EPA
450/3-91-026
©EPA
United States Office of Air Quality
Environmental Protection Planning and Standards
Agency Research Triangle Park NC 27711
EPA-450/3-91-026
December 1991
Air
Alternative Control
Techniques Document --
Nitric and Adipic Acid
Manufacturing Plants
DALLAS, TsL'
-------
EPA-450/3-91-026
Alternative Control
Techniques Document-
Nitric and Adipic Acid
Manufacturing Plants
Emission Standards Division
U. S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
December 1991
-------
ALTERNATIVE CONTROL TECHNIQUES DOCUMENTS
This report is issued by the Emission Standards Division,
Office of Air Quality Planning and Standards, U. S. Environmental
Protection Agency, to provide information to State and local air
pollution control agencies. Mention of trade names and
commercial products is not intended to constitute endorsement or
recommendation for use. Copies of this report are available—as
supplies permit—from the Library Services Office (MD-35),
U. S. Environmental Protection Agency, Research Triangle Park,
North Carolina 27711 ([919] 541-2777) or, for a nominal fee, from
the National Technical Information Service, 5285 Port Royal Road,
Springfield, VA 22161 ([800] 553-NTIS).
-------
-------
TABLE OF CONTENTS
Page
LIST OF TABLES iv
LIST OF FIGURES vi
1.0 INTRODUCTION 1-1
2.0 SUMMARY 2-1
2.1 SUMMARY FOR NITRIC ACID PLANTS 2-1
2.2 SUMMARY FOR ADIPIC ACID PLANTS 2-3
3.0 DESCRIPTION OF NITRIC/ADIPIC ACID MANUFACTURING . . 3-1
3.1 NITRIC ACID MANUFACTURING 3-1
3.1.1 Uses and Industry Characterization ... 3-1
3.1.2 Production Process 3-2
3.1.3 Plant Design 3-6
3.1.4 Concentrated Nitric Acid Process .... 3-10
3.2 ADIPIC ACID MANUFACTURING 3-12
3.2.1 Uses and Industry Characterization . . . 3-12
3.2.2 Production Process 3-14
3.3 REFERENCES FOR CHAPTER 3 3-17
4.0 CHARACTERIZATION OF NOX EMISSIONS 4-1
4.1 NOX EMISSIONS FROM NITRIC ACID MANUFACTURING . 4-1
4.1.1 NOX Formation 4-1
4.1.2 Factors Affecting NOX Emission Levels . 4-2
4.1.3 Uncontrolled NOX Emission Levels .... 4-4
4.2 NOX EMISSIONS FROM ADIPIC ACID MANUFACTURING . 4-5
4.2.1 NOX Formation 4-5
4.2.2 Factors Affecting NOX Emission Levels . 4-6
4.2.3 Uncontrolled NOX Emission Levels .... 4-6
4.3 REFERENCES FOR CHAPTER 4 4-7
ii
-------
TABLE OF CONTENTS
Page
5.0 CONTROL TECHNIQUES FOR NITROGEN OXIDES FROM
NITRIC/ADIPIC ACID MANUFACTURING 5-1
5.1 NITRIC ACID MANUFACTURING 5-1
5.1.1 Extended Absorption 5-1
5.1.2 Nonselective Catalytic Reduction .... 5-7
5.1.3 Selective Catalytic Reduction 5-15
5.1.4 Control Technique Performance Summary . 5-23
5.1.5 Other Control Techniques 5-25
5.2 ADIPIC ACID MANUFACTURING 5-34
5.2.1 Extended Absorption 5-34
5.2.2 Thermal Reduction 5-38
5.2.3 Other Control Technique 5-43
5.2.4 Control Technique Performance Summary . 5-44
6.0 CONTROL COSTS 6-1
6.1 COSTS OF CONTROL TECHNIQUES USED IN NITRIC
ACID PLANTS 6-2
6.1.1P Extended Absorption 6-2
6.1.2 Nonselective Catalytic Reduction .... 6-6
6.1.3 Selective Catalytic Reduction 6-12
6.2 COSTS OF CONTROL TECHNIQUES USED IN ADIPIC ACID
PLANTS 6-17
6.2.1 Extended Absorption 6-19
6.2.2 Thermal Reduction 6-21
6.3 REFERENCES 6-25
7.0 ENVIRONMENTAL AND ENERGY IMPACTS 7-1
7.1 NITRIC ACID MANUFACTURING 7-1
7.1.1 Air Pollution 7-1
7.1.2 Solid Waste Disposal 7-4
7.1.3 Energy Consumption 7-4
7.2 ADIPIC ACID MANUFACTURING 7-6
7.2.1 Air Pollution 7-6
7.2.2 Energy Consumption 7-8
7.3 REFERENCES FOR CHAPTER 7 7-9
iii
-------
LIST OF TABLES
TABLE 2-1.
TABLE 2-2.
TABLE 2-3.
TABLE 2-4.
TABLE 5-1.
TABLE 5-2.
TABLE 5-3.
TABLE 5-4.
TABLE 5-5.
TABLE 5-6.
TABLE 5-7.
TABLE 6-1.
TABLE 6-2.
Page
NO EMISSIONS AND COST COMPARISON OF
ALTERNATIVE CONTROL TECHNIQUES USED
IN NITRIC ACID PLANTS 2-2
ENVIRONMENTAL AND ENERGY IMPACTS OF
ALTERNATIVE CONTROL TECHNIQUES USED
IN NITRIC ACID PLANTS 2-4
NOX EMISSIONS AND COST COMPARISON OF
ALTERNATIVE CONTROL TECHNIQUES USED IN
ADIPIC ACID PLANTS 2-4
ENVIRONMENTAL AND ENERGY IMPACTS OF
ALTERNATIVE CONTROL TECHNIQUES USED IN
ADIPIC ACID PLANTS 2-4
NITROGEN OXIDES EMISSIONS FROM NITRIC ACID
PLANTS USING EXTENDED ABSORPTION 5-6
NITROGEN OXIDES EMISSIONS FROM NITRIC ACID
PLANTS USING NONSELECTIVE CATALYTIC
REDUCTION 5-14
NITROGEN OXIDES EMISSIONS FROM NITRIC ACID
PLANTS USING RHONE-POULENC SCR
TECHNOLOGY 5-21
NITROGEN OXIDES EMISSIONS FROM NITRIC ACID
PLANTS USING BASF SCR TECHNOLOGY 5-22
SUMMARY OF NOX CONTROL TECHNIQUES
PERFORMANCE NITRIC ACID PLANTS 5-24
NITROGEN OXIDES EMISSIONS FROM ADIPIC ACID
PLANTS USING THERMAL REDUCTION 5-42
SUMMARY OF NOX CONTROL TECHNIQUES
PERFORMANCE FOR ADIPIC PLANTS 5-45
CAPITAL COST SUMMARY FOR NITRIC ACID PLANTS
USING EXTENDED ABSORPTION FOR NOX
CONTROL 6-4
ANNUAL COST SUMMARY FOR NITRIC ACID PLANTS
USING EXTENDED ABSORPTION FOR NOX
CONTROL 6-5
-------
LIST OF TABLES
TABLE 6-3.
TABLE 6-4.
COST EFFECTIVENESS FOR MODEL PLANTS USING
EXTENDED ABSORPTION FOR NOX CONTROL . . .
ANNUAL COST SUMMARY FOR NITRIC ACID USING
NONSELECTIVE CATALYTIC REDUCTION FOR
Page
6-7
NOX CONTROL 6-9
TABLE 6-5. COST EFFECTIVENESS FOR MODEL PLANTS USING
NONSELECTIVE CATALYTIC REDUCTION FOR
NOX CONTROL 6-11
TABLE 6-6. CAPITAL COST SUMMARY FOR NITRIC ACID
PLANTS USING SELECTIVE CATALYTIC REDUCTION
FOR NOX CONTROL 6-13
TABLE 6-7. ANNUAL COST SUMMARY FOR NITRIC ACID PLANTS USING
SELECTIVE CATALYTIC REDUCTION FOR NOX
CONTROL 6-16
TABLE 6-8. COST EFFECTIVENESS FOR NITRIC ACID PLANTS
USING SELECTIVE CATALYTIC REDUCTION FOR
NOX CONTROL 6-18
TABLE 6-9. ANNUAL COSTS FOR AN ADIPIC ACID PLANT USING
EXTENDED ABSORPTION FOR NOX CONTROL . . . 6-20
TABLE 6-10. ANNUAL COSTS FOR ADIPIC ACID PLANTS USING
THERMAL REDUCTION FOR NOX CONTROL .... 6-23
TABLE 6-11. COST EFFECTIVENESS FOR ADIPIC ACID PLANTS
USING THERMAL REDUCTION FOR NOX CONTROL . 6-24
TABLE 7-1. NOX EMISSIONS FROM NITRIC ACID
MANUFACTURING PLANTS 7-2
TABLE 7-2. ANNUAL ELECTRICITY REQUIREMENTS FOR
EXTENDED ABSORPTION AND ANNUAL FUEL
REQUIREMENTS FOR NSCR 7-5
TABLE 7-3. NOX EMISSIONS FROM ADIPIC ACID
MANUFACTURING PLANTS 7-7
-------
LIST OF FIGURES
Figure 3-1.
Figure 3-2.
Figure 3-3.
Figure 3-4.
Figure 3-5.
Figure 3-6.
Figure 5-1.
Figure 5-2.
Figure 5-3.
Figure 5-4.
Figure 5-5.
Figure 5-6.
Figure 5-7.
Figure 5-8.
Figure 5-9.
Figure 5-10.
Figure 5-11.
Figure 5-12.
Basic nitric acid production process . . .
Single-pressure nitric acid manufacturing
process
Dual-pressure nitric acid manufacturing
process
Nitric acid concentration using extractive
distillation
Nitric acid concentration using the direct
strong nitric process (Unde process) . . .
Basic adipic acid manufacturing
process
Extended absorption system using one large
absorber for NOX control at nitric acid
plant
Extended absorption system using second
absorb*
plants
absorber for NOX control at nitric acid
Nonselective catalytic reduction system for
NOX control at nitric acid plants ....
Selective catalytic reduction system for
NOX control at nitric acid plants ....
SCR catalyst performance as a function of
NH3/NOX mole ratio
SCR catalyst performance as a function of
area velocity ,
Process flow diagram for the Goodpasture
process
Flow diagram for the MASAR process ....
Schematic diagram of the CDL/VITOK NOX
removal process
Molecular sieve system
Extended absorption for NOX control at an
adipic acid plant
Thermal reduction unit for NOX control at
an adipic acid plant
Page
3-3
3-8
3-9
3-11
3-13
3-15
5-3
5-4
5-9
5-16
5-19
5-20
5-26
5-29
5-31
5-33
5-36
5-40
Vi
-------
1.0 INTRODUCTION
Congress, in the Clean Air Act Amendments of 1990 (CAAA),
amended Title I of the Clean Air Act (CAA) to address ozone
nonattainment areas. A new Subpart 2 was added to Part D of
Section 103. Section 183(c) of the new Subpart 2 provides that:
Within 3 years after the date of the enactment of the
[CAAA], the Administrator shall issue technical
documents which identify alternative controls for all
categories of stationary sources of ... oxides of
nitrogen which emit, or have the potential to emit
25 tons per year or more of such air pollutant.
These documents are to be subsequently revised and updated as the
Administrator deems necessary.
Nitric and adipic acid manufacturing have been identified as
categories of stationary sources that emit more than 25 tons of
nitrogen oxides (NOX) per year. This alternative control
techniques (ACT) document provides technical information for use
by State and local agencies to develop and implement regulatory
programs to control NOX emissions from nitric and adipic acid
manufacturing facilities. The decision to include both
categories in a single ACT document is based on similarities in
the process sources of NOX emissions from nitric and adipic acid
plants.
The information in this ACT document was generated from
previous EPA documents and literature searches and contacts with
acid manufacturers, engineering and construction firms, control
equipment vendors, and Federal, State, and local regulatory
agencies. Chapter 2.0 presents a summary of the findings of this
study. Chapter 3.0 provides process descriptions and industry
1-1
-------
characterizations of nitric and adipic acid manufacturing.
A discussion of NOX emission levels is presented in Chapter 4.0.
Alternative control techniques and achievable controlled emission
levels are discussed in Chapter 5.0. Chapter 6.0 presents
control costs and cost effectiveness for each control technique.
Environmental and energy impacts associated with using NOX
control techniques are discussed in Chapter 7.0.
1-2
-------
2.0 SUMMARY
The purpose of this document is to provide technical
information that State and local agencies can use to develop
strategies for reducing NOX emissions from nitric and adipic acid
manufacturing plants. This section presents a summary of the
information contained in this document, including uncontrolled
and controlled NOX emissions data, ACT'S, capital and annual
costs, and cost effectiveness. Section 2.1 presents a summary of
the information relating to nitric acid plants. Section 2.2
presents a summary of the information relating to adipic acid
plants.
2.1 SUMMARY FOR NITRIC ACID PLANTS
Approximately 65 plants in the United States produce nitric
acid. The ammonia-oxidation process is the most commonly used
process for producing weak (50 to 70 percent) nitric acid. The
absorption tower, common to all ammonia-oxidation nitric acid
production facilities, is the primary source of NOX emissions.
Three control techniques are predominantly used to reduce the
level of NOV emissions in the absorber tail gas: (1) extended
Ji
absorption, (2) nonselective catalytic reduction (NSCR), and
(3) selective catalytic reduction (SCR). This section presents a
summary of NOX control performance, control cost data, and
environmental impacts for each of the three control techniques
applied to each of three model plants.
Table 2-1 is a summary of NOX emissions and a cost
comparison of the three alternative NOX control techniques used
in model plants sized at 200, 500, and 1,000 tons of nitric acid
2-1
-------
TABLE 2-1. NO,. EMISSIONS AND COST COMPARISON OF ALTERNATIVE
CONTROL TECHNIQUES USED IN NITRIC ACID PLANTS
Plant
size,
toos/d
200
500
1,000
Uncon-
trolled NOX
emissions,
tons/yr*
718
1,800
3,590
Control
technique
Extended
Absorption"
NOX
removed,
tons/yr
679
1,700
3,400
Costs, $103
Capital
919
1,610
2,470
Annual
202
250
257
Cost
effectiveness,
$/ton NOX
removed
297
147
76
200
500
1,000
718
1,800
3,590
NSCRC
701
1,760
3,510
1,070
1,860
2,820
501
1,020
1,780
715
580
507
200
500
1,000
718
1,800
3,590
SCRd
616
1,550
3,090
314
409
553
188
442
714
305
285
231
250
898
SCRe
873
548
252
289
aBased on the following: (1) uncontrolled NOX emissions factor of 20 Ib/ton, (2) plant operating 359 days per
year.
