5
1970 Low Odor
1968
1974
post-1965
1968
1955
post-1965
1963
pre-1965
1963
1965
post-1965
pre-1965
1966
BLO
BLO
BLO
BLO
BLO
BLO
BLO
BLO
BLO
2-700;500 1973;66;62 BLO
350
450
800
833
420
590
pre-1965
1963
BLO
0.
0.
0.
2.
2.
2.
2.
0.
2.
0.
2.
15
5
6
1
1
1
1
6
1
6
1
1974 Low Odor
pre-1965
1972
1969
350(each) 1963
800
600
386
1974
pre-1965
1970
BLO
Low Odor
Low Odor
BLO
Low Odor
BLO
Low Odor
2.
1
0.1E
0.1E
0.5
c.r
0.1:
A-5
-------
-ims Kiln
Size TRS n
Ho. of Tons (CaO) Level'' j
Units ; Per Day.: f/j ADP ! No.
'.
3 ' |. .0.8
I
3 525 | 0.8
j
3 400 i 0.2
0.2
1 80
0
1 60 0.05
i
1 i 0.8
1 75
2 471 !
i
1 150
1 300
1 200
n.?
9
34
1
11
1
3
1
7
6
3
34
2
1
1
2
4
2
r
u 1
1 150 o |i
0.05
1 100 0.1
1
1
1
1
igsster ?
Type'
...LtS1/.e).
ti
(950)
B (1775)
C (600)
B (720)
C
B
(320)
C
(600)
B
B
(300)
C
(1650)
B (1250)
C (250)
'C
C
(660)
C
C
. 0290)
C
(840)
C
C
(600)
C
(600)
C
(400)
1 Control1
Technique:.
inc.
i nc .
inc.
inc.
inc.
inc.
inc.
inc,
inc.
Inc.
inc.
inc.
inc.
K.'l ti,i'! :-<_ i i'..c t . n,~
TRS , ! v..j <>. ;.'.!! ''J f r>ol;r Sto:.[; Washer TRS
Lcvs:4 Co-.itroi \..!-.'-\: Capacity l-'asnc,- i,w01
VT ADP.. "" Ti>ci,ii-.;i.o './. ' '}? no..... ADFP;' . StcJtjjs ;//r/l
1.5
1.5
0.02
0.02
0.02
1.5 ;
1.5 ':
0.02 '
1.5
0.02
0.02 ;
0.02
1.5
0.02 i
0.02 i
i
i
0.02
0.02
0.02
0.02
3 1.0
c i.o
4 inc. 0.02
1 inc. 0.02
1 inc. 0.02
1 1-0
T 1.0
2 inc. 0.02
4 1.0
inc. 0.02
1 inc. 0.02
i.o ;
1.0
2 inc. 0.02
1 inc. 0.02 ;
i
3 4 0.3
0.3
0.3
1 3 0.3
1 2 0.3
1
4 '
1 2
4
2 2
i,
II
0.02
1 inc. 0.02
inc. 0.02
1 inc. 0.02
n.nz
1 4 T.3
A-6
-------
Mai ne
Maryland .
Michigan
*
Minnesota
\-
Mississippi
Montana
New
Hampshire
N. York :
N.Carolina
V-,
..
Company
- | * .. ..
Oxford
S. D. Warren
Westvaco
Mead
Scott
Boise Cascade
Potlatch
I. P.
I. P. ;
I. P. ;
St. Regis
Hoerner-Waldorf
'
Brown
I. P.
Champion
Federal
Hoerner-V/alacrf
Weyerhaeuser
Weyerhaeuser
. '. ... - . _ .
'j locii.tion
Rumford
Westbrook
Luke
Escanaba
.Muskegon
i
Int'l Falls
Cloquet
1
Moss Point
,
Hatchez :
, i
Vicksburg
f
Monti cello
Missoula
i
Berlin-Gorham
Ticonderoga
Canton
Riegelwood
Roanoke Rapids
New Bern
Plymouth
Mill
Size
Ave. Kraft
Production
,-( Kraft )
(Capacity) No. of I
! tP
-------
Line Kiln
Size ..... TRS . --, ..Digester. ? .-.-,.-, ,
To.is (CEO) Level ! Type i Control
a. ot io.Ts (UJ) Level
Units L Per. Day J */"LADP_
1 120 0.8
1 90 0-8
0.8
1 220 0.05
1 70(Fluo- 0.05
i solids)
0.05
1 100 0.8
0.05
0.05
0.05
1 410 0.2
3 300 0.8
0.8
/ 0.05
2 300 0.8
2 280
1 200
1 225 j<
3 315 O.t
. NQ,._._
6
7
10
6
1
5
8
ft
2
3
8
8
1
18
11
11
7
231
lypc ! Lontroi : Level
_[Size) iTechniqueJ/T ADP
i
B inc. ! 0.02
(365)
B inc. O.Oi.'
(315)
B inc. i 0.02
B inc. 0.02
(700)
C inc. 0.02
(240)
B 1.5
B 1.5
(400)
inc. 0.02
inc. 0.02
C inc. 0.02
C inc. 0.02
(1650)
C (900) inc. 0.02
B (700)
1
- B inc. 0.02
L.K
i
C , inc. ' 0.02
B 1.5
(1250)
B 1.5
00 each)
B 1.5
(1000)
B inc. 0.02
(800)
B inc. 1 02
(1500)
(,!:! cm 1 LO'.-'I
!)0. K--:!iq-.IC_;'/T /!?)'
1 inc. 0.02
1 ii>~. 0.02 j
i i
inc. 0.02
1 inc. 0.02
1 inc. 0.02
1 1.0
1 1.0
inc. 0.02
inc. 0.02
inc. 0.02
2 inc. 0.02
4 inc. 0.02
1.0
inc. 0.02
3 scrub. 0.08
3 1.0
2 1.0
1 inc. 0.02
5 inc. LJ
Capacity washer Love:
Co. AD':rO Sxavs .r'/i A.V
2 4 0.3
1 . 3
1 3
1 3
2 3
2
2 3
! 2 0.02
3 3 0.3
3 3
1 3
A-8
-------
Ohio
Mill !