Average control efficiency, 94.6 percent. Based on actual operating data.
cAverage control efficiency, 97.7 percent. Based on actual operating data.
^Control efficiency, 86 percent (required to reduce uncontrolled NOX emission level down to new source
performance standard (NSPS) level, 3.0 Ib/ton). Estimates provided by Engelhard Corporation.
eControl efficiency, 97.2 percent. Based on actual operating data.
2-2
-------
produced per day. Annual uncontrolled NOX emissions were
calculated based on an uncontrolled emission factor of 20 pounds
per ton (Ib/ton) of nitric acid produced. Annual NOX emissions
reductions were calculated using the average control efficiency
for each control technique. The average control efficiencies
used in the calculations are as follows:
1. Extended absorption—94.6 percent;
2. NSCR—97.7 percent; and
3. SCR—86 percent and 97.2 percent (see Table 2-1).
Table 2-2 summarizes the environmental impacts of the NOX control
techniques used in nitric acid manufacturing plants.
2.2 SUMMARY FOR ADIPIC ACID PLANTS
Four plants in the United States produce adipic acid. Three
of the plants, producing over 98 percent of the total output,
manufacture adipic acid using the cyclohexane-oxidation process.
The NOX absorption tower, common to all three plants, is the
major source of NOX emissions. Two control techniques are used
to reduce the level of NOX emissions in the absorber tail gas:
(1) extended adsorption and (2) thermal reduction. The fourth
plant, which produces adipic acid as a byproduct of caprolactam
production, uses the phenol-hydrogenation process. The major
sources of NOV emissions from this plant are nitric acid storage
Ji
tanks and the adipic acid reactors. Fumes containing NOX from
these sources are recovered by suction and recycled to the
caprolactam process. This section presents a summary of NOX
control performance, control cost data, and environmental impacts
for extended absorption and thermal reduction.
Table 2-3 is a summary of NOX emissions and a cost
comparison the two alternative control techniques used in the
three adipic acid plants. Annual uncontrolled NOV emissions were
Ji
calculated based on an uncontrolled emission factor of 53 Ib/ton
of adipic acid produced. Annual NOX emission reductions were
calculated using controlled emission factors estimated from
reported data and data obtained from an adipic acid screening
study performed in 1976. Table 2-4 summarizes the environmental
2-3
-------
TABLE 2-2. ENVIRONMENTAL AND ENERGY IMPACTS OF ALTERNATIVE
CONTROL TECHNIQUES USED IN NITRIC ACID PLANTS
Control
technique
Extended
adsorption
NSCR
SCR
Environmental impacts
Air
Reduces NOX; no secondary
impacts
Reduces NOX; possible HC
and CO emissions
Reduces NOX; possible
ammonia emissions
Liquid
None
None
None
Solid
None
Catalyst disposal
(3- to 8-yr life)
Catalyst disposal
(2- to 10-yr life)
Energy
Pumps and refrigeration
Natural gas consumption; heat
recovery possible
Pumps, fans; minimal energy
consumption
TABLE 2-3. NOX EMISSIONS AND COST COMPARISON OF ALTERNATIVE
CONTRdL TECHNIQUES USED IN ADIPIC ACID PLANTS
Plant size,
1CP tons/yr
190
300
350
Uncontrolled
NOX emissions,
tons/yr
5,040
7,950
9,280
Control technique
Extended
adsorption
Thermal reduction
Thermal reduction
NO
removed,
tons/yr
4,330
6,480
7,560
Costs, $103
Capital
2,830
7,050
8,000
Annual
425
3,240
3,720
Cost
effectiveness,
$/ton NOX
removed
98
500
492
TABLE 2-4. ENVIRONMENTAL AND ENERGY IMPACTS OF ALTERNATIVE
CONTROL TECHNIQUES USED IN ADIPIC ACID PLANTS
Control technique
Extended absorption
Thermal reduction
Environmental impacts
Air
Reduces NOX; no abatement
ofN2O
Reduces NOX; possible HC
and CO emissions
Liquid
None
None
Solid
None
None
Energy
Pumps and refrigeration
used
Natural gas consumption;
heat recovery possible
2-4
-------
and energy impacts of the NOX control techniques used in adipic
acid manufacturing plants.
2-5
-------
3.0 DESCRIPTION OF NITRIC/ADIPIC ACID MANUFACTURING
This chapter describes nitric and adipic acid manufacturing.
Section 3.1 deals primarily with "weak" nitric acid and its uses,
production processes, and industry characterization.
Concentrated nitric acid, though produced in considerably lesser
quantities, is also presented with a brief process description.
Adipic acid manufacturing is described in Section 3.2.
Similarly, this section characterizes the adipic acid industry,
discusses various uses of adipic acid, and describes the two
principal production processes.
3.1 NITRIC ACID MANUFACTURING
Nitric acid, HNO3, is considered to be one of the four most
important inorganic acids in the world and places in the top
10 chemicals produced in the United States. This nearly
colorless, liquid acid is (1) a strong acid due to its high
proportion of hydrogen ion, (2) a powerful oxidizing agent,
attacking most metals except gold and the platinum metals, and
(3) a source of fixed nitrogen, which is particularly important
to the fertilizer industry.1
3.1.1 Uses and Industry Characterization
The largest use, about 70 percent, of nitric acid is in
producing ammonium nitrate. This compound is primarily used for
fertilizer.
The second largest use of nitric acid, consuming 5 to
10 percent, is for organic oxidation in adipic acid
manufacturing. Terepthalic acid (an intermediate used in
polyester) and other organic compounds are also obtained from
organic oxidation using nitric acid.2'3'4 Nitric acid is also
used commercially for organic nitrations. A principal use is for
3-1
-------
nitrations in explosives manufacturing, but nitric acid nitration
is also used extensively in producing chemical intermediates such
as nitrobenzene and dinitrotoluenes.
In 1990 there were 67 nitric acid production facilities in
the United States, including government-owned munitions plants.
Twenty-four of these plants had a capacity of at least
180,000 tons per year, as compared to only 13 plants with such
capacity in 1984. Total plant capacity was about 11.3 million
tons of nitric acid as of January 1990.4'5 Actual production has
remained steady from 1984 to 1988, with an average annual
production of about 7.5 million tons of acid.6
Since a principal use of nitric acid is to produce ammonium
nitrate for fertilizer, the heaviest concentrations of nitric
acid production facilities are located in agricultural regions,
primarily in the Midwest, the South Central, and the Gulf States.
3.1.2 Production Process
Nitric acid is commercially available in two forms: weak
(50 to 70 percent nitric acid) and concentrated (greater than
95 percent nitric acid). Different processes are required to
produce these two forms of acid. For its many uses, weak nitric
acid is produced in far greater quantities than is the
concentrated form. Concentrated nitric acid production is
discussed in Section 3.1.4.
Virtually all commercial production of weak nitric acid in
the United States utilizes three common steps: (1) catalytic
oxidation of ammonia (NH3) to nitric oxide (NO), (2) oxidation of
nitric oxide with air to nitrogen dioxide (NO2), and
(3) absorption of nitrogen dioxide in water to produce "weak"
nitric acid.2 The basic process is shown in Figure 3-1.
3.1.2.1 Oxidation of Ammonia. The first step of the acid
production process involves oxidizing anhydrous ammonia over a
platinum-rhodium gauze catalyst to produce nitric oxide and
water. The exothermic reaction occurs as follows:8
4NH3 + SO2 ~^ 4NO + 6H20 + heat
3-2
-------
VT
1/1
u4 ae
0 Ul
ii
r
OB
^U
s
Si
o
So
si
(0
to
0)
o
o
M
cu
c
o
•H
4J
O
T3
O
o
rtJ
o
•H
•P
-H
c
o
W
(0
09
I
n
0)
^
3
Cn
3-3
-------
This extremely rapid reaction proceeds almost to completion,
evolving 906 kilojoules per mole (kJ/mole) (859 British thermal
units per mole [Btu/mole]) of heat. Typical ammonia conversion
efficiency ranges from 93 to 98 percent with good reactor
design.8
Air is compressed, filtered, and preheated by passing
through a heat exchanger. The air is mixed with vaporized
anhydrous ammonia and passed to the converter. Since the
explosive limit of ammonia is approached at concentrations
greater than 12 mole percent, plant operation is normally
maintained at 9.5 to 10.5 mole percent.9 In the converter, the
ammonia-air mixture is catalytically converted to nitric oxide
and excess air. The most common catalyst consists of 90 percent
platinum and 10 percent rhodium gauze constructed from squares of
fine wire.9 Up to 5 percent palladium is used to reduce costs.2
Operating temperature and pressure in the converter have
been shown to have an influence on ammonia conversion
efficiency.8 Generally, reaction efficiency increases with gauze
temperature. Oxidation temperatures typically range from 750° to
900°C (1380° to 1650°F). Higher catalyst temperatures increase
reaction selectivity toward NO production, while lower catalyst
temperatures are more selective toward less useful nitrogen (N2)
and nitrous oxide (N20).9 The high-temperature advantage is
offset by the increased loss of the precious metal catalyst.
Industrial experience has demonstrated and the industry has
generally accepted conversion efficiency values of 98 percent for
atmospheric pressure plants at 850°C (1560°F) and 96 percent for
plants operating at 0.8 megaPascals (MPa) (8 atmospheres [atin])
and 900°C (1650°F).2
As mentioned earlier, the ammonia oxidation reaction is
highly exothermic. In a well-designed plant, the heat byproduct
is usually recovered and utilized for steam generation in a waste
heat boiler. The steam can be used for liquid ammonia
evaporation and air preheat in addition to nonprocess plant
requirements.
3-4
-------
As higher temperatures are used, it becomes necessary to
capture platinum lost from the catalyst. Consequently, a
platinum recovery unit is frequently installed on the cold side
of the waste heat boiler. The recovery unit, composed of
ceramic-fiber filters, is capable of capturing 50 to 75 percent
of the lost platinum.10
3.1.2.2 Oxidation of Nitric Oxide. The nitric oxide formed
during the ammonia oxidation process is cooled in the cooler/
condenser apparatus, where it reacts noncatalytically with oxygen
to form nitrogen dioxide and its liquid dimer, dinitrogen
tetroxide.4 The exothermic reaction, evolving 113 kJ/mole
(107 Btu/mole), proceeds as follows:3
2NO + 02 "«— 2N02 •<— N204 + heat
This slow, homogeneous reaction is highly temperature- and
pressure-dependent. Lower temperatures, below 38°C (100°F), and
higher pressures, up to 800 Jcilopascals (kPa) (8 atm) , ensure
maximum production of NO2 and minimum reaction time.4
Furthermore, lower temperatures and higher pressures shift the
reaction to the production of N2O4, preventing the reverse
reaction (dissociation to NO and O2) from occurring.2
3.1.2.3 Absorption of Nitrogen Dioxide. The final step for
producing weak nitric acid involves the absorption of NO2 and
N204 in water to form nitric acid (as N204 is absorbed, it
releases gaseous NO) . The rate of this reaction is controlled by
three steps: (1) the oxidation of nitrogen oxide to NO2 in the
gas phase, (2) the physical diffusion of the reacting oxides from
the gas phase to the liquid phase, and (3) the chemical reaction
in the liquid phase.7 The exothermic reaction, evolving
135 kJ/mole (128 Btu/mole), proceeds as follows:2
3N02(g) + H20(0 2HN03(aq) + N0(g) + heat
3-5
-------
The absorption process takes place in a stainless steel
tower containing numerous layers of either bubble cap or sieve
trays. The number of trays varies according to pressure, acid
strength, gas composition, and operating temperature. Nitrogen
dioxide gas from the cooler/condenser effluent is introduced at
the bottom of the absorption tower, while the liquid dinitrogen
tetroxide enters at a point higher up the tower. Deionized
process water is added at the top, and the gas flows
countercurrent to both liquids. Oxidation occurs in the free
space between the trays, while absorption takes place in the
trays. Because of the high order of the oxidation process in
absorbers, roughly one-half the volume of the absorber is
required to absorb the final 3 percent of nitrogen oxide gas
concentration.9 Because lower temperatures are favorable for
maximum absorption, cooling coils are placed in the trays.
Nitric acid in concentrations of 55 to 65 percent is withdrawn at
the bottom of the tower.
Secondary air is used to improve oxidation in the absorption
tower and to bleach remaining nitrogen oxides from the product
acid. Absorption efficiency is further increased by utilizing
high operating pressure in the absorption process. High-pressure
absorption improves efficiency and increases the overall
absorption rate.
Absorber tail gas is reheated using recovered process heat
and expanded through a power recovery turbine. In a well-
designed plant, the exhaust gas turbine can supply all the power
needed for air compression with excess steam available for
export.10
3.1.3 Plant Design
Corrosive effects of nitric acid under pressure precluded
the use of pressures greater than atmospheric in early plant
designs. With the advent of corrosion-resistant materials,
nitric acid producers were able to take advantage of the
favorable effects of increased pressure in the NO oxidation and
absorption processes. All modern plants incorporate increased
pressure at some point in the process. Currently, two plant
3-6
-------
pressure designs are in use: single-pressure and dual-pressure
processes.
3.1.3.1 Single-Pressure Process. The single-pressure
process is the most commonly employed method of nitric acid
production in the United States. This process uses a single
pressure—low (atmospheric), medium (400 to 800 kPa [4 to
8 atm])—or high (800 to 1,400 kPa [8 to 14 atm]) in both the
ammonia oxidation and nitrogen oxides absorption phases of
production. The majority of new smaller capacity (less than
300 tons per day) nitric acid plants use the high-pressure
process. Operating at atmospheric pressure offers advantages
over higher-pressure processes: the catalyst lasts longer
(6 months) and ammonia conversion efficiency is increased. These
advantages are far outweighed, however, by low absorption and NO
oxidation rates (prompting the need for several large absorption
towers).8 Atmospheric plants still in existence generally
operate in a standby capacity, and no new atmospheric plants are
likely to be built.7 The medium-pressure process utilizes a
single higher pressure throughout the process. Though ammonia
conversion efficiency and catalyst life are somewhat decreased,
the economic benefits of medium pressure downstream are
substantial. Single-pressure-type plants require significantly
smaller, less expensive equipment for oxidation, heat exchange,
and absorption.7 A simplified single-pressure process flow
diagram is shown in Figure 3-2.
3.1.3.2 Dual-Pressure Process. The dual-pressure process
combines the attributes of low-pressure ammonia oxidation with
high-pressure absorption, thus optimizing the economic benefits
of each. Popularized in Europe, this process is finding
increasing utility in the United States. A simplified dual-
pressure process flow diagram is shown in Figure 3-3.
In the dual-pressure process, ammonia oxidation is usually
carried out at pressures from slightly negative to about (400 kBa
[4 atm]).2 This maintains the advantages of high ammonia
conversion efficiency and extended catalyst life. The heat of
reaction is recovered by the waste heat boiler, which supplies
3-7
-------
ii i
HI Jj ^|
IE _ _ d3auosav
23 | n A i
iSg ? T
<
0
<
m 2
3» i« i • GC
^j ™~
-
1 ^
0
L
3
T
j
OUXvaUaVAd
» j
^
^ g
n H
^ "
NITRIC ACII
CTOR AND ABSORBER OPERATE
AT THE SAME PRESSURE:
e nitric acid manufacturing process.1
Lu ^
ffi w
(0
C 0)
^ ^,
J ?
0)
r-H
C
•H
CO
I
•
rg
1
a)
3 -S1
c fc
•* •
3-8
-------
H39HOS8V
oc
1
Q
8
o
I
OC
dOlVtiOdVAB
(O
HI
tr
a.
W
(0
0)
o
o
iH
cu
o
R)
T3
•H
O
(0
o
•H
*H
4J
•H
C
0)
(Q
(0
0)
3
Q
n
n
0)
^
3
CC
<
3-9
-------
steam for the turbine-driven compressor. After passing through
the cooler/condenser, the gases are compressed to the absorber
pressure of 800 to 1,400 kPa (8 to 14 atm). Absorption is
further enhanced by internal water cooling, which results in acid
concentrations up to 70 percent and absorber efficiency to
96 percent. Nitric acid formed in the absorber is usually routed
through an external bleacher where air is used to remove (bleach)
dissolved oxides of nitrogen. The bleacher gases are then
compressed and passed through the absorber. Using excess ammonia
oxidation heat, tail gas is reheated to about 200°C (392°F) and
expanded in the power-recovery turbine.*'^•^
Atmospheric ammonia conversion is limited (due to low gas
loading at atmospheric pressure) to about 100 tons per day of
equivalent acid.2'9 Consequently, for large plants, several
ammonia converters and waste heat boilers are required.
Moreover, nitrous gas compression requires the use of stainless-
steel compressors. These costs require an investment for dual-
pressure plants from one and one-half to two times the amount for
single-pressure plants. However, these costs are offset by
improved ammonia efficiency, reduction of platinum catalyst loss,
higher absorption efficiency, and higher power recovery.^'7
3.1.4 Concentrated Nitric Acid Process
In some instances, such as organic nitrations, nitric acid
concentrations as high as 99 percent are required. Nitric acid
forms an azeotrope with water at 68.8 weight percent (simple
distillation will not separate the water from the acid). The
method most commonly employed in the United States for attaining
highly concentrated nitric acid is extractive distillation.
Another method, the direct strong nitric process, can produce 95
to 99 percent nitric acid directly from ammonia.2'8 However,
this process has found limited commercial application in the
United States.
The extractive distillation method uses concentrated
sulphuric acid as a dehydrating agent to produce 98 to 99 percent
nitric acid. The process is shown in Figure 3-4. Strong
sulfuric acid (typically 60 percent concentration) mixed with 55
3-10
-------
TAIL GAS TO
ATMOSPHERE (VOLUME S)
COUNTERCURRENT
CONDENSER
VAPJiR
FEED
w I
HN03 '
EHYORATING
COLUMN
\
CONOENSATE
BLEACHER
NO'N-~
CONDENSABLE
GAi
TO COOLER AND STORAGE
ES
74.3 N2
20.402
1.0NO + N02
4JH20
r
ABSORFnON
COLUMN
VAPOR
STEAM-
COIL -
.AIR
BOILER
TO COOLER
MHHHOj
Figure 3-4.
Nitric acid concentration using extractive
distillation.13
3-11
-------
to 65 percent nitric acid enters the top of a packed tower and
flows countercurrent to ascending vapors. Ninety-nine percent
nitric acid vapor containing small amounts of NOX is recovered at
the top of the tower. The vapors are then bleached and
condensed, leaving weak nitric acid, NOX, and oxygen. The gases
are subsequently passed to an absorber, where they are converted
to nitric acid and recovered.2'8
The direct strong nitric acid process (DSN) produces
concentrated nitric acid directly from ammonia. While several
DSN processes exist, the Uhde process has demonstrated commercial
application in the United States. The Uhde process is shown in
Figure 3-5. Air and gaseous ammonia are mixed and reacted. Heat
of reaction produces steam in the burner/waste-heat boiler. Upon
cooling, the reaction products condense to form weak nitric acid.
After separating the liquid nitric acid, the remaining NO is
oxidized to NO2 by passing through two oxidizing columns. The
vapors are then compressed and cooled to form liquid dinitrogen
tetroxide. At a pressure of 5 MPa (50 atm), the liquid N204
reacts with 02 to form strong nitric acid of 95 to 99 percent
concentration. Because NOX from the absorber is a valuable raw
material, tail gas emissions are scrubbed with water and
condensed N204. The scrubber effluent is then mixed with the
concentrated acid from the absorber column. The combined product
is oxidized in the reactor vessel, cooled, and bleached,
producing concentrated nitric acid.8
3.2 ADIPIC ACID MANUFACTURING
Adipic acid, COOH-(CH2)3-COOH, was the 48th-highest-volume
chemical produced in the United States in 1985 and is considered
one of the most important commercially available aliphatic
dicarboxylic acids. Typically, it is a white crystalline solid,
soluble in alcohol and acetone.15
3.2.1 Uses and Industry Characterization
Ninety percent of adipic acid manufactured in the United
States is used to produce nylon 6/6 fiber and plastics. Esters
used for plasticizers and lubricants are the next largest
3-12
-------
onv
3iVSN3ClNOOiai3V3iaJLIN
«-'a>«a^
2Sa5*|iij*
31
W
W
0)
o
o
a
o
•H
k
-P
O
M
4J
CO
-M
O
0)
0)
-C-*
ilSlssS
1 ^o x < C vi
ailVM 033d
/
vfc
*
t
uaiioa
VINCWWV
Sn03SV9
'
c
•H
5 W QJ
a 3 °
< °
s o a
•H
. -P
-------
consumer. Small quantities of adipic acid are also used as food
acidulants.8'16
There are four adipic acid manufacturing facilities in
operation: (1) Allied-Signal, Inc., in Hopewell, Virginia, with
an annual production capacity of 15,000 tons; (2,3) DuPont
Chemicals in Orange and Victoria, Texas, with annual production
capacities of 190,000 and 350,000 tons, respectively; and
(4) Monsanto Chemical Company in Pensacola, Florida, with an
annual production capacity of 300,000 tons.5 Total annual
production reached 865,000 tons in 1989.17
3.2.2 Production Process
Two methods of producing adipic acid are currently in use.
The basic process is shown in Figure 3-6. Ninety-eight percent
of adipic acid produced in the United States is manufactured from
cyclohexane in a continuous operation. Cyclohexane is air-
oxidized, producing a cyclohexanol-cyclohexanone (ketone-alcohol,
or KA) mixture. This mixture is then catalytically oxidized
using 50 to 60 percent nitric acid, producing adipic acid.
Phenol hydrogenation followed by nitric acid oxidation is the
lesser-used method.8/16
3.2.2.1 Oxidation of Cvclohexane. In commercial use, two
approaches predominate the air oxidation of cyclohexane process:
cobalt-catalyzed oxidation and borate-promoted oxidation. A
third method, the high-peroxide process, has found limited
commercial use.
Cobalt-catalyzed air oxidation of cyclohexane is the most
widely used method for producing adipic acid. Cyclohexane is
oxidized with air at 150° to 160°C (302° to 320°F) and 810 to
1,013 kPa (about 8 to 10 atm) in the presence of the cobalt
catalyst in a sparged reactor or multistaged column contactor.
Several oxidation stages are usually necessary to avoid over-
oxidizing the KA mixture. Oxidizer effluent is distilled to
recover unconverted cyclohexane then recycled to the reactor
feed. The resultant KA mixture may then be distilled for
improved quality before being sent to the nitric acid oxidation
3-14
-------
5
1
a .
P
Ul
ui 5
51
if
1 s
* Q«g
o *
e
2
in
«H
i>
s u
Z m
o
I £
y Q a"
S| * §8
I %
g
H Ul
ii
1
Ii
> Q
si
1
||K
318^
2£gs
lif
1
ee
fc
K
M
1
l~
U O
-i
3
C?
•H
fa
3-15
-------
stage. This process yields 75 to 80 mole percent KA, with a
ketone to alcohol ratio of 1:2.16
Borate-promoted oxidation demonstrates improved alcohol
yields. Boric acid reacts with cyclohexanol to produce a borate
that subsequently decomposes to a thermally stable borate ester,
highly resistant to further oxidation or degradation. Another
key feature of the borate-promoted oxidation system is the
removal of byproduct water from the reactors using inert gas and
hot cyclohexane vapor. Reaction yields of 87 percent and a K:A
ratio of 1:10 have been achieved.16
The high-peroxide process is an alternative to maximizing
selectivity. Noncatalytic oxidation in a passivated reactor
results in maximum production of cyclohexylhydroperoxide. This
is followed by controlled decomposition to KA. Achievable
reaction yield is as high as 84 percent KA.16
3.2.2.2 Phenol Hvdroaenation. Phenol hydrogenation is
another method of producing cyclohexanol and cyclohexanone.
Molten phenol is typically hydrogenated at 140°C (284°F) and 200
to 1800 kPa (2 to 18 atm) hydrogen pressure over a nickel,
copper, or chromium oxide catalyst. These catalysts
predominantly yield cyclohexanol. Cyclohexanone, typically an
intermediate product for manufacturing caprolactam, is favored by
using a palladium catalyst. Cyclohexanol yield is typically 97
to 99 percent; however, given sufficient reactor residence time,
conversion efficiency of 99+ percent is achievable.16'19'20
3.2.2.3 Nitric Acid Oxidation of Cyclohexanol-
Cyclohexanone. The second step in commercial production of
adipic acid is nitric acid oxidation of the cyclohexanol-
cyclohexanone mixture. The reaction proceeds as follows:8
cyclohexanol + nitric acid ~~^ adipic acid + NO2 + H2O + heat
cyclohexanone + nitric acid ~~ adipic acid + NOX + H20 + heat
3-16
-------
As the reaction is highly exothermic, heat of reaction is usually
dissipated by maintaining a high ratio (40:1) of nitric acid to
KA mixture.19
Nitric acid (50 to 60 percent) and a copper-vanadium
catalyst are reacted with the KA mixture in a reactor vessel at
60° to 80°C and 0.1 to 0.4 MPa. Conversion yields of 92 to
96 percent are attainable when using high-purity KA feedstock.
Upon reaction, nitric acid is reduced to nitrogen oxides: N02,
NO, N20, and N2> The dissolved oxides are stripped from the
reaction product using air in a bleaching column and subsequently
recovered as nitric acid in an absorption tower.16'19
The stripped adipic acid/nitric acid solution is chilled and
sent to a crystallizer, where crystals of adipic acid are formed.
The crystals are separated from the mother liquor in a centrifuge
and transported to the adipic acid drying and/or melting
facilities. The mother liquor is separated from the remaining
uncrystallized adipic acid in the product still and recycled to
the reactors.
3.3 REFERENCES FOR CHAPTER 3
1. Keleti, C. (ed.). The History of Nitric Acid. In: Nitric
Acid and Fertilizer Nitrates. New York, Marcel Dekker, Inc.
1985. pp. 2, 19-23.
2. Newman, D.J. Nitric Acid. In: Kirk-Othmer Encyclopedia of
Chemical Technology. New York, John Wiley & Sons. 1981.
pp. 853-871.
3. Control Techniques for Nitrogen Oxides Emissions From
Stationary Sources: Revised 2nd Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-83-002. January 1983.
Ch. 6: pp. 35-46.
4. Review of New Source Performance Standards for Nitric Acid
Plants. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/8-84-011.
April 1984. Ch. 2: pp. 1-13.
5. SRI International. Directory of Chemical Producers, United
States of America. Menlo Park, CA. 1990. pp. 809-811.
3-17
-------
6. Inorganic Fertilizer Materials and Related Products. In:
Current Industrial Reports. U.S. Department of Commerce,
Bureau of Census. Washington, DC. 1988. 3 pp.
7. Nitric Acid Plant Inspection Guide. Prepared for U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-340/1-84-013. August 1984. p. 9.
8. Reference 1, pp. 31-71.
9. Ohsol, E.O. Nitric Acid. In: Encyclopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1990. pp. 150-155.
10. Nitric Acid Plant Inspection Guide. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. EPA-340/1-84-013. August 1984.
11. Reference 4, p. 7.
12. Reference 4, p. 9.
13. Reference 3, p. 8
14. Reference 3, p. 9.
15. Sax, N.I., and R.J. Lewis, Sr. (eds.). Hawley's Condensed
Chemical Dictionary. New York, Van Nostrand Reinhold
Company. 1987. p. 24.
16. Adipic Acid. In: Encyclopedia of Chemical Technology,
Kirk-Othmer. New York, John Wiley & Sons. 1978.
pp. 513-528.
17. Adipic Acid. In: Synthetic Organic Chemicals, U.S. Sales
and Production. U.S. International Trade Commission.