Size
Ave. Kraft
Production !
( Kraft- ) .Recovery Furnace^
(Capacity) No. of' Rating [ Yea> , Control JTRS Level".
-.-.Company. '_ _Loca,tjon. [.__tp.d_.L_l)niis_ Manuf.J_tp_d J.nstaJlec':'ec;'.rn>.;,; #/T ADP ;
: ' , I!
,3TBC-
Grief
Mead ;
-
Oklahoma
Heyerhaeuser
t !
Iregon
'American Can
'!
Boise Cascade
'Crown Z
Georgie-Pacific
,I.P. !
Western Kraft
-- Weyerhaeuser
-_
Penna. !
Appleton
P.H. Glatfelter
1
1; ,
'I
: Penntech
''
S. Caro-
lina
Bowater
'I. P.
S. Carolina
Westvaco
; '
' ''
l'.
Mass i Ion
(200)
Chillicothe COO
Valliant
Halsey
(540)
1300
(1300)
300
(300}
St. Helens '856'
(850)
Clatskmie '690'
Toledo
Gardiner
Al bany
(916)
\ * w /
1075
(1075)
600
(545)
. \+r tt* f
500
(550)
Springfield 1150"
; (1050)
|
Roaring Springs 180
180
. Spring Grove . 500
t
i (500)
i
!
Johnsonburg , 190
Catawba
; (iso)
!
\
940
(040}
Georgetown 1830
Florence
(1750)
6CO
(G75V
Charleston 2000 '
|
(1989)
"2
1
1
2
1
3
2
2
2
2
2
1
1
2
2
2
4
CE
CE
B&W
B&W
CE
B&W
CE
B&W
CE
B&W
CE
B&W
CE
B&W
CE
B&W
B&W
B&W
CE
B&W
366 pre-1965 BLO
175 .pre-1965
1500 post-1965 Low Odor
j
'
400 1967 Low Odor
450 1966 BLO
465 post-1965
800 1964 BLO
350(each)1-postl965 BLO
2-prel965
420 1972 Low Odor
420 pre-1965 BLO
600 1969 Low Odor
165 1965
800 (each) pre-1 965 ' BLO
.post-1965
122 1960
83 1950
.400&150 pre-1965 BLO
748 1969 Low Odor
160 pre-1965 BLO
600 1964 BLO
450 1957
900(each) 1966 BLO
1963
1000 1972 Low Odor
410 1962
1000 pre-1965 BlO
360 1955
250 1948
250 1945
15.0 |
fc
2.1 1
i
1
1
0.1E ', '
i
0.1E
0.5
0.5
0.1E
n.5
0.1E
0.15
15.0
2.1
2.1
2.1
0.5
!
0.15
2.1 ;
1
f
'<
\
A-9
-------
Lime Kiln
... . Size .-TRS,, .-., Dige,ster 2-- - i
No! cf Tons (CaO) Leva; < \ Type ! Control ;
Uni ts iPer .Day..; W. AD?_ L . > - --i-U-izs) - Technique
0.8
1 250 0.8
1 0.8
1 0.2
3 0.2
1 250 . 0.2
3 260 ! 0.2
1
! .0.05
1 0.2
'/.
0.2
1 . 0.8
I
1 0.05
Hue-
solid)
2 50 0.8
' 1 0.8
0.8
1 150 0.8
4 CC5 0.8
8 B inc.
(6CO)
3 C inc.
(1000)
, (500)
(100)
2 C inc.
: (300)
3 ' B inc.
2 i C
2 i C inc.
(916)
11 B (650) inc.
1 C (115)
inc.
6 B inc.
7 B (380) inc.
1 C (770)
5 B
8 B (285)
,C (250)
16 B
(170)
6 B
(940)
5 B
(625)
15 B (1COO) inc.
1 C (700)
TRS "j.
Level4
?/T..ADP_. .
1.5
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
1
| 1.5
1.5
1.5
', 1-5
i
1 1.5
1.5
0.02
i unin-ei TOL i -,...
vaporalor ll0 , 3rowr
Conlrol Lewi
No. ...Technique ?/T ;".!.'? Mo.
1.0 |
2 scrub 0.08 f
I
I
1 inc. 0.02 1
1 inc. 0.02 I 1
L.K.
2 inc. 0.02
1 inc. 0.02 I
3 inc. 0.02 I 3
! inc. 0.02 I
1 inc. 0.02 J 2
inc. 0.02 1
1.0 1
2 1.0 2
1 1.0 J 1
:
2 1.0 1
1 .0
1 1.0 2
4 inc. 0.02
Brown Stock Masher TRS
Capacity Kasher Level'1
i!o. . .. ADFPn ...Stages #/7 AD!'
0.3
0.3
C.3
inc.
Px.F.
(2-3)
(1-4)
(3,4)
0.02
0.3
A-10
-------
Tennessee
Texas
Virginia
Washington
Mill
Size
Ave. Kraft
Production
( Kraft ) .. ...Recovery Furnace
(Capacity) No. cf i Rating '' " Year
ControrTRS Level
Company . ._.; ......Location... _.._J..Jpd L-Units _.]laLul-' _t.Pd_...).nsta.llerl.schnique. -?/T AC?