Washington, DC. 1989. 1 p.
18. Compilation of Air Pollutant Emission Factors: Volume l:
Stationary Point and Area Sources. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. AP-42. September 1985. p. 5.1-2.
19. Luedeke, V.D. Adipic Acid. In: Encyclopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1977. pp. 128-146.
20. Cyclohexanol/Cyclohexanone. In: Organic Chemical
Manufacturing: Volume 6: Selected Processes. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-80-028a. December 1980.
pp. III-2-7.
3-18
-------
21. Screening Study to Determine Need for Standards of
Performance for New Adipic Acid Plants: Final Report.
GCA/Technology Division. Bedford, MA. Publication
No. GCA-TR-76-16-G. July 1976. p. 12.
3-19
-------
4.0 CHARACTERIZATION OF NOX EMISSIONS
This section presents a description of NOX formation and
emission levels from nitric and adipic acid manufacturing.
Section 4.1 describes uncontrolled NOX emissions from nitric acid
manufacturing. The uncontrolled NOX emissions from manufacturing
adipic acid are described in Section 4.2.
4.1 NOV EMISSIONS FROM NITRIC ACID MANUFACTURING
Ji
Nitric acid production is one of the larger chemical
industry sources of NOX. Unlike NOX found in combustion flue
gas, NOX from nitric acid production is part of the process
stream and is recoverable with some economic value. Vent gas
containing NOX is released to the atmosphere when the gas becomes
too impure to recycle or too low in concentration for recovery to
be economically practical.1
Section 4.1.1 describes how NOX is formed as a result of the
basic ammonia oxidation process of nitric acid manufacturing.
Several factors affect the level of NOX emissions from a typical
nitric acid plant. These factors are presented in Section 4.1.2.
Finally, Section 4.1.3 discusses the sources of NOX emissions and
typical levels of uncontrolled NOX emissions. Furthermore, this
section describes how tail gas plume color and opacity are
related to the level of NOX in the gas.
4.1.1 NOX Formation
The chemical reactions for each of the nitric acid
production process steps (Chapter 3) demonstrate that NOV must
Ji
first be created before nitric acid can be produced. The first
reaction,
4NH3 + 502 -* 4NO + 6H20 + heat, Eq. 1
4-1
-------
shows NO forming from the reaction of NH3 and air. The NO is
then oxidized in the second step,
2NO + O2 -» 2NO2 ?=iN2O4 + heat, Eq. 2
producing NO2. The NO2 is subsequently absorbed in water to
produce nitric acid. However, as the absorption reaction,
3NO2(g) + H20(P) <^2HNO3(aq) + NO(g) + heat, Eq. 3
shows, one mole of NO is produced for every three moles of N02
absorbed, making complete absorption of the NOX impossible. The
unabsorbed NOX, if not controlled, is emitted in the absorber
tail gas.
4.1.2 Factors Affecting NO.. Emission Levels
Many interrelated factors affect the efficiency of the
absorber and the level of uncontrolled NOX emissions. These
factors are described below.
As noted in the previous section, the production of nitric
acid necessarily results in the formation of NO. Using bleacher
air, NO must be reoxidized to NO2 prior to being reabsorbed. Two
limiting factors are present. First, reoxidation of NO to N02 is
a very slow reaction. As more air is added, the reaction becomes
increasingly slower as the reactants become diluted with excess
nitrogen. Second, increased temperatures due to the exothermic
absorption reaction tend to reverse the reaction equation
(Equation 3).2 These factors impose economic limits on
absorption efficiency and, consequently, must be addressed when
considering absorber design.
Maximum absorber efficiency is a primary concern of process
designers. Higher absorber efficiency translates to lower NOX
emissions. Maximum efficiency is achieved by operating at low
temperatures, high pressure, low throughput, and low acid
strength with a long residence time.2 Altering any of these
design criteria affects the level of NOX emissions. Furthermore,
proper operation and maintenance practices are vital to
minimizing Npx emissions.
Low temperature (less than 38°C [100°F]) is a key factor for
high absorption efficiency but is also one that is difficult and
expensive to control.3 The difficulty of maintaining a low
4-2
-------
temperature arises from the addition of heat from two sources:
heat of reaction and ambient heat. Heat from the exothermic
absorption reaction is carried away by cooling water that is
circulated through the absorption tower. However, high ambient
temperature reduces the heat removal capacity of heat transfer
equipment.4 This, in turn, reduces absorber efficiency and
increases NOX emissions.
Operating pressure is another important consideration for
increasing absorber efficiency. Gas volume in the tower
contracts as the absorption reaction proceeds; therefore,
completion of the reaction is aided by increased pressure.2 As
mentioned in Chapter 3, most new nitric acid plants use high
pressure (800 to 1,400 kPa [8 to 14 atm]) in the absorption tower
to increase absorber efficiency.
Nitric acid plants are designed for a specified production
rate, or throughput. Throughput ranges from 50 to 1,000 tons per
day (100 percent nitric acid). Operating outside of the optimal
throughput affects the levels of NOX emissions. Increasing the
production rate typically increases the NOX emission rate by
decreasing residence time in the absorption tower. Typical
residence time for absorption of NOV in water is on the order of
Ji
seconds for NO2 absorption and minutes for NO+O2 absorption
reaction (NO does not absorb into water).5 Decreasing the
residence time minimizes the oxidation of NO to NO2 and decreases
the absorption of NO2. Conversely, operating below design
throughput increases residence time, and lower NOV emissions
A.
would be expected.6
It is not always true that NOX emissions are a function of
plant rate. Since the hot gas expander acts as a restriction
device in the tail gas system, increasing the rate actually
increases the pressure and conversely lowers emissions because of
greater absorption efficiency. The absorber volume requirement
is a function of the cube of the absorber pressure; therefore,
unless the tail gas is vented or bypassed around the expander,
NOX will be lower leaving the absorber if all other variables
remain the same.7
4-3
-------
Acid strength is another factor designed into the process.
Increasing acid strength beyond design specifications (e.g.,
60 percent nitric acid) typically increases NOX emissions. Lower
emissions would be expected from reduced acid strength.6
Finally, good maintenance practices and careful control of
operations play important roles in reducing emissions of NOV.
A.
Repairing internal leaks and performing regular equipment
maintenance help to ensure that NOX levels are kept to a design
minimum.1
4.1.3 Uncontrolled NO.. Emission Levels
a
The main source of atmospheric NOV emissions from nitric
4t
acid manufacturing is the tail gas from the absorption
tower.1'6'^ Uncontrolled NOX emission levels vary from plant to
plant due to differences in plant design and other factors
previously discussed. Typically uncontrolled emission levels of
3,000 ppm (with equal amounts of NO and NO2) are found in low-
pressure (atmospheric) plants. Medium- and high-pressure plants
exhibit lower uncontrolled emission levels, 1,000 to 2,000 ppm,
due to improved absorption efficiency.6'8'9 These levels apply
to single- and dual-pressure plants.
Typical uncontrolled NOX emissions factors range from 7 to
43 kg/Mg (14 to 86 Ib/ton) of acid (expressed as 100 percent
HNC^). This range includes atmospheric, medium-, and high-
pressure plants. Factors that affect the emission rate are
discussed in Section 4.1.2. The average emission factor (from
AP-42) for uncontrolled tail gas emissions is 22 kg/Mg
(43 Ib/ton) of acid.9 As discussed in Chapter 3
(Section 3.1.3.1), atmospheric plants operate only in a standby
capacity and no new atmospheric plants are likely to be built.
Using the average NOX concentration (1,500 ppm) for medium- and
high-pressure plants, an uncontrolled NOX emission factor of
10 kg/metric ton (20 Ib/ton) can be calculated. This emission
factor will be used throughout this text for uncontrolled NOV
X.
emissions from nitric acid plants. This emission factor is
typical for steady-state, continuous operation. Startups,
shutdowns, and malfunctions increase the uncontrolled emission
4-4
-------
levels.6 A typical NOX emission level from concentrated nitric
acid production is 5 kg/Mg (10 Ib/ton) of 98 percent nitric
acid.9
Color and opacity of the tail gas plume are indicators of
the presence and concentration of NOX, specifically N02 (NO is
colorless). A reddish-brown plume reveals the presence of N02.
Plume opacity is directly related to N02 concentration and stack
diameter. The rule of thumb is that the stack plume has a
reddish-brown color when the NO2 concentration exceeds 6,100 ppm
divided by the stack diameter in centimeters.1
Nitrogen oxides emissions may occur during filling of
storage tanks.9 However, there is no information on the
magnitude of these emissions.
4.2 NOV EMISSIONS FROM ADIPIC ACID MANUFACTURING
a
Nitrogen oxides created in the adipic acid production
process, like those created in the production of nitric acid, are
considered part of the process stream and are recoverable with
some economic value. Tail gas from the NOV absorber is released
X>
to the atmosphere when the gas becomes too low in concentration
for recovery to be economically practical.
Section 4.2.1 describes how NOX is formed as a result of the
KA oxidation process (using nitric acid) used in producing adipic
acid. Factors affecting the level of uncontrolled NOX emissions
in the absorber tail gas are discussed in Section 4.2.2.
Section 4.2.3 describes the source of NOX emissions and presents
data showing typical levels of uncontrolled NOV emissions.
Ji
4.2.1 NOX Formation
Adipic acid is produced by oxidizing a ketone-alcohol
mixture (cyclohexanone-cyclohexanol) using nitric acid as
follows:10'11
Cyclohexanone + nitric acid -* adipic acid + NOX + water Eq. 1
Cyclohexanol + nitric acid -» adipic acid + NOV + water Eq. 2
a
The oxidation process creates oxides of nitrogen in the form of
NO, N02, and N20, with some N2 also forming.11'12
The NOX is stripped from the reaction product using air in a
bleaching column, and NO and NO2 are subsequently recovered as
4-5
-------
nitric acid in an absorption tower. The N2 and N2O are released
to the atmosphere. The absorption tower functions in the same
manner as the absorption tower used in the nitric acid production
process. Nitrogen oxides, entering the lower portion of the
absorber, flow countercurrent to a water stream, which enters
near the top of the absorber. Unabsorbed NOX is vented from the
top while diluted nitric acid is withdrawn from the bottom of the
absorber and recycled to the adipic acid process.
4.2.2 Factors Affecting NO^ Emission Levels
The absorption tower used in adipic acid production
functions in the same manner as the NOX absorber used in nitric
acid production. Consequently, factors affecting uncontrolled
NOX emissions from both absorbers are expected to be similar.
These factors are described in detail in Section 4.1.2 and
include the following: high absorber pressure, low temperature
in the absorber, long residence time, and low throughput.
4.2.3 Uncontrolled NO^ Emission Levels
The main source of atmospheric NOX emissions from adipic
acid manufacturing is the tail gas from the absorption
tower.10'11 Other sources of NOX emissions include nitric acid
storage tanks and off-gas from the adipic acid refining process.
However, NOV emissions from these two sources are minor in
Ji
comparison. All four adipic acid manufacturing plants were
contacted in order to obtain uncontrolled NOX emissions data.
The data received did not contain any uncontrolled NOX emissions
factors. However, one plant did report uncontrolled NOX
concentrations of 7,000 parts per million by volume (ppmv) in the
tail gas of the KA oxidation absorber.13 The 1976 screening
study reported uncontrolled NOX emission rates for two plants
(capacities of 150,000 and 175,000 tons/yr of adipic acid) as
1,080 and 1,400 pounds per hour.5
The AP-42 cites an emission factor of 27 kg per metric ton
of adipic acid produced (53 Ib/ton) for uncontrolled NOX
emissions in the absorption tower tail gas.1^ This emission
factor represents NOX in the form of NO and N02 only. Large
quantities of nitrous oxide (N2O) are also formed during the
4-6
-------
oxidation process. The effect of N2O on the ozone layer is
currently under investigation by the Air and Energy Engineering
Research Laboratory. However, one plant reports that the N20
produced at that facility is recovered by a private company to be
used in dental offices.15
The adipic acid refining process, which includes chilling,
crystallizing, and centrifuging, is a minor source of NOX
emissions. The AP-42 cites an uncontrolled NOX emission factor
of 0.3 kg per metric ton (0.6 Ib/ton) of adipic acid produced for
the refining process.^ No emissions factor for the nitric acid
storage tanks was reported; however, one plant cited an
uncontrolled NOX concentration of 9,000 ppmv.13
4.3 REFERENCES FOR CHAPTER 4
1. Control Techniques for Nitrogen Oxides Emissions From
Stationary Sources: Revised 2nd Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-83-002. January 1983.
Chapter 6: pp. 1, 10-11.
2. Blackwood, T. R., and B. B. Crocker. Source Control—
Chemical. In: Handbook of Air Pollution Control
Technology, S. Calvert and H. M. Englund (eds.). New York,
John Wiley and Sons. 1984. p. 654.
3. Ohsol, E. 0., Nitric Acid. In: Encyclopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1990. pp. 150-155.
4. Telecon. Vick, K., Farmland Industries, with Lazzo, D.,
Midwest Research Institute. February 27, 1991. NOX
controls for nitric acid plants.
5. Screening Study to Determine Need for Standards of
Performance for New Adipic Acid Plants: Final Report.
GCA/Technology Division. Bedford, MA. Publication
No. GCA-TR-76-16-G. July 1976. p. 34
6. Nitric Acid Plant Inspection Guide. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. EPA-340/1-84-013. August 1984.
7. Letter from Boyd, D. E., Weatherly, Inc., to Neuffer, B.,
EPA/ISB. October 9, 1991. Comments on draft ACT.
4-7
-------
8. Review of New Source Performance Standards for Nitric Acid
Plants. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/8-84-011.
April 1984. Chapter 2: pp. 10, 11.
9. Compilation of Air Pollutant Emission Factors: Volume I:
Stationary Point and Area Sources. U. S. Environmental
Protection Agency. Research Triangle Park, NC.
September 1985. Section 5.9:pp. 1-6.
10. Reference 1, pp. 6-39.
11. Reference 4, p. 11.
12. Luedeke, V. D. Adipic Acid. In: Encylopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1977. p. 137.
13. Response to questionnaire from Miller, M., E.I. DuPont de
Nemours and Company, to Neuffer, B., EPA/ISB. June 18,
1991. NOX controls for adipic acid plants.
14. Reference 7, p. 5.1-4.
15. Telecon. McCloud, B., Monsanto Chemical Company, with
Neuffer, B., EPA/ISB. April 10, 1991. NOX controls for
adipic acid plants.
4-8
-------
5.0 CONTROL TECHNIQUES FOR NITROGEN OXIDES FROM
NITRIC/ADIPIC ACID MANUFACTURING
This chapter describes the techniques used to control NOV
•A
emissions from nitric and adipic acid manufacturing plants.
Section 5.1 discusses control techniques for nitric acid
manufacturing and Section 5.2 discusses control techniques for
adipic acid manufacturing. Each of these sections describes the
control techniques, discusses factors affecting the performance
of each control, and presents data illustrating the achieved
levels of control for each device.
5.1 NITRIC ACID MANUFACTURING
Several control techniques have been demonstrated that
reduce NOX emissions from nitric acid manufacturing plants. Of
the available control techniques, three methods are used
predominantly: (1) extended absorption, (2) nonselective
catalytic reduction (NSCR), and (3) selective catalytic reduction
(SCR). All three of these control techniques are suitable for
new and existing plant applications. Sections 5.1.1, 5.1.2, and
5.1.3 describe these control techniques, discuss factors
affecting their performance, and provide data that demonstrate
the level of achievable NOX control. In Section 5.1.4, a table
is presented that summarizes the level of control and control
efficiency. Section 5.1.5 briefly describes other NOV control
Ji
techniques with more limited use: (1) wet chemical scrubbing
(ammonia, urea, and caustic), (2) chilled absorption (CDL/VITOK
and TVA), and (3) molecular sieve adsorption.
5.1.1 Extended Absorption
Extended absorption reduces NOX emissions by increasing
absorption efficiency and is achieved by either installing a
5-1
-------
single large tower, extending the height of an existing
absorption tower, or by adding a second tower in series with the
existing tower.1 Increasing the volume and the number of trays
in the absorber results in more NOX being recovered as nitric
acid (1 to 1.5 percent more acid) and reduced emission levels.2
Extended absorption can be applied to new and existing plants;
however, it is considered an add-on control only when applied to
existing plants. Typically, retrofit applications involve adding
a second tower in series with an existing tower. New plants are
generally designed with a single large tower that is an integral
component of the new plant design. New nitric acid plants have
been constructed with absorption systems designed for
99.7+ percent NOX recovery.1
The following sections discuss extended absorption used as a
NOX control technique for nitric acid plants. Section 5.1.1.1
describes single- and dual-tower extended absorption systems.
Factors affecting the performance of extended absorption are
discussed in Section 5.1.1.2; and Section 5.1.1.3 presents
emissions test data and discusses NOV control performance.
4v
5.1.1.1 Description of Extended Absorption Systems.
Figure 5-1 is a flow diagram for a typical nitric acid plant with
an extended absorption system using a single large (typically 100
to 130 feet tall) tower.1'3 Following the normal ammonia oxida-
tion process as described in Chapter 3, NOX is absorbed in the
"extended" absorption tower. The lower portion (approximately
40 percent of the trays) of the tower is cooled by normal cooling
water available at the plant site. The remaining trays are
cooled by water or coolant to approximately 2° to 7°C (37° to
45°F), which is usually achieved by a closed-loop refrigeration
system using Freon or part of the plant ammonia vaporization
system.1'5'6 Absorber tail gas is then heated in a heat
exchanger, which utilizes the heat of the ammonia conversion
reaction. The heat is subsequently converted to power in a
turboexpander.
Figure 5-2 is a flow diagram for a nitric acid plant with an
extended absorption system using a second absorption tower. The
5-2
-------
*• ft
CC 111
ESftS
liis
\
3
Sj
5 IE
§5
ii
*\
( •»
« f • N
V i x
\ •
\
i .
1
i[_r
v*«
1
(
II
r
s
x" »x x'
/ \ \ ^ r \
' 1 1 N x 1
' i N '
XX ' ' XX 1
v , I % |
•
1
5
i
s
3
S
s
v_
1
§
c
^ /^""^
ii
>
\
1
1s
1
/
/
/
•C ^S
Ui U|
'N.
00
^Ul
K O
$1
K o 5
ill
4J
O
U
O
2
0)
O
(0
.0
0)
0)
to
o
s
0)
U
O
(0
A
(0
tJ
0)
a>
•P
x
w
"Z_
V
i
in
a)
b
5-3
-------
Of*
O •
0) CO
(0 4->
to
3 TJ
•H
J*
to u
CO M
.p
C-H
o c
•H
^ I I i
O rH
to o
•o o
0) U
T3
C X
0) O
-P 2
X
-------
second tower is the "extended" portion of the absorption system.
Following the normal ammonia oxidation process as described in
Chapter 3, NOX is absorbed in the first absorption tower. The
tail gas from the first absorber is routed to the base of the
second absorber. As the gas flows countercurrent to the process
water in the second absorber, the remaining NOX is absorbed to
form additional nitric acid. The weak acid from the second
absorber is then recycled to the upper trays of the first
absorber. Consequently, no liquid effluent waste is generated.
The weak acid entering the top of the first tower absorbs rising
NOV gases, producing the product nitric acid. Tail gas from the
Ji
second absorber is heated in a heat exchanger and recovered as
power generated in a turboexpander. In order to minimize the
size of the second absorption tower, inlet gas to the first
absorber is generally pressurized to at least 730 kPa (7.3 atm)
and additional cooling is provided. One company's process uses
two cooling water systems to chill both absorbers. The entire
second absorber and approximately one-third of the trays of the
first absorber are cooled by refrigerated water at about 7°C
(45°F). The remaining trays in the first absorber are cooled by
normal plant cooling water. 1'5/*}
5.1.1.2 Factors Affecting Performance. Specific operating
parameters must be precisely controlled in order for extended
absorption to reduce NOX emissions significantly. Because this
control technique is essentially an extension of the absorber, a
component common to all weak nitric acid production processes,
the factors that affect its performance are the same as those
that affect uncontrolled emissions levels as discussed in detail
in Chapter 4. These factors include maximum NOX absorption
efficiency achieved by operating at low temperature, high
pressure, low throughput and acid strength (i.e., throughput and
acid concentration within design specifications), and long
residence time.
5.1.1.3 Performance of Extended Absorption. Table 5-1
illustrates the levels to which extended absorption can reduce
NOV emissions from nitric acid plants. The emission factors are
a
5-5
-------
TABLE 5-1.
NITROGEN OXIDES EMISSIONS FROM NITRIC ACID PLANTS
USING EXTENDED ABSORPTION10
Plant
C
D
G
I
J
H
E
Absorber
Single
Single
Single
Single
Single
Dual
Dual
Absorber inlet
pressure, atm
9
9
NA
NA
9
9
7
Nitric acid
production
rate, tons/d
271
538
375
300
530
1,056
220
Acid
strength, %
56
57
62
55
56
54
57
Average
emission factor
Ib/ton acid
1.3
2.75
2.55
2.74
2.13
2.81
1.8
Control
efficiency, %a
97
93.6
94
93.6
95
93.5
96
NA = not available.
aThese figures calculated using average uncontrolled emissions level of 43 Ib/ton (from AP-42).
Notes: The following are provided for comparative purposes.
1. From AP-42, NOX emission levels from nitric acid plants
a. Emissions:
uncontrolled~22 kg/metric ton; 43 Ib/ton
extended absorption—0.9 kg/metric ton; 1.8 Ib/ton
b. Control efficiency:
uncontrolled—0 %
extended absorption—95.896
2. From NSPS, allowable NOX emission levels from nitric acid plants
Emissions:
1.5 kg/metric ton; 3.0 Ib/ton
5-6
-------
based on compliance tests (using EPA Method 7) performed on seven
new plants using extended absorption that are subject to the new
source performance standards (NSPS) since the 1979 review.
Actual production capacities during testing ranged from 200 to
960 metric tons (220 to 1,060 tons per day [tons/d]) expressed as
100 percent nitric acid. Acid concentration is similar for six
of the plants, ranging from 54 to 57 percent, while one plant
produces acid at a concentration of 62 percent. Five plants
operate with a single large absorption tower, and two use a
second tower.
The emission factors range from 0.59 to 1.28 kg of NOV per
a
metric ton (1.3 to 2.81 Ib/ton). No trends are indicated
relating NOX emission levels to plant size, production capacity,
or acid strength. Additionally, there is no correlation between
absorber design (single vs. dual) and controlled emission levels.
However, the emissions data do illustrate the effectiveness of
extended absorption on reducing NOX emissions. From AP-42, the
average uncontrolled emissions level for nitric acid plants is
22 kg per metric ton (43 Ib/ton) of nitric acid.9 Furthermore,
AP-42 gives an average control efficiency of 95.8 percent for
extended absorption. From the emissions data in Table 5-1, the
control efficiency for extended absorption at the seven plants
ranges from 93.5 to 97 percent. For further comparison, the data
demonstrate that for all seven plants, extended absorption
reduces NOX emissions below the NSPS level of 1.5 kg per metric
ton (3.0 pounds per ton).
5.1.2 Nonselective Catalytic Reduction
Nonselective catalytic reduction uses a fuel and a catalyst
to (1) consume free oxygen in the absorber tail gas, (2) convert
NO2 to NO for decolorizing the tail gas, and (3) reduce NO to
elemental nitrogen. The process is called nonselective because
the fuel first depletes all the oxygen present in the tail gas
and then removes the NOX. Nonselective catalytic reduction was
widely used in new plants between 1971 and 1977. It can achieve
higher NOX reductions than can extended absorption. However,
rapid fuel price escalations caused a decline in the use of NSCR
5-7
-------
for new nitric acid plants, many of which opted for extended
absorption.
Despite the associated high fuel costs, NSCR offers
advantages that continue to make it a viable option for new and
retrofit applications. Flexibility adds to the attractiveness of
NSCR, especially for retrofit considerations. An NSCR unit
generally can be used in conjunction with other NOX control
techniques. Furthermore, NSCR can be operated at any pressure.5
Additionally, heat generated by operating an NSCR unit can be
recovered in a waste heat boiler and a tail gas expander. The
heat recovered can supply the energy for process compression
needs with additional steam available for export.11
The following sections discuss NSCR used as a NOX control
technique for nitric acid plants. Section 5.1.2.1 describes an
NSCR system including its components and operation. Factors
affecting the performance of NSCR units are discussed in
Section 5.1.2.2, while Section 5.1.2.3 presents data and
discusses NOX control performance.
5.1.2.1 Description of Nonselective Catalytic Reduction
Systems. Figure 5-3 is a flow diagram for a typical nitric acid
plant using nonselective catalytic reduction. Absorber tail gas
is heated to the required ignition temperature using ammonia
converter effluent gas in a heat exchanger, and fuel (usually
natural gas) is added. Available reducing fuels and associated
ignition temperatures are as follows:5'6
Fuel Temperature. °C (°F)
Natural gas (methane) 450-480 (842-896)
Propane/butane/naphtha 340 (644)
Ammonia plant purge gas/hydrogen 250 (482)
Carbon monoxide 150-200 (302-392)
The gas/fuel mixture then passes through the catalytic reduction
unit where the fuel reacts in the presence of a catalyst with NOX
and oxygen to form elemental nitrogen, water, and carbon dioxide
when hydrocarbon fuels are used.
The following reactions occur when natural gas is used as
the reducing fuel:5
5-8
-------
(0
3
SitS
$
WJ
§1 g—.
C
0)
•P
Ul
>1
M
c •
O M
•H 4J
•P C
O (0
3 «J
•H
4J H
O 0
0) M
rH 4J
0) C
U) O
C O
O
55 o*
n ^i
I O
in IM
0)
5-9
-------
CH4 + 202 •* C02 + H20 + heat (oxygen consumption) Eg. 1
CH4 + 4N02 -» 4NO + C02 + 2H2O + heat (decolorizing) Eq. 2
CH4 + 4NO -* 2N2 + C02 + 2H2O + heat (NOX reduction) Eq. 3
The second reaction is known as the decolorizing step. Though
total NOX emissions are not decreased, the tail gas is
decolorized by converting reddish-brown NO2 to colorless NO. Not
until the final reaction does NOX reduction actually occur.
Heat from the catalytic reduction reactions is recovered as
power in a turboexpander. Depending on the type of NSCR unit,
single-stage or two-stage, heat exchangers or quenchers may be
required to reduce the outlet gas temperature of the NSCR unit
because of thermal limitations of the turboexpander. Temperature
rise associated with the use of NSCR is discussed in greater
detail in the following paragraphs.
Catalyst metals predominantly used in NSCR are platinum or
mixtures of platinum and rhodium. Palladium exhibits better
reactivity and is cheaper than platinum. However, palladium
tends to crack hydrocarbon fuels to elemental carbon under upset
conditions that produce excessively fuel-rich mixtures (greater
than 140 percent of stoichiometry). Consequently, excess oxygen
reacts with deposited carbon and produces a surface temperature
sufficiently high to melt the ceramic support. Platinum
catalysts have been known to operate over extended periods of
time at 150 to 200 percent of stoichiometry (fuel: O2) on natural
gas without exhibiting coking.12 Catalyst supports are typically
made of alumina pellets or a ceramic honeycomb substrate,
although the honeycomb is preferred due to its higher gas space
velocities. Gas space velocity is the measure of the volume of
feed gas per unit of time per unit volume of catalyst. The gas
space velocity (volumetric flue gas flow rate divided by the
catalyst volume) is an indicator of gas residence time in the
catalyst unit. The lower the gas space velocity, the higher the
residence time, and the higher the potential for increased NOX
reduction. Typical gas space velocities are 100,000 and
30,000 volumes per hour per volume for honeycomb and pellet-type
substrates, respectively.5/12
5-10
-------
The reactions occurring within the reduction unit are highly
exothermic. Exit temperature typically rises about 130°C (266°F)
for each percent of oxygen consumed when hydrocarbon fuels are
used. Alternatively, if hydrogen fuel is used, the corresponding
temperature rise is 150°C (302°F) for each percent of oxygen
consumed. Due to catalyst thermal limitations, the final
reduction reaction must be limited to a temperature of 843°C
(1550°F). This corresponds to a maximum tail gas oxygen content
of about 2.8 percent to prevent catalyst deactivation.5
Therefore, the gas must be cooled if oxygen content exceeds
2.8 percent.