Bowater
Packaging
Champion
I. P.
Owens-Illinois
Southland
Southland
Temple-Fostex
Chesapeake
Continental
Union Camp
Westvaco
Boise Cascade
Crown Zellerbach
Crown Zellerbach
Longview
St. Regis
Weyerhaeuser
Calhoun
Counce
Pasadena
Texarkana
Orange
Houston
Lufkin
Evadale
West Point
Hopewel 1
Franklin
Covington
Wall ola
Caraas
Port Townsend
Longview
Tacoma
Everett
bOO
(BOO)
(775)
850
(820)
610
1000
(900)
650
(500)
, 400
(400)
1250
(1200)
1150
(1150)
896
(900)
1430
(1500)
1048
(1000)
460
(460)
780
(760)
420
(420)
1600
(]928}
1090
(1040)
360
(375)
2
1
2
1
2
1
2
3
3
2
3
1
2
2
1
3
2
1
CE
CE
B&W
B&W
B&W
CF
CE
R&l>'
CE
CE
CE
CE
CE
B&W
B&W
CE
CE
CE
CE
CE
600&320
420
550
. 550
750
550(each)
500
pre-1965
post-1965
1971
1955
1969
1965
post-1965
BLO
BLO
BLO
(oxyqen)
BLO
175(each)pre-1965
534
1100
530
900
400&200
1966
post-1965
pre-1965
oost-1965
ore-1965
375(each)pre-1965
580
580&350
1320
250
165
350
660
725
1100
2-700
863
467
365
post-1965
pre-1965
post-1965
1960
1957
1955
post-1965
post-1965
post-1965
pre-1965
oost-1965
pre-1965
BLO
BLO
BLO
BLO
BLO
BLO
BLO
Low Odor
BLO
Low Odor
BLO
oost-1965 BLO
2.
2.
0.
0.
2.
2.
15.
2.
0.
0.
0.
0.
0.
0.
0.
0.
1
1
5
5
1
1
0
1
5
5
5
5
5
5
5
5
0.5
0.5
A-11
-------
I.ini3
;o . of
Uni.t.s
1
3
1
1
1
3
2
3
3
2
3
4
2
1
Kiln
Size TRS .
Tons (CaO) Level
! P_er_.i)ay
35D
260
13)
85
36)
300
44 ">
(total)
40)
(total)
200
500
196 & 80
140
Digester 9
i Type11
#/J AD?.;_..No_. j. .(Size)
0.8
0.8
0.05
-0.2
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.2
0.2
0.2
0.2
0.2
0.2
6
5
9
5
2
1
6
9
1
8
1
13
2
12
2
10
5
1
9
1
9
1
18
4
4
2
6
B
(500)
B
B
100 each
B
C
(1000)
C
(500)
B
(400)
BO 200)
C (200)
B (600)
C (600)
B 900)
C (.230)
B (950)
C [800 )
B
B
C (150)
B
C
B
C
B (1600)
C (600
B (240)
C (690)
B (550)
Control
rnr i i i L t 1 r 1 '.- t. i ' L - I- i f»^ p -, IIP' TIlr
TRS rv,..' .-., -, in,- ' /. "''<-'' >^oci, !,i'S;,^'r Tljj
Level ^ ' Co'itir.l Level' rapacity (''asher Level''
Technique,?/! ADP :!°- - T'jch:,ic:i:- ,'.'/ 1" Air ;i0. AOT!'D Slopes .;//T A[)P
inc.
inc.
inc.
inc.
inc.
Power
inc.
inc.
inc. .
inc.
O.C2 \ 1 inc. 0.02 j 1 4 0.3
9
1.5
0.02
-0.02.
1.5
1.5'
1.5
1.5
0.02
0.02
0.02
0.02
0.02
0.02
0.02
i
: 0.02
0.02
0.02
1 ,
\ 1
1.0
2 inc. 0.02 j
1
jj
inc. 0.02
1 1.0
1 1.0
2 1.0
3 1.0
3 0.02
1
2 0.02
4 0.02
0.02
2 inc. 0.02
3 inc. 0.02
inc; 0.02
6 inc. 0.02
2 inc. 0.02
1 inc. 0.02
4
3
3 3
A-12
-------
Mill
Size
Ave. Kraft
Production
- .....: ( Kraft
| ! (Capacity
Company 'i _. Loca.tio.n '.__ tpd._ .
Uashim - Ueyerhaeuser Longvicw 650
ton ' (30f>)
nssc
ch. rcc.
plant
Wisconsin Consolidated Wise. Rapids 400
(400)
ch. rec.
plant
Great Northern Nekoosa 310
(330)
ch. roc.
plant
Hammeriirill Kaukauna 400
(400)
Mosinee I1osir.ee 174
(175)
) . . Recovery
) No ."of [
! ..Units.... Manuf-L
i B&W
CE
i
Pinnace
Hating j " Year
tpd installed
1200 1972
350 pre-1965
2 CE 400(each)post-1965
2 CE
1 B&W
1 B&W
350 pre-1965
165 pre-1965
390 1960
250 1973
Control TRS Level
:;chnique. .-//T AtiP
Low Odor 0,5
BLO
BLO 0.5
0.5
BLO 0.5
Low Odor 0.5
New Mills (Planned or under construction)
Scott Paper - Skowhegan, Maine - 750 TPD
Potlatch Corp. - McGehee, Arkansas - 500 TPD
A-13
-------
Kiln
Size - TRS ,
... of Tons (CaO) Level
Jn.its Per Day #/T AD?
0.2
0.8
0.8
0.8
0.8
Digester ?. -\
Type i Control Level
Ti!3
No. .:_(Size) TechniqupiVT ADP !
/. .'> o P
! '1 tip 11.'- effect T,.c
TRS rvopnrcio- Tlte 4 Bruv.v. Stock Hasher
4 Control LOVIM Capacity Kashor Level
Jrc'iniquc f/T AD? _ No. .. ADTPD " .-Stages .$/j /,;;
.
"
12
1
B
C
scrubber
inc.
B
(17S)
0.02
1.1
1.5
1.5
1.5
inc.
0.02
1.0
1.0
1.0
1.0
0.3
Source
Recovery Furnacs
[
Lime Kiln
Digester
Multiple-effect
Evaporator
Brown Stock Washer
CONVERSION1 ';
Ib/T flDP
0.15
0.5
0.6
2.1
15.0
0.05
0.1
0.?
0.8
0.02
1.1
1.5
0.02
0.08
1.0
0.02
0.3
P ^ ; r:
Eriisioii Rate
g/Ko APP
0.075
0.25
0.3
1.05
7.5
0.025
0.05
0.1
0.4
0.01
0.55
0.75
0.01
0.04
0.5
0.01
0.15
ppm
5
17.5
20
70
550
10
20
40
170
<5
7000
9500
<5
350
6700
<5
30
A-14
-------
APPENDIX B
DATA SUMMARY
KRAFT PULP MILLS
Recovery Furnaces, Smelt Dissolving Tanks, Lime Kilns, and Incinerators
Results are summarized for tests conducted by EPA at 6 kraft pulp mills.
At these mills a total of 9 TRS tests; 3 recovery furnaces, 2 smelt dissolving
tanks, 3 lime kilns, and one incinerator,were conducted by EPA. Emission
data obtained from operators or state agencies are also reported for some
of the facilities.
TRS EMISSION DATA
Incinerator:
The incinerator handles the noncondensable gases from a continuous
digester system and a multiple-effect evaporator system. The
continuous digester was producing 670 tons of pulp per day.
The Incinerator was operating at 1000°F with a retention time
for the gases of at least 0.5 seconds. Natural gas is fired in
the incinerator.
Recovery Furnaces:
A. Conventional type recovery furnace designed for an equivalent
pulp production rate of 657 tons per day. TRS emissions are
controlled by using black liquor oxidation and maintaining proper
*
furnace operation. The furnace was operating near its design
capacity during the EPA test period. Continuous monitoring data
were also obtained from the operator.
B-l
-------
B. Low-odor type recovery furnace designed for an equivalent pulp
production of 300 tons per day. During the EPA testing, the
furnace was operating at a rate of about 345 tons of pulp per
day. TRS emissions are controlled by eliminating the direct contact
evaporator and maintaining proper furnace operation. Noncondensable
gases from the brown stock washer system are burned in this furnace.
Continuous monitoring data were also obtained from the state agency.
D. Conventional type recovery furnace designed for an equivalent pulp
production rate of 602 tons per day. TRS emissions are controlled
by black liquor oxidation and maintaining proper furnace operation.
H. Low-odor type recovery furnace operating at an equivalent pulp
production rate of about 200 tons per day. TRS emissions are
controlled by maintaining proper furnace operation. Data were
obtained from the state agency.
K. Low-odor type recovery furnace designed for an equivalent pulp
production rate of about 863 tons per day. TRS emissions are
controlled by maintaining proper furnace operation. Data were
obtained from state agency.
Smelt Dissolving Tanks
D. A wet fan type scrubber is employed to control the particulate
emissions. Weak wash liquor is used as the scrubbing medium.
The associated recovery furnace operates at an equivalent pulp
production rate of 570 tons per day.
E. A wet fan type scrubber is employed to control the particulate
emissions. Fresh water is used as the scrubbing medium. The
associated recovery furnace operates at an equivalent pulp production
rate of 770 tons per day.
P-2
-------
Lime Kilns
D. Rotary lime kiln operating at an equivalent pulp production rate
of 570 tons per day. TRS emissions are controlled by maintaining
proper kiln combustion and proper lime mud washing. Noncondensable
gases from the multiple-effect evaporators are burned in the kiln.
E. Rotary lime kiln operating at an equivalent pulp production rate
of about 770 tons per day. TRS emissions are controlled by
maintaining proper combustion in the kiln, maintaining proper
lime mud washing, and using a caustic solution in the particulate
scrubber. Noncondensable gases from the digesters, multiple-effect
evaporators, condensate stripper, and miscellaneous storage tanks
are burned in the kiln. Continuous monitoring data were also obtained
from the operator.
K. Rotary lime kiln operating at an equivalent pulp production rate
of about 320 tons per day. TRS emissions are controlled by main-
taining proper combustion in the kiln and proper lime mud washing.
Noncondensable gases from the digesters, multiple-effect evaporators,'
and turpentine system are burned in the kiln.
0. Rotary lime kiln not tested by EPA. Continuous monitoring data
was obtained from the local agency. TRS emissions are controlled
by maintaining process combustion in the kiln.
B-3
-------
Table 19 - TRS and SOg Emissions from Incineration
FACILITY - Incinerator
Summary of Results
Run Number
Date - 1972
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM (XI000) 2610
Flow rate - DSCF/ton
Temoerature - °F
Water vanor - Vol. %
C02 - Vol. % dry
62 - Vol. % dry
CO - ppm
TRS Emissions
pom
Ib/hr
Ib/ton of nulp
S_0g Emissions
nnm
Ib/hr
Ib/ton of nulo
1
10/5
240
2610
805
6.3
2.6
11.8
0
2.8
1.5
0.06
25
9.4
0.4
2
10/6
240
2223
805
4.3
2.4
12.0
0
0.4
0.2
0.007
306
96.9
3.8
3
10/7
240
2302
805
5.4
2.1
12.7
0
1.6
0.6
0.02
1050
358
13.9
4
12/13
240
-
-
-
9.0
15.7
0
0.9
0.4
0.02
«"
-
_
B-4
-------
Table 20 - TRS and S02 Emissions from Recovery Furnace A
FACILITY ~ Recovery Furnace A
Summary of Results
Run Number
Date - 1972
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM (XI000) 142
Flow rate - DSCF/ton
Temperature - °F
Water vanor - Vol. %
C02 - Vol. % dry
02 - Vol. % dry
CO - ppm
TRS Emissions
DDm
Ib/hr
Ib/ton of nulo
SO? Emissions
DDm
Ib/hr
Ib/ton of nulo
1
6/3
240
142
314
25.5
10.4
10.7
153
2.0
1.5
45
85.0
2 3
6/4 6/5
240 240
145
304
25.3
8.2 10.7
11.4 11.4
93 84
1.4 1.4
1.1 1.1
116 79
- -
4
6/6
240
148
303
21.9
11.8
10.1
95
1.5
1.2
118
mf
5
6/7
240
-
.