Energy recovery imposes greater temperature constraints due
to construction material thermal limitations (650°c [1200°F]) of
the turboexpander. To compensate for these temperature
limitations, two methods of nonselective catalytic reduction have
been developed, single-stage and two-stage reduction.
Single-stage units can only be used when the oxygen content
of the absorber tail gas is less than 2.8 percent. The effluent
gas from these units must be cooled by a heat exchanger or
quenched to meet the temperature limitation of the turboexpander.
Because of the specific temperature rise associated with the
oxygen consumption and NO^ removal, two-stage units with an
internal quench section are used when the oxygen content is over
3 percent.2 Two systems of two-stage reduction are used. One
system uses two reactor stages with interstage heat removal. The
other two-stage reduction system involves preheating 70 percent
of the feed to 482°C (900°F), adding fuel, and passing the
mixture over the first-stage catalyst. The fuel addition to the
first-stage is adjusted to obtain the desired outlet temperature.
The remaining 30 percent of the tail gas feed, preheated to only
121°C (250°F), is used to quench the first-stage effluent. The
two streams plus the fuel for complete reduction are mixed and
passed over the second-stage catalyst. The effluent gas then
passes directly to the turboexpander for power recovery. This
system eliminates the need for coolers and waste-heat boilers;
5-11
-------
however, performance of the two-stage system has been less
satisfactory than that of the single-stage system.5'8
5.1.2.2 Factors Affecting Performance. Factors that can
affect the performance of an NSCR unit include oxygen content of
the absorber tail gas; fuel type, concentration, and flow
distribution; type of catalyst support; and inlet NOX
concentration. The oxygen content of the tail gas entering the
catalytic unit must be known and controlled. As mentioned in the
previous section, excess oxygen content can have a detrimental
effect on the catalyst support and turboexpanders. Even minor
oxygen surplus can lead to catalyst deactivation.
The type of fuel selected is based largely upon
availability. However, it is important to select a fuel that is
compatible with the thermal constraints of the catalytic
reduction system. The temperature rise resulting from oxygen
consumption is higher for hydrogen than for hydrocarbon fuels.2
Fuel concentration is also important in achieving maximum NOV
a
reduction. Natural gas must be added at 10 to 20 percent over
stoichiometry to ensure completion of all three reduction
reactions. Less surplus fuel is required when hydrogen is used.5
Poor control of the fuel/oxygen ratio can result in carbon
deposition on the catalyst, thereby reducing its effectiveness.
Excessive fuel consumption can be minimized by close control of
fuel/tail gas mixing and adequate flow gas distribution into the
catalyst bed (to prevent rich or lean gas pockets).12
Although similar catalyst metals are typically used,
differences in catalyst support can have an effect on the system
performance. Honeycomb supports offer relatively low pressure
drop and high space velocity. The increased surface area of the
honeycomb structure allows greater exposure of the tail gas to
the catalytic material, thereby resulting in improved NOX
conversion. However, honeycombs are more easily damaged by
overheating. Alternatively, pellet beds have proved to be more
durable but offer less gas space velocity. Furthermore, catalyst
fines from pellet beds have been reported to cause turboexpander
blade erosion.12'13
5-12
-------
Malfunctions upstream of the catalytic reduction unit will
also affect the level of NOX reduction. Upsets in the absorption
column that result in NOX concentrations in the 9,000 to
10,000 ppm range can inhibit catalytic activity by chemisorption
(weak chemical bonds formed between the gas and the catalyst
surface). The effects of chemisorption of NO2 are not permanent,
however, and the bed recovers immediately after the upstream
abnormality is corrected.12
5.1.2.3 Performance of Nonselective Catalytic Reduction.
Table 5-2 illustrates the level of control that has been
demonstrated by five nitric acid plants using NSCR as the
exclusive means of NOX control. Production capacities range from
50 to 819 metric tons (55 to 900 tons) per day (expressed as
100 percent nitric acid). Both pellet bed and honeycomb catalyst
supports are equally used, although single-stage units are the
predominant NSCR method. Two common fuel types are used:
natural gas (methane) and ammonia plant purge gas (65 percent
hydrogen).
The emissions data for plants A and £ are taken from test
reports and represent the average of multiple test runs (EPA
Method 7) at each plant. Emissions data for plants B, C, and D
are taken from summaries of test reports and represent the
average of three test runs (EPA Method 7). Emission factors
range from 0.2 to 1.0 kg of NOV per metric ton (0.4 to
a
2.3 Ib/ton) of nitric acid (expressed as 100 percent acid). On
limited data, no trends are apparent relating the catalytic unit
(i.e., the number of stages, fuel type, and catalyst support) to
emission factors. However, it should be noted that the plant
operating at 127 percent of its design production capacity has
the highest NOX emission factor. Regarding fuel type, AP-42
cites NOX emission factors of 1.5 pounds per ton for purge gas
and 0.6 pounds per ton for natural gas. A possible correlation
can be made between control efficiency and the rate of acid
production. As discussed in Chapter 4, production rates in
excess of design can adversely affect absorber efficiency.
Consequently, the NOX concentration of the gas at the inlet of
5-13
-------
TABLE 5-2. NITROGEN OXIDES EMISSIONS FROM NITRIC ACID PLANTS
USING NONSELECTIVE CATALYTIC REDUCTION
Plant
A14
B*
C«
D15
E16
Design
capacity,
tons/d
195
350
55
55
900
Actual
production, %
design
89
107
127
100
NA
No. of
stages
1
2
1
1
NA
Fuel
Natural gas
Natural gas
Purge gas
Purge gas
Natural gas
Catalyst
support*
NA
H
P
H
P
Emission
factor.
lb/ton°
1.13
0.4
2.3
0.7
0.4
Control
efficiency, %c
97.4
99.1
94.7
98.4
99.1
NA = not available.
aH = honeycomb; P = pellet.
''From test reports (EPA method 7).
cThese figures calculated using average uncontrolled emissions level of 43 Ib/ton (from AP-42).
Notes: The following is provided for comparative purposes.
1. From AP-42, NOX emission levels for nitric acid plants using NSCR
a. Natural gas-0.2 kg/metric ton; 0.4 Ib/ton
b. Hydrogen-0.4 kg/metric ton; 0.8 Ib/ton
c. Natural gas/hydrogen (25%/75%)--0.5 kg/metric ton; 1.0 Ib/ton
2. From AP-42, control efficiency for nitric acid plants using NSCR
a. Natural gas-99.1 %
b. Hydrogen-97-99.8%
c. Natural gas/hydrogen (2596/7596)-98-98.5%
3. From NSPS, allowable NOX emission levels from nitric acid plants
Emissions:
1.5 kg/metric ton; 3.0 Ib/ton
5-14
-------
the NSCR unit may be increased to the point of inhibiting
catalyst activity (discussed in Section 5.1.2.2), resulting in
decreased control efficiency.
The data in Table 5-2 indicate NOX control efficiencies
ranging from 94.7 to 99.1 percent. This demonstrated level of
control is consistent with the control efficiency data presented
in AP-42.
5.1.3 Selective Catalytic Reduction
Selective catalytic reduction uses a catalyst and ammonia in
the presence of oxygen to reduce NOX to elemental nitrogen. The
process is called selective because the ammonia preferentially
reacts with NOX in the absorber tail gas. The following sections
discuss SCR used as a NOX control technique for nitric acid
plants. Section 5.1.3.1 describes an SCR system including its
components and operation. Factors affecting the performance of
SCR units are discussed in Section 5.1.3.2. Section 5.1.3.3
presents emission test data and discusses NOV control
X>
performance.
5.1.3.1 Description of SCR Systems. Figure 5-4 is a flow
diagram for a typical nitric acid plant using SCR. Following the
normal ammonia oxidation process, absorber tail gas is passed
through a heat exchanger to ensure that the temperature of the
gas is within the operating temperature range (discussed below)
of the SCR unit. The gas enters the SCR unit, where it is mixed
with ammonia (NH3) and passed over a catalyst, reducing the NOX
to elemental nitrogen (N2).
The reactions occurring in an SCR unit proceed as
follows:1'13
8NH3 + 6N02 -* 7N2 + 12H2O + heat Eq. 4
4NH3 + 6NO -» 5N2 + 6H20 + heat Eq. 5
4NH3 + 302 -» 2N2 + 6H20 + heat Eq. 6
Reactions 4 and 5 proceed at much faster rates than Reaction 6.
Therefore, NOX is reduced without appreciable oxygen removal.
Proper operation of the process requires close control of the
tail gas temperature. Reduction of NOX to N2 must be carried out
within a narrow temperature range, typically 210° to 410°C (410°
5-15
-------
(0
4J
10
•a
•H
o
o
•H
M
4J
•H
C
*J
10
c
o
o
o
o
1
(fl
c
o
•H
4J
O
•o
0)
o
(0
o
0)
•H
-p
o
0)
rH
0)
en
I
IT)
0)
M
g.
•H
5-16
-------
to 770°F).1^ The optimum operating temperature range varies with
the type of catalyst used. The SCR catalysts are typically
honeycombs or parallel plates, allowing the flue gas to flow
through with minimum resistance and pressure drop while
maximizing surface area. Several catalyst materials are
available. In general, precious metal catalysts (e.g., platinum,
palladium) yield higher conversions of NOX to N2 with low excess
ammonia usage at lower temperatures than the base metal oxides
(e.g., titanium, vanadium) or zeolites.12'19 However,
titania/vanadia catalysts are most commonly used in nitric acid
plants.20
Reducing NOV using SCR results in a reduction in acid yield
Ji
and increased ammonia use.12 Acid yield is slightly reduced
because NOV is destroyed rather than recovered as with extended
Ji
absorption. Although ammonia is an expensive reagent, less fuel
is required than for NSCR because complete O2 consumption is not
required. Furthermore, ammonia is readily available since it is
consumed as feedstock in the nitric acid process.8
Several advantages of SCR make it an attractive alternate
control technique. The SCR process can operate at any pressure.
The lack of pressure sensitivity makes SCR a viable retrofit
control device for existing low-pressure nitric acid plants.5
Selective catalytic reduction is also well suited for new plant
applications. Cost savings are a primary benefit of SCR.
Because the temperature rise through the reactor bed is small (2°
to 12°C [36° to 54°F]), energy recovery equipment is not
required. The need for waste-heat boilers and high-temperature
turboexpanders as used for NSCR is eliminated.5
5.1.3.2 Factors Affecting Performance. Three critical
factors affect the NOX removal efficiency of SCR units:
(1) NH3/NOX mole ratio, (2) gas stream temperature, and (3) gas
residence time.20 The reaction equations in the previous section
show that the stoichiometric ratio of NH3 to NOX is 1:1.
Therefore, stoichiometric quantities of ammonia must be added to
ensure maximum NOX reduction. Ammonia injected over
5-17
-------
stoichiometric conditions permits unreacted ammonia to be
emitted, or to "slip." Figure 5-5 illustrates NOX removal
efficiency and NH3 slip as a function of NH3/NOX mole ratio.
Ammonia slip can be monitored and is easily controlled to levels
below 20 ppm (where odor may become a problem).*9
Catalyst activity varies according to the catalyst
composition and temperature. The active temperature range of
catalysts used in nitric acid plants are typically 210° to 330°C
(410° to 626°F).17 The gas temperature in the SCR reactor
chamber must be within the active temperature range of the
catalyst to obtain efficient operation. At lower temperatures,
ammonium nitrate salts can be formed, causing possible damage to
the downstream turboexpander and piping system. Above 270°C
(518°F), NO can be produced by the reaction between NH3 and O2 as
follows:13
4NH3 + 502 -•• 4NO + 6H2O + heat Eq. 7
Older plants may require preheating of the tail gas prior to the
SCR unit in order to accommodate the catalyst temperature
limitations.20
Gas residence time is primarily a function of the flue gas
flow and the catalyst volume or surface area. Residence time is
expressed as space velocity in m3/hr/m3 or area velocity in
m3/hr/m2. Figure 5-6 illustrates NOX removal efficiency and NH3
slip as a function of area velocity. As the area velocity
increases, the residence time of the gas within the catalytic
unit decreases. Consequently, NOX removal efficiency decreases
and unreacted ammonia begins to slip.
5.1.3.3 Performance of Selective Catalytic Reduction.
Selective catalytic reduction is used in many nitric acid plants
in Europe and Japan. However, only three nitric acid plants
using SCR have been identified in the United States: (1) First
Chemical Corp. in Pascagoula, Mississippi, (2) E.I. DuPont de
Nemours in Orange, Texas, and (3) E.I. DuPont de Nemours in
Victoria, Texas.
Tables 5-3 and 5-4 illustrate the levels of NOX reduction
achieved by European plants using SCR. The data are from
5-18
-------
100
G 80
o
60
uj 40
CC
20
0.6
NOx REMOVAL EFFICIENCY
NH3SUP
0.7
0.8 0.9
NHj/NOx RATIO
20 ~
10
a
OT
f
1.0
Figure 5-5. SCR catalyst performance as a function of NH3/NO}
mole ratio.18
5-19
-------
Ill
I
Ul
100
80
60
40
20
NO* REMOVAL EFFICIENCY
NH3SUP
20 ^
10
10 15 20
AREA VELOCITY, Nm3/m2/H
Figure 5-6. SCR catalyst performance as a function of area
velocity.18
5-20
-------
TABLE 5-3. NITROGEN OXIDES EMISSIONS FROM NITRIC ACID PLANTS
USING RHONE-POULENC SCR TECHNOLOGY17
Location
Greece
Greece
Greece
Finland
Norway
Start
date
1985
1985
1985
1986
1987
NOV reduction, ppm
Inlet
1,300
1,500
1,200
1,500
1,200
Outlet
200
200
200
200
200
Control
efficiency,
%a
84.6
86.7
83.4
86.7
83.4
Emission
factor.
lb/tonb
2.87
2.87
2.87
2.87
2.87
^Calculated based on inlet/outlet data.
"Calculated based on NSPS ratio of 3.0 lb/ton:209 ppm.