-
12.9
10.1
102
0.7
0.6
50
*
B-5
-------
Table 21 - TRS and S02 Emissions from Recovery Furnace B
FACILITY - Recovery Furnace B
Summary of Results
Run Number
Date - 1972
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM (XI 000) 85
Flow rate - DSCF/ton
Temnerature - °F
Water vanor - Vol . %
C02 - Vol. % dry
Og - Vol. % dry
CO - opm
TRS Emissions
pom
Ib/hr
Ib/ton of oulo*
S02 Emissions
pom
Ib/hr
Ib/ton of nulo
1
7/13
240
85
395
0
.05
2
7/14
240
84
400
12.3
8.T
0
0.2
0.7
.01
3
7/15
240
86
415
12.4
7,. 6
0
0.5
o:i
.02
4 5
7/18 7/19
240 240
12.7 12.0
7.7 8.0
0 0
0.3 0.4
0.2 0.2
.01 .01
6
7/20
240
12.4
8.0
0
0.3
0.2
.01
0.9
* Based on 334.5 ATDP/day
B-6
-------
Table 22 - TRS and S02 Emissions from Recovery Furnace D
FACILITY - Recovery Furnace D
Summary of Results
Run Number 1 2345
Date - 1972 11/11 IT/12 11/13 11/14 11/15
Test Time - minutes 240 240 240 240 240
Production Rate - TPH - - - -
Stack Effluent
Flow rate - DSCFM (X1000) 73.2 73.2 73.2 73.2 73.2
Flow rate - DSCF/ton
Temoerature - °F
Water vaoor - Vol. % 35 35 35 35 35
C02 - Vol. % dry
02 - Vol. % dry
CO - ppm
TRS Emissions
DDm 3.1 2.8 3.9 7.0 2.8
Ib/hr 55.1 48.9 53.7 12.5 46.0
Ib/ton of nulo - - - -
SO? Emissions
pom 15.5 ' 1.0 22.9 5.0 14.2
Ib/hr 162 10 239 52 149
Ib/ton of nulo - -
B-7
-------
Table 23
ADDITIONAL TRS EMISSION DATA
FOR RECOVERY FURNACES*
Month
July 1971
Aug.
Sept.
Oct.
Nov.
Dec.
Jan. 1972
Feb.
March
April
May
June
July
Aug.
Sept.
Oct.
*Tested by
Recovery
Furnace A
TRS Concentration 1
(ppm, daily average
basis)
Maximum Average f
6.0
20.0
5.0
10.9
. 4.4
9.8
5.5
3.3
2.5
5.3
5.5
8.2
9.8
9.0
4.9
6.1
3.1
2.4
1.5
2.8
1.3
1.8
1.6
1.3
1.0
2.0
2.1
3.8
3.7
3.3
2.9
2.2
Recovery Furnace B
Month
April 1972
May
June
July
Aug.-
Oct.
Nov.
Dec.
Jan. 1973
Feb.
March
Apri 1
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
TRS Concentration
(ppm, daily average
basis)
Maximum Average
1.4
2.3
2.8
4.6
5.0
1.9
0.7
1.0
1.5
2.6
2.4
1.5
1.6
1.9
1.6
. 3.1
1.8
2.0
1.6
3.4
0.7
1.2
1.5
1.1
1.5
0.7
0.4
0.7
0.8
1.0
0.9
0.8
1.0
1.1
1.0
1.2
0.8
0.9
0.8
1.6
operators using barton titrators.
B-8
-------
Table 23 (cont.)
ADDITIONAL TRS EMISSION DATA
FOR RECOVERY FURNACES
Recovery Furnace A
TRS Concentration
(ppm, daily average
basis)
Month Maximum Average
-
!
Recovery Furnace H
TRS Concentration
(ppm, daily average
basis)
Month Maximum Average
April 1972 3 2.1
Hay 4 2.1
June 7 3.5
June 1972 8 3.1
July 4 2.4
Aug. 4 1.9
Sept. 2 1.3
Oct. 6 1.8
Month
Jan. 1974
Feb.
March
April
May.
June
Month
Aug. 1973
Sept.
Oct.
Nov.
Dec.
Jan. 1974
Feb.
March
April
May
Recovery Furnace B
TRS Concentration
(ppm, daily average . . .
basis)
Maximum Average
1.4 0.8
1.9 1.3
5.0 1.6
2.4 1.2
1.8 1.0
1.5 1.0
Recovery Furnace K
TRS Concentration
(ppm, daily average
basis)
Maximum Average
6.2 1.0
32.0 5.2
7.3 2.4
17.0 4.1
1.2 0.7
1.8 0.6
2.4 1.0
9.7 2.3
3.0 1.4
3.4 1.4
B-9
-------
Table 23-A
TRS EMISSION DATA FOR A CROSS RECOVERY FURNACE*
Days TRS (4-hour)
Sulfldit.y Average TRS Maximum 4-Hour Emissions Greater
Month Range (%) Em'ssions (ppm) TRS Emissions (ppm) than 25 ppm
Oct
Nov
Dec
Jan
Feb
76
76
76
77
77
22
28
28
27
27
- 36
- 33
- 34
- 36
- 35
12.
24.
9.
7.
8.
5
3
5
7
0
54
51
43
36
48
.5
.2
.2
.5
.0
2
4
1
0
1
* Tested by operator using barton tatrator.