Example:
X Ib/ton = outlet, „.
2.87 Ib/ton
5-21
-------
TABLE 5-4. NITROGEN OXIDES EMISSIONS FROM NITRIC ACID PLANTS
USING BASF SCR TECHNOLOGY18
Location
Germany
Germany
Germany
Germany
Germany
Germany
Sweden
Sweden
Sweden
Portugal
Sweden
France
Portugal
Norway
Belgium
Start
date
1975
1975
1976
1977
1977
1979
1979
1980
1982
1982
1982
1982
1982
1983
1985
Capacity,
tons/d
270
270
225
270
270
270
225
225
300
360
390
920
360
450
650
NOX reduction, ppm
Inlet
450-800
450-800
1,300
450-800
450-800
500
2,000-2,500
2,000-2,500
550
500
2,000-3,000
850-950
500
500
200
Outlet
<150
<150
<400
<150
<150
<200
<500
<500
<200
<200
<500
<500
<200
<200
<110
Control
efficiency, %a
67-81
67-81
>69
67-81
67-81
>60
75-80
75-80
>64
>60
75-83
41-47
>60
>60
>45
Emission
factor, lb/tonb
<2.15
<2.15
<5.74
<2.15
<2.15
<2.87
<7.18
<7.18
<2.87
<2.87
<7.18
<7.18
<2.87
<2.87
<1.58
aCalculated based on inlet/outlet data.
bCalculated based on NSPS ratio of 3.0 lb/ton:209 ppm. Example:
x Ib/ton = outlet,
Ib/to
n
.3 Ib/ton
~,e,.u
- 2'15 lb/t°D
5-22
-------
European technical papers discussing NOV reductions using SCR
Ji
from two process licensors.17'18 Descriptions of the test data
and test methods are not reported. Control efficiencies are
calculated using the inlet and outlet test data and range from 44
to 86.7 percent. Emission factors are calculated using a ratio
established for the nitric acid NSPS (3.0 lb/ton:209 ppm) and the
outlet NOX concentration. The emission factors range from less
than 0.72 to 3.3 kg of NOX per metric ton (1.58 to 7.18 Ib/ton)
of nitric acid. The data do not indicate a trend relating SCR
control performance to inlet NOX concentration. It should be
noted, however, that high emission factors (greater than
3.0 Ib/ton) may indicate less stringent standards rather than low
SCR control efficiency.
First Chemical Corporation in Pascagoula, Mississippi, is a
new nitric acid manufacturing facility producing 250 tons per day
of nitric acid. Selective catalytic reduction is used in
conjuction with extended absorption for NOX control. Compliance
testing, using EPA Method 7, was performed in April 1991. A
summary of the compliance testing data is as follows:21
NOX emission factor: 0.29 kg/metric ton (0.57 Ib/ton);
NOV concentration: less than 60 ppm; and
Ji
Stack plume opacity: zero percent.
No information was obtained regarding the uncontrolled (exit the
NOX absorber) NOX level. However, because First Chemical is a
new facility, it is reasonable to assume an uncontrolled NOX
emission factor of at least 10 kg per metric ton (20 Ib/ton).7
Based on this uncontrolled NOV emission factor of 10 kg per
Ji
metric ton (20 Ib/ton), the controlled NOX emission factor
(0.29 kg/metric ton [0.57 Ib/ton]) represents a control
efficiency for SCR of 97.2 percent. Again for comparative
purposes, the NOX emission data from First Chemical (0.57 Ib/ton;
<60 ppm) demonstrate that SCR is capable of reducing NOV
a
emissions to well below NSPS levels (3.0 Ib/ton; 209 ppm).
5.1.4 Control Technique Performance Summary
Table 5-5 summarizes the NOX control data presented in
Tables 5-1 through 5-4. For each control technique, the
5-23
-------
TABLE 5-5. SUMMARY OF NO CONTROL TECHNIQUE PERFORMANCE NITRIC
ACID PLANTS
Control technique
Extended adsorption
NSCR
SCR (European)*
SCR (U.S.)b
Emission factor, kg/metric (Ib/ton)
Range
0.59-1.28 (1.3-2.81)
0.2-1.05 (0.4-2.30)
0.72-3.26(1.58-7.18)
0.29 (0.57)
Average
1.05 (2.3)
0.5 (1.0)
<1.67(<3.67)
Control efficiency, %
Range
93.5-97.0
94.7-99.1
44-86.7
97.2
Average
94.6
97.7
70.8
&SCR data are from European plants where less stringent (compared with U.S. standards) standards are
imposed. The SCR is used to bring NOX emissions down to required levels only.
''Based on compliance test data from a single plant using SCR with extended absorption (First Chemical
Corporation).
5-24
-------
following data are presented: range of achievable control,
average achievable control, range of control efficiency, and
average control efficiency.
5.1.5 Other Control Techniques
Several other control techniques for nitric plants have been
developed and demonstrated. However, poor NOX control
performance or other disadvantages have excluded these controls
from common use. These NOX control techniques are (1) wet
chemical scrubbing, (2) chilled absorption, and (3) molecular
sieve adsorption. Each of these techniques is described briefly
below.
5.1.5.1 Wet Chemical Scrubbing. These processes use
ammonia, urea, or caustic chemicals to "scrub" NOV from the
Ji
absorber tail gas, converting the NOX to nitrates or nitrites by
chemical reaction.
5.1.5.1.1 Ammonia scrubbing. Goodpasture, Inc., developed
an ammonia scrubbing process in 1973 that is suitable to retrofit
existing plants with inlet NOX concentrations of up to
10,000 ppm. Feed streams to this process are ammonia and water.
Ammonium nitrate is produced as a byproduct of this process.
Successful operation of this process requires that ammonium
nitrite formation be kept to a minimum and any ammonium nitrite
that forms must be oxidized to ammonium nitrate.
A flow diagram for this process is shown in Figure 5-7. The
entire process is conducted in a single packed bed absorption
tower with three sections operated in concurrent flow. In the
Goodpasture process, there are three distinct sections of the
absorption tower:
1. A gas absorption and reaction section operating on the
acidic side;
2. A second gas absorption and reaction section operating
on the ammonic side; and
3. A final mist collection and ammonia recovery section.
Tail gas enters the first or acidic section of the tower,
where NOX in the gas stream is converted to nitric acid. Ammonia
is added to the process in the second section in sufficient
5-25
-------
TAIL
GAS IN
PRODUCT
AMMONIUM
NITRATE
SOLUTION
PH
RECORD,
CONTROL
"
STEAM
CONOENSATE
Figure 5-7. Process flow diagram for the Goodpasture process.22
5-26
-------
amounts to maintain the pH at a level of 8.0 to 8.3. In this
section of the tower, ammonia reacts with NOX in the gas stream
to form ammonium nitrate and ammonium nitrite; the proportion of
each depends on the oxidation state of the NOX. Product solution
from the second section is fed to the first, where ammonium
nitrite is oxidized to ammonium nitrate by the acidic conditions,
and ammonium nitrate is formed directly from the reaction of free
ammonia with nitric acid. The resulting solution is split into
two streams. One stream is withdrawn from the process as product
solution, while the other is fed to the second or ammoniacal
section of the tower. Feed streams to the third and final
section of the tower consist of process water or steam condensate
in sufficient quantities to maintain the product ammonium nitrate
solution in the 30 to 50 percent concentration range, and a small
amount of solution from the acidic section to control the pH to
approximately 7.0. In this section of the process, entrained
droplets are removed, and any free ammonia is stripped from the
solution. Product solution withdrawn from the first section of
the process contains 35 to 40 percent ammonium nitrate and
0.05 percent ammonium nitrite. The ammonium nitrite can be
oxidized by heating the solution to 115°C (240°F) or by simply
holding it in a tank for 24 hours without heating.
Ammonia scrubbing systems have operated reliably. An
advantage of this process is that the pressure losses are only
6.8 to 13.0 kPa (1-2 psi), which allows the process to be easily
retrofitted for control of existing low-pressure nitric acid
plants. Special precautions must be taken, however, to prevent
deposition of ammonium nitrate on the power-recovery
turboexpander blades. One potential disadvantage of the process
is that the requirement for 85 percent ammonium nitrate solutions
by modern fertilizer plants can necessitate additional
evaporators to concentrate the 35 to 55 percent ammonium nitrate
solution recovered as a byproduct from the Goodpasture process.5
The Goodpasture process is designed to reduce inlet NOV
X
concentrations as high as 10,000 ppm (65 kg per metric ton
[144 lb/ton]) to within NSPS limits (1.5 kg per metric ton
5-27
-------
[3.0 lb/ton]).5 However, nitric acid plants that use this
process have not been identified. Therefore, no test data are
available.
5.1.5.1.2 Urea scrubbing. The MASAR process serves as a
representative example of urea scrubbing. A flow diagram for the
MASAR process is shown in Figure 5-8. The process control device
consists of a three-stage absorption column with gas and liquid
chillers on the feed gas and recirculated solvents. Liquid
ammonia or some other form of refrigeration is used as the
cooling medium. The chemical reaction mechanisms proposed for
urea scrubbing are as follows:
HN02 + CO(NH2)2 *=?N2 + HNCO + H20 Eq. 8
HNCO + HN02 *?N2 + C02 + H2<5 Eq. 9
HNCO + H20 + H + *=?NH4 + C02 Eq. 10
Under actual process operating conditions, the last reaction
listed above predominates so that the overall reaction is:
HN02 + CO (NH2)2 + HN03 - N2 + C02 + NH4NO3 + H20 Eq. 11
In the MASAR process, absorber tail gas is first cooled in a
gas chiller, where condensation occurs and forms nitric acid.
Normal plant absorber feedwater is chilled in the top section of
the MASAR absorber and is then fed to the bottom section, where
it flows countercurrent to the incoming chilled tail gas in the
packed bed. After additional NOX is scrubbed from the tail gas,
the scrubbing water is recirculated through a chiller to remove
reaction heat; this weak acid stream is used as feed to the
nitric acid plant absorber. In the middle section of the MASAR
absorber, the tail gas is scrubbed with the urea-containing
solution, forming nitric acid and nitrous acid that reacts to
form CO(NH2), N2, and H2O. Recirculation of the scrubbing
solution causes the concentration of nitric acid and ammonium
nitrate to rise. Therefore, a bleed stream is required to keep
the system in balance. Makeup urea/water solution is fed to the
scrubbing system at a rate sufficient to maintain a specified
minimum urea residual content. To maintain temperature control
in the middle section, the recirculated scrubbing solution is
pumped through a chiller to remove the heat of reaction. Prior
5-28
-------
TAIL GAS
.TO.
NITRIC ACIO
PLANT
FEED WATER-
SECTION
i
SPENT MASAR
—SOLUTION—
(BLOW DOWN)
LIQUID
CHILLER
CONCENTRATED MASAR
SOLUTION
I
SECTION
B<
PUMP
TAIL GASES
FROM PLANT
LIQUID
CHILLER'
TAIL GAS
CHILLER
•*•
\
SECTION
FEED WATER
TO NITRIC ACIO PLANT
ABSORBEB COLUMN
PUMP
MASAR
ABSORBER
Figure 5-8. Flow diagram of the MASAR process.23
5-29
-------
to leaving the MASAR unit, the tail gas is again scrubbed with
plant absorber feed water in the top section.5
This process has been reported to reduce NOV emissions from
H
4,000 to 100 ppm (26 to 0.7 kg per metric ton [57 to 1.4 lb/ton])
and can theoretically be designed for zero liquid discharge.8 In
practice, however, liquid blowdown of 16 kg/hr (35 Ib/hr) of urea
nitrate in 180 kg/h (396 Ib/hr) of water is estimated for a plant
with a capacity of 320 Mg of nitric acid/d (350 tons/d).
5.1.5.1.3 Caustic scrubbing. Caustic scrubbing involves
treatment of the absorber tail gas with solutions of sodium
hydroxide, sodium carbonate, or other strong bases to absorb NOX
in the form of nitrate or nitrite salts in a scrubbing tower.
Typical reactions for this process are:
2NaOH + 3N02 *=> 2NaNO3 + NO + H2O Eq. 12
2NaOH + NO + N02 *52NaN02 + H2O Eq. 13
One disadvantage of this process is that disposal of the spent
scrubbing solution can require waste-water treatment. Also, the
cost of the caustic can become prohibitive.1'5'8
5.1.5.2 Chilled Absorption. Chilled absorption provides
additional cooling to the absorption tower. This process is
frequently used in addition to other control techniques such as
extended absorption. The principal advantage of chilled
absorption is improved absorber efficiency due to lower
absorption temperature. However, chilled absorption by itself
typically cannot reduce NOX emissions to the level that any of
the three primary control techniques can achieve. Two types of
chilled absorption are the CDL/VITOK and the Tennessee Valley
Authority (TVA) processes.
In the CDL/VITOK process, tail gas enters the absorber,
where the gases are contacted with a nitric acid solution to both
chemically oxidize and physically absorb NOX. A flow diagram of
this process is shown in Figure 5-9. The reaction of NO to N02
is catalyzed in the main absorber. The upper portion of the
absorber is water-cooled to improve absorption. The nitric acid
solution from the absorber is sent to a bleacher where air
removes entrained gases and further oxidation occurs. The
5-30
-------
PURIFIED
TAjL GAS
COOLING
WATER
RETURN
FEED 6AS/I
LIQUID"
FROM HEAT
EXCH.
ABSORBER
BLEACH AIR
RECOVERED AGIO
COOIIMQ
WATER
NH3
VAPORIZER
•MAKE4JFWAHR
JUMP
Figure 5-9. Schematic diagram of the
CDL/VITOK NOX removal process.24
5-31
-------
bleached nitric acid solution is then either sent to storage or
recirculated to the absorber after makeup water is added. The
process uses a closed-loop system to chill the recirculated acid
solution and tower cooling water by ammonia evaporation.