-------
Table 24 - TRS Emissions frorr Smelt Dissolving Tank D
FACILITY - Smelt Dissolving Tank D
Summary of Results
Run Number 1 2 3
Date - 1973 10/31 11/1 11/2
Test Time - minutes 240 240 240
Production Rate - TPH 25.1 25.9 25.6
Stack Effluent
Flow rate - DSCFM 9000 8880 9400
Flow rate - DSCF/ton 21514 20571 22031
Temperature - °F
Water vaoor - Vol. % 37 41 40
C02 - Vol. * dry
02 - Vol. % dry
CO - ppm
TRS Emissions
pom 8.1 8.8 6.9
Ib/hr 0.43 0.44 0.38
Ib/ton of oulD 0.017 0.017 .015
B-10
-------
Table 25 - TRS Emissions from Smelt Dissolving Tank E
Run Number
Date - 1973
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM
Flow rate - DSCF/ton
Temoerature - °F
Water vaoor - Vol. %
C02 - Vol. % dry
02 - Vol. X dry
CO - ppm
TRS Emissions
pom
Ib/hr
Ib/ton of DU!D
FACILITY- Smelt Dissolving Tank E
Summary of Results
1 2 3
9/18 9/19 9/20
240. 240 240
30.1 34,1 31.3
19542
38954
26
2.4
0.27
0.009
18740
32974
26
1.9
0.20
.006
19100
36613
23.3
2.7
0.28
.009
B-n
-------
1
11/5
240
2
11/7
240
3
11/7
240
4
11/7
240
5
11/8
240
6
11/8
240
Ti&le 26 - T3S Emissions from Lime Kiln D
FACILITY - Lime Kiln D
Summarv of Results
Run Number
Date - 1973
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM (Xinno)
Flow rate - DSCF/ton
Temperature - °F
Water vaoor - Vol. % 43 35 40 38 41 31
C02 - Vol. % dry
02 - Vol. % dry
CO - Dpm
TRS Emissions
pom 3.5 24.1 2.8 5.7 4.6 17.8
Ib/hr
Ib/ton of DU!D
B-12
-------
1
9/24
240
2
9/25
240
3
9/26
240
4
9/26
240
5
9/27
240
6
9/27
240
Table 27 - TRS Emissions from Lime Kiln E
FACILITY- Lime Kiln E
Summary of Results
Run Number
Date - 1973
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - OSCFM (XI000)
Flow rate - DSCF/ton
Temperature - °F
Mater vaoor - Vol. %
0)3 - Vol. % dry 9.4 10.2 10.0 9.8 8.2 9.8
Og - Vol. % dry 13.2 11.0 12.2 12.0 13.1 11.8
CO - ppm
TRS Emissions
pDm 1.7 0.8 0.5 0.4 0.3 0.5
Ib/hr
Ib/ton of DU!D
B-13
-------
Table 28 - TRS arH S02 Emissions from Lime Kiln K
FACILITY - Lime Kiln K
Summarv of Results
Run Number
Date - 1974
Test Time - minutes
Production Rate - TPH
Stack Effluent
Flow rate - DSCFM (XI000) 13.8
Flow rate - DSCF/ton
Temperature - °F
Mater vaoor - Vol. %
C02 - Vol. % dry
02 - Vol. % dry
CO - ppm
TRS Emissions
com
Ib/hr
Ib/ton of DU!D
SO? Emissions
pom
Ib/hr
Ib/ton of nulo
1
4/5
240
13.8
142
21.8
13.0
7.6
0
4.6
0.34
52
7.2
2
4/5
240
13.8
142
21.8
13.0
7.6
0
12.0
0.88
42
5.8
3
4/9
240
14.0
146
22.9
14.2
7.1
0
4.5
0.33
25
3.5
4
4/9
240
13.4
152
26.0
14.2
7.1
0
4.8
0.34
18
2.4
5
4/10
240
13.6
155
25.8
14.6
6.4
0
4.0
0.29
16
2.2
6
4/10
240
14.2
154
26.8
14.2
7.2
0
5.2
0.39
37 I
5.2
B-14
-------
Table 29
ADDITIONAL TRS EMISSION DATA
FOR LIME KILNS*
Month
May 1973
June
.July
.Aug.
Sept.
Oct.
Nov.
Dec.
Jan. 1974
Feb.
March
April
May
L1me Kiln E
TRS Concentration
(ppm, daily average)
Maximum Average
1.4
3.4
2.1
1.4
10.1
7.1
5.9
8.9
3.4
2.6
0.7
3.1
2.9
0.3
0.7
0.4
0.3
1.5
1.0
0.8
1.0
0.6
0.2
0.1
0.6
0.7
Month
Jan. 1973
Feb.
March
April
May.
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan. 1974
Feb.
March
April
May
Lime Kiln 0
TRS Concentr
(ppm, daily
Maximum
14
20
14
32
16
10
99
12
17
34
12
22
30
33
30
40
25
Average
tion
average)
Average
6.8
9.3
7.6
9.6
4.7
3.4
4.5
3.8
5.0
8.2
5.7
9.8
17.9
21.1
19.3
16.2
12.3
= 9.7
*Tested by operators using barton titrators.
B-15
-------
TABLE 29 (CONTINUED)
Lime Kiln P
TRS Summary: 4-Hour Averages
4-Hour Averages Monitored
Month
February '75
March '75
April '75
May '75
<5 ppm
45
65
63
53
>5/ <10 ppm >10/ <20 ppm
26 9
25 7
16 12
25 13
01 01
fa fa
>20 ppm >40 ppm
20 12
7 2
9 5
8 2
B-16
-------
"i '.LI: i
IMPAC" OF CONTROLLING TH:'. VARIOUS TRS SOURCES CM
AMBIENT TRS CONCENTRATION FROM A 907 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(One-hour average)
ug/m3
Frequency -
% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolvino
lank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1) Process Controls
2) Process Controls +
Hiah Eff. Mud Washing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scrubbina
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Incineration
"Uncontroi led Distance
Level (9 0.3 km 0.3 0.6
2465
2465
6000
6000
1175
1175
256
256
256
113
96
96
172
4400
aann
100 64
25 16
4680 1716
o*
96 60
0*
64 23
16 6
8 3
8
19 6
nc
\j\.
22
2200
nb.- .
from Source (km) Greater than 1/2
1.0 1.5 2.0 the Maximum
38 16
10 4
1090
36
*
18 15 12
443
222
3 2
3 1 1
8 6
306 165
w
28
28
3
4
-
1
1
1
34
25
3
25
Percent
Reduction
95.9
99.0
22.0
inn n
IUU. U
91.8
100.0
75.0
93.8
96.9
92.9
80.2
100.0
87.2
50.0
inn n
a
i i
CO
-o
m
8
f-H
O
^Z
CO
-H
O
1 1
i '1
GO
ya
m
CO
c
i
I
GO
^
\j
-o
m
z
o
1 1
X
o
Reduction from uncontrolled average level
Gases are assumed burned in th» lime kiln. The levels from the lime kiln
include unburned TRS portion of these oases.
-------
TABLh
IMPACT OF CONTROLLING, THF VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 907 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(24-hour average)
ug/m3
Frequency -
% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolvino
Tank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1) Process Controls
2) Process Controls +
Hiqh Eff. Mud Washing
3) Process Controls + High
Eff. Mud Washina +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Tnri npratinn
Uncontrol led
Level @ 0.3 km
643
643
580
580
115
115
56
56
56
30
25
25
16
840
Rdn
Distance
0.3 0.6
26
6
450
Ob'
10
Ob-
14
4
2
2
5
OC-
2
420
nb-
16
4
150
6
7
2
1
2
from Source (km)
1.0 1.5 2.0
9 4
2 1
68
*k
3
432
1 1 1
1 1 1
1 1
1 1 1
1 1
66 34
^
Greater than 1/2
the Maximum
29
29
48
55
1
1
1
26
20
46
30
Percent
Reduction
96.0
99.0
22.4
100.0
91.3
100.0
75.0
92.9
96.4
93.3
80.0
100.0
87.5
50.0
inn n
Reduction from uncontrolled averaae level.
Gases are assumed burned in the lime kiln. The levels from the lime kiln
include unburned TRS portion of these oases.
NO vpnt nases
-------
TABL C-3
IMPACT or CONTROLLING THE VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 4S4 MEGAGRAMS/DAY
KRAFT PULP MILL
t
Maximum Ambient Concentration: pg/m3 Frequency - 2
(10 second average) * of Concentrations *
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving
Tank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1) Process Controls
2) Process Controls +
High Eff. Mud Mashing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Incineration
uncontro: led
Level 0 0.3 km
5250
5250
9800
QOnn
7OUU
1950
i o*\n
1 7*f U
440
440
440
370
210
7in
C. IU
270
9100
9100
Distance
0.3
210
53
7645
Ob
U
156
nb
U***
110
28
14
25
42
nC
u
34
4550
Ob.
0.6
135
35
2790
98
40
10
5
14
from so
1.0
80
20
1775
58
30
8
4
9
7
13
635
urce (Km) 'Greater than 1/Z
1.5 2.0 the Maximum
34 33
8 33
3
4
25 21 1
6 5 1
3 3 1
6 19
33 30
8 3
340 16
Percent
Reduction3
96.0
99.0
22.0
inn n
1 UU> U
92.0
inn n
IUU. \t
75.0
93.6
96.8
93.2
80.0
inn n
IUU. U
87.4
50.0
100.0
' Reduction from uncontrolled average level
1 Gases are assumed burned in the lime kiln. The levels from the lime kiln
include unburned TRS portion of these oases.
No ven* oases.
-------
TABL_ L-4
IMPACT OF CONTROLLING THE VARIOUS TRS SOURCtS ON
AMBIENT TRS CONCENTRATION FROM A 454 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(One-hour average)
ug/m3
Frequency -
% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving
Tank
Condensate Stripning
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1 ) Process Controls
2) Process Controls +
High iff. Mud Washing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
.
Uncontrolled Distance
Level
18
987
322
1 1 1
3 2
2 1 1
^
4 3
195 105
33
33
3
4
1
1
1
19
30
-
3
16
Percent
Reduction
96.0
99.0
22.0
inn n
1 UU. U
92.0
i nn n
1 UU. U
75.0
93.4
96.3
92.9
80.3
100.0
86.9
50.0
l nn n
Reduction from uncontrolled averace level
Gases are assumed burned in tne lime kiln. The levels fron the lime kiln
include unburned TRS portion of these oases.
No vent oases
-------
fABLi C-5
IMPACT Oh CONTROLLING hir VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 454 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(24-hour average)
Frequency -
% of Concentrations
'
iource
t'eove1". Furna..'-
Multiple-Ffter*
Evaporator
Limp Kiln
(. on11""! IP'" Unique'
i>L'.' (ii1 nnin)
*L<- ' L pom)
Scrut-'.inv
Incineration
-^rubber
Incineration
1 ) Process Controls
P) Process Controls -t
Hiah Eff. Mud Washina
TJncnntrol led Distance from Sourco (km] breater than 1/Z
Level » 0.3 km 0.3 0.6 1.0 1.5 2.0 the Maximum
430
430
290
290
60
60
32
32
17
4
226
ot>-
5
00-
8
2
11
3
76
3
4
1
6
1
34
2
2 2
1 1
2
1
>
1
1
27
27
48
55
4
4
Percent .
Reduction
96.0
99.1
22.1
100.0
91.7
100.0
75.0
93.8
3) Process Controls + High
Eff. Mud Washinn +
Caustic Scrubbina
32
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving
Tank
Condensate Stripping
System
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
17
17
17
8
420
A9n
1
3 1
nr
i
210
nb
1
1
28
1
1
17
-^_
52
19
46
37
96.9
94.1
82.4
100.0
56.0
100.0
Reduction from uncontrol led average level
Gases are assumed burned in trv lime kiln.
includp unburned TRS portion nf these
The level1., from the lime kiln
-------
TABLE I 6
IMPACT OF CONTROLLING THE VARIOUS TRS SOURCES Ofi
AMBIENT TRS CONCENTRATION FROM A 1350 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(10 second average)
ug/m3
Frequency -
% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolving
Tank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1 ) Process Controls
?) Process Controls +
Hi ah Cff. Hud Washing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scruobing
Incineration
Incineration
Molecular Oxvqen
Fresh Waier
Scrubber
Incineration
Uncontrolled Distance
Level P 0.3 km 0.3 0.6
11 ,785
11,785
29,000
29,000
5,750
R 7"5n
J , / JU
1 ,280
1,280
1,280
555
490
4pn
HJU
840
21 ,000
?i .nnn
470 305
120 75
22,930 8370
Oh
470 295
Oh
U
320 115
80 29
40 14
37
98 33
rtr
u\*
105
10,500
nb
from Source (km) Greater than 1/2
1.0 1.5 2.0 the Maximum
180 75
45 20
5325
175
88 73 62
22 18 16
11 9 8
14 9
14 7 5
^.
39 24
1460 780
i.i -. .1 *>
28
28
3
4
1
1
1
34
25
3
25
Percent
Reduction
96.0
99.0
20.9
100. 0
91.8
i nn n
1 UU . U
75.0
93.8
96.9
93.3
80.0
i nn n
I UU. U
87.5
50.0
inn n
Reduction from uncontrolled averaoe level
Gases are assumed burned in tne lime kiln. The levels fron tne lime kiln
include unburned TRS portion of these cases.
No vent case-
-------
I ABU C-7
IMPACT OF CONTROLLING THE VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 1350 MEGAGRAMS/OAY
KRAFT PULP HILL
Maximum Ambient Concentration:
(One-hour average)
Frequency -
% of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
Evaporator
Lime Kiln
Brown Stock
Washer System
Black Liquor
Oxidation System
Smelt Dissolvinn
Tank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1 ) Process Controls
2) Process Controls +
Hiqh Eff. Mud Washing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Tnri noraflnn
uncontrolled Distance
Level 1? 0.3 km 0.3 0.6
3,750
3,750
9,000
Q nnn
y , uuu
1,775
1 ,775
385
385
385
175
150
150
255
6,600
fi.finn
150 97
37 24
7,020 2,575
Ob
154 90
Oh
96 34
24 9
12 4
12
30 10
(1C
32
3,300
nb
from Source (km) 'Greater than 1/2
1.0 1.5 2.0 the Maximum
57 24
14 6
1,640
53
27 22 19
665
332
4 3 ._.
5 2 2
12 9
460 250
v
28
28
3
4
1
1
1
34
25
-
3
25
Percent
Reduction
96.0
99.0
22.0
100.0
91.3
100.0
75.0
93.8
96.9
94.8
80.0
100.0
87.5
50.0
inn n
Reduction from uncontrolled average level
Gases are assumed burned in the lime kiln. The levels from thp lime kiln
include unburned TRS portion of these pases.
N-- vent aises.
-------
TABU i. 8
IMPACT OF CONTROLLING THE VARIOUS TRS SOURCES ON
AMBIENT TRS CONCENTRATION FROM A 1350 MEGAGRAMS/DAY
KRAFT PULP MILL
Maximum Ambient Concentration:
(24-hour average)
yg/m3
Frequency -
of Concentrations
Source
Recovery Furnace
Digester
Multiple-Effect
tvaporator
Lime Kiln
Brown Stock
Washer System
Blacl- Liquor
Oxidation System
Smelt Dissolving
lank
Condensate Stripping
System
Control Techniques
BLO (20 ppm)
BLO (5 ppm)
Scrubber
Incineration
Scrubber
Incineration
1 ) Process Controls
2) Process Controls +
Hiqh Eff. Mud Washing
3) Process Controls + High
Eff. Mud Washing +
Caustic Scrubbing
Incineration
Incineration
Molecular Oxygen
Fresh Water
Scrubber
Tnrinprflti nn
lincontrol led Distance
Level (? 0.3 km 0.3 0.6
965
965
870
870
175
175
80
80
80
45
40
40
25
1,300
i ..inn -
40 25
10 6
680 230
nb
15 9
Oh
20 10
5 3
3 2
3
8 3
Or
3
650
nb
from Source (km) .Greater than 1/2
1.0 1.5 2.0 the Maximum
15 5
4 1
100
5
643
2 1 1
1 1 1
1 1
1 1 1
1 1
87 53
29
29
48
55
1
1
1
26
20
46
30
Percent
Reduction
95.9
99.0
21.8
100.0
91.4
100.0
75.0
93.8
96.3
93.3
80.0
100.0
88 .'0
50.0
inn n
Reduction from uncontrolled averacie level
Gases are assumed burned in the lime kiln. The levels from the lime kiln
include unburned TRS portion of these cases.
No vent aasp .
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-^0/9-7R-nm»
4. TITLE AND SUBTITLE
Draft Guideline Document: Control of TRS
from Existing Kraft Pulp Mills
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Air Quality Planning and Standa
Environmental Protection Agency .
Research Triangle Park, North Carolina 27
12. SPONSORING AGENCY NAME AND ADDRESS
DAA for Air Quality Planning and Standard
Office of i^ir and .Waste Management
U. S» Environmental Protection Agency
Research Triangle Park, North Carolina 27
3. RECIPIENT'S ACCESSIOI*NO.
5. REPOFW DATE
-. _. January 1978
EllLU,wJ.UU~ 6 PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
rds
11. CONTRACT/GRANT NO.
711
13. TYPE OF REPORT AND PERIOD COVERED
S
14. SPONSORING AGENCY CODE
?11 EPA/200/04
IS. SUPPLEMENTARY NOTES -
This document discusses the proposed guidelines for existing mills and the
resulting environmental and economic effects.
16. ABSTRACT - - -
Guidelines to aid the states .in. their preparation of plans for the control of
emissions of total reduced, sulfur (TRS) from existing kraf t pulp mills are
being proposed under the authority of section lll(d) of . the .Clean Air Act.
TRS emissions from kraf t pulp mills are extremely .odorous., and there are
numerous instances of poorly controlled mills, creating .public. odor problems.
Adoption of the proposed emission guidelines by the states would result in
an .overall reduction of about 80 percent in nationwide TRS emissions from
kraft pulp mills.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Air pollution
Pollution control
Kraft pulp mills .
Total reduced sulfur
Particulate matter
Emission Guidelines
Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air pollution control
19. SECURITY CLASS (This Report) 21. NO. OF PAGES
Unclassified 210
20. SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220-1 (9-73)
C-9
-------
-------