One variation in this system proposed by CDL/VITOK includes
adding an auxiliary bleacher operating in parallel with the
primary unit. Another variation uses a secondary absorber with
its own bleacher.8
The TVA designed and installed refrigeration for NOX
abatement purposes on a nitric acid plant. This process uses
ammonia from the ammonia oxidation process in a closed loop to
cool the top trays of the absorber. Bleacher effluent gases are
also recycled to the absorption tower. Effectiveness of the TVA
process relies on high absorber inlet pressure. This process
reduces product acid concentration.8'25
5.1.5.3 Molecular Sieve Adsorption. The molecular sieve
process has been successful in controlling NOX emissions from
existing plants. However, no new nitric acid plants have been
built that use this form of NOX control.8 The principal
objections to the process are high capital and energy costs, the
problems of coupling a cyclic system to a continuous acid plant
operation, and bed fouling.
The pressure drop through the sieve bed is rather high and
averages 34 kPa (5 psi). The average concentration of NOX in the
treated tail gas discharged to the atmosphere is 50 ppm.
Figure 5-10 shows a flow diagram of a typical molecular
sieve system. The fundamental principle behind molecular sieve
control is selective adsorption of NOX followed by recycle of the
NOX back to the nitric acid plant adsorption tower. The first
step of the process is to chill the absorber tail gas to between
7° and 10°C (45° and 50°F); the exact temperature required is
governed by the NOX concentration in the tail gas stream. Next,
the chilled gas is passed through a mist eliminator to remove
entrained water droplets and acid mist. Weak acid is collected
in the mist eliminator to remove entrained water and acid mist.
This collected weak acid is either recycled to the absorption
5-32
-------
ABSORBER
(EXISTIM8)
TAIL6AS
CONTAININ6 NOX
CLEAN DRY
HOT GAS
CONTAININQ
OESORBEQM02
NO OXIOATIONJO MO2
AND N0;/H20
ADSORPTION
REGENERATION
HEAT
EXCHANGER
POWER
RECOVERY
(EXISTING)
(EXISTING)
•HOT GAS
Figure 5-10. Molecular sieve system
26
5-33
-------
tower or stored. Partially dried tail gas then passes to the
sieve bed, where several operations proceed simultaneously:
1. Dessicant contained in the bed removes the remaining
moisture from the gas stream;
2. NO in the tail gas is converted catalytically to N02;
and
3. NO2 is selectively adsorbed.
Regeneration is accomplished by thermal-swinging (cycling) the
adsorbent/catalyst bed after it is nearly saturated with N02.
Regeneration gas is obtained by heating a portion of the treated
tail gas in an oil- or gas-fired heater. This gas is then used
to desorb NO2 from the bed for recycle back to the nitric acid
plant absorption tower. Both adsorption and regeneration of the
bed require approximately 4 hours.5
5.2 ADIPIC ACID MANUFACTURING
Adipic acid is produced at four plants in the United States.
This section presents a discussion of two NOX control techniques
used at three of the plants: extended absorption and thermal
reduction. A third technique, fume removal by suction, is
uniquely applied by the fourth plant at which adipic acid is a
byproduct.
Sections 5.2.1 and 5.2.2 present discussions of extended
absorption and thermal reduction, respectively. These sections
describe the control techniques, discuss factors affecting their
performance, and provide emissions data that demonstrate the
level of achievable NOX control. Section 5.2.3 describes the NOX
fume removal and recycle system used at the Allied-Signal plant
in Hopewell, Virginia.
5.2.1 Extended Absorption
Extended absorption is used at one plant to reduce NOX
emissions from adipic acid manufacturing by increasing the
absorption efficiency of the NOX absorber. Increased NOX
absorption efficiency is achieved by increasing the volume of the
absorber, which extends the residence time of the NOx-laden gas
with absorbing water, and by providing sufficient cooling to
remove the heat released by the absorption process.
5-34
-------
Extended absorption is suitable for new and retrofit adipic
acid plant applications because a NOX absorption tower is an
integral part of all adipic acid manufacturing processes.
Extended absorption was installed as a retrofit control on the
adipic acid plant that uses this control technique.27
The following sections discuss extended absorption used as a
control technique for adipic acid plants. Section 5.2.1.1
describes the extended absorption system. Section 5.2.1.2
discusses factors affecting the performance of extended
absorption. Levels of achievable NOX emission reductions and the
performance of extended absorption are presented in
Section 5.2.1.3.
5.2.1.1 Description of Extended Absorption. Figure 5-11 is
a flow diagram for the nitric acid reaction portion of a typical
adipic acid plant using extended absorption for NOX control.
Following the nitric acid oxidation of the KA (ketone-alcohol)
oil, NOX is stripped from the product solution using air and
steam in a bleacher. The NOV is then recovered as a weak nitric
A,
acid solution in an absorption tower containing bubble-cap trays.
Nitrogen oxides enter the lower portion of the absorption
tower and flow countercurrent to descending process water, which
enters near the top of the absorption tower. Two processes occur
within the absorption tower: (1) NO is oxidized to NO2, and
(2) NO2 is absorbed in water, forming nitric acid. Heat created
by these processes reactions is removed by cooling water
circulating in internal coils within the trays. The strength of
the nitric acid recovered from the bottom of the absorption tower
is about 20 percent.28 This weak nitric acid is recycled to the
nitric acid reactor. The tail gas exits the top of the absorber
and is discharged to the atmosphere.
5.2.1.2 Factors Affecting Performance. Several factors
that affect the performance of an extended absorber include high
pressure, low temperature, long residence time, and low
throughput. These factors are discussed in detail in Chapter 4
(Section 4.1.2). One adipic acid manufacturer that uses extended
absorption for NOX control cites two main design criteria for
5-35
-------
UJ
DC
UJ
a
iff
Sp
81
o ^
\
•o
•H
u
(0
o
•O
(0
4J
(0
woe
LUUJ
X
O
DC
UJ
Z
o
3
CO
UJ
fe
5
UJ
o
u
i
14
O
U-l
o
•H
o
(Q
T3
(0
•O
0)
•a
c
0)
3IC ACID
ACTOR
I
in
a)
14
o
2
5-36
-------
effective absorber performance: long residence time and low
temperature.2 8
The primary purpose for increasing the size of an absorber
is to increase the residence time. Increasing the residence time
of NOX in the absorber does the following: (1) allows sufficient
time for N02 to be absorbed (approximately l second), and
(2) allows more time for NO (relatively insoluble) to be oxidized
to readily soluble NO2 (minutes). The residence time can also be
increased by using O2 rather than air as a bleaching agent.
Low temperature is another key factor in increasing NOV
J\.
absorption in the absorber. The lower the temperature, the
faster and more efficient the NOX absorption.28 To maintain
efficient operation, heat of reaction is removed by circulating
cooling water through coils in the absorber trays. At one plant,
enough cooling water is circulated through the absorberber such
the gas temperature rises 1°C (1.8°F).28 By maintaining a low
temperature, the absorption process occurs more readily and the
required residence time is also decreased.
5.2.1.3 Performance of Extended Absorption. Extended
absorption is used to control NOX at one adipic acid plant in the
United States. This plant produces approximately 190,000 tons of
adipic acid per year using the cyclohexane oxidation method.29 A
summary of the results of NOX emissions tests was provided by the
plant.
Nitrogen oxides monitoring was conducted over a 3-day period
in 1988 to determine the level of NOV emissions from the NOV
*» X
absorber, located downstream of a nitric acid reactor
(Figure 5-11). On-line instruments used to monitor NOX were
calibrated using EPA methods. The NOX absorber was operating at
maximum rates with cooling water temperature around 20°C (68°F).
Samples were withdrawn from piping at the exit of the absorber.
The testing showed that NOX varied from 500 to 1,500 ppm off the
column.28 With State permit limits at about. 4,500 ppm, the tests
show that extended absorption is capable of achieving permitted
levels of NOX control. An emission factor for NOX from the
absorber was not available. However, calculations using the
5-37
-------
permit level of 700 tons per year and the plant production
capacity indicate a NOX emission factor of at least 3.7 kg of NOX
per metric ton (7.4 Ib/ton) of adipic acid produced. By equating
the State permit levels (700 tons/yr - 4,500 ppm) and applying
that equivalence ratio to the NOV concentrations determined from
Ji
the monitoring data (500 to 1,500 ppm), a range of annual NOX
emissions and emission factors can be calculated. Using the
method just described, the annual NOX emissions for this adipic
acid plant ranged from 77 to 233 tons per year. By dividing the
annual NOX emissions by the annual adipic acid production for
this plant, the NOX emission factors are found to range from
0.41 to 1.23 kg per metric ton (0.81 to 2.45 Ib/ton) of adipic
acid produced.
5.2.2 Thermal Reduction
Thermal reduction is used to control NOX emissions from
adipic acid manufacturing by reacting the NOX in the absorber
tail gas with excess fuel in a reducing atmosphere.30 This
technique of NOX reduction is used at two adipic acid plants.
One plant reduces NOX in a powerhouse boiler, while the other
uses a thermal reduction furnace. However, both techniques can
be considered similar and are treated as such in this section.30
The following sections discuss thermal reduction used as a
control technique for adipic acid plants. Section 5.2.2.1
describes the thermal reduction process. Factors affecting the
performance of thermal reduction are presented in
Section 5.2.2.2. Levels of controlled NOX emissions and
performance of thermal reduction are presented in
Section 5.2.2.3.
5.2.2.1 Description of Thermal Reduction. Thermal (or
flame) reduction reduces NOX by reaction with excess fuel in a
reducing environment. In a typical thermal reduction unit, the
NOX~laden stream and excess fuel (usually natural gas) mixture
passes through a burner where the mixture is heated above its
ignition temperature. The hot gases then pass through one or
more chambers to provide sufficient residence time to ensure
complete combustion. For economic reasons, heat recovery is an
5-38
-------
integral part of thermal reduction units.31 A heat recovery
steam generator typically is used for heat recovery.
The thermal reduction unit used at one plant consists of two
cylindrical towers, 20 feet high and 8 feet in diameter, through
which the gas flows at a rate of 15,000 lb/hr.32 Figure 5-12 is
a simplified flow diagram of an adipic acid plant using thermal
reduction for NOX control. Thermal reduction reduces NOX in
three steps. First, the absorber tail gas is mixed with excess
fuel and burned at high temperature (1090°C [2000°F]) to form
CO2, N2, and H20 in two reactions as follows:
CH4 + 4N02 -» 4NO + C02 + 2H20 Eq. 14
CH4 + 4NO -» 2N2 + C02 + 2H20 Eq. 15
In the second step, the gases are cooled to approximately 760°C
(1400°F), usually by heat exchange. In the third step of the
process, air is admitted and the excess fuel is burned at the
lower (760°C [1400°F]) temperature. Burning the excess fuel at
this temperature prevents atmospheric nitrogen fixation, called
thermal NOX.33 Two adipic acid plants that use thermal reduction
to control NOX produce steam with the heat generated from their
NOV control systems.32'34 For example, one plant with an annual
a
adipic acid production capacity of 300,000 tons per year produces
approximately 50,000 lb/hr of steam from its thermal reduction
unit.40
In addition to NO and N02, adipic acid manufacturing also
produces large quantities of N20. This N20 can be removed
upstream of the NOX absorber and recovered for medical use. If
not recovered for resale, the N2O generally decomposes in the
thermal reduction unit to nitrogen and oxygen; however, some NOV
Ji
is created as a result of the decomposition. There is no data to
quantify the percentage of NOV reformation in the thermal
Jt
reduction unit, although the net effect of this control technique
is that NOX emissions do not exceed the amount of NOX fed to the
unit.32
5.2.2.2 Factors Affecting Performance. Thermal reduction
is essentially a two-step combustion process burning fuel, air,
and NOX. The NOX reduction process occurs after complete
5-39
-------
UJ
UJ
5
(O
oc
<
UJ
X o
UJ Zj
|
4J
<8
•O
•H
O
(Q
O
•H
T3
R)
C
id
OC
UJ
CD
C
4J
O
O
§
•H
3
c
o
•H
•P
O
3
•O
0)
*v
B
-------
combustion of the air. The effectiveness of this NOX reduction
process relies on two factors: temperature and excess fuel.
Temperature is an important criterion of thermal reduction
unit design. Temperature greatly affects the rate at which the
combustion/NOv reduction reactions occur. The higher the
Ji
temperature, the faster the reactions proceed.31 Faster reaction
time reduces the amount of residence time necessary for complete
combustion, thereby reducing the required size of the unit.
However, since fuel costs are the major operating expense for
thermal reduction units, economics dictates the balance between
operating temperature and unit size. Typical operating
temperature for a thermal reduction unit is 1090°C (2000°F).
Thermal reduction units typically burn natural gas (methane)
in a fuel-rich mode to create a reducing environment. Excess
fuel is required (1) to maintain temperature and (2) to reduce
NOV. Enough fuel must be admitted to the burners to promote the
a
initial combustion process. As the temperature in the combustion
chamber increases, the combustion reaction becomes increasingly
self-sustaining. Then, only enough fuel to ensure complete
combustion of the air is needed. However, to reduce NOX, excess
fuel is required to react with the oxygen component of NOX,
forming C02 and water vapor while reducing NOX to elemental
nitrogen. The amount of excess fuel required depends on the NOV
a
concentration inlet of the thermal reduction unit and the
operating temperature.
5.2.2.3 Performance of Thermal Reduction. Thermal
reduction is used to control NOX emissions at two adipic acid
plants in the United States. Current NOX emissions data are
available for only one plant. However, a study of adipic acid
plants performed in 1976 presents NOX emissions data for both
plants.
Table 5-6 presents the available NOV emissions data for the
a
two adipic acid plants using thermal reduction. Both plants
produce adipic acid using the cyclohexane oxidation process. The
controlled NOX emission rate for Plant B (371 Ib/hr) is the
average of 21 tests performed in 1989. The measured NOV emission
J\*
5-41
-------
TABLE 5-6.
NITROGEN OXIDES EMISSIONS FROM ADIPIC ACID PLANTS
USING THERMAL REDUCTION
Plant
B
C
Annual
production
capacity,
tons/yr
350,000
300,000
Fuel
Natural gas and No. 6
fuel oil
Natural gas
NOX emissions
rate, Ib/hr
371b
112d
Annual NOX
emissions,
ton/yr
1,630°
490
NOX emission
factor, Ib/ton
9.3
3.3
Efficiency,
percent8
69
94
5-42
-------
rates for the 21 tests ranged from 191 to 608 pounds of NOX per
hour. The sources of uncontrolled NOX emissions in Plant B were
tail gas from the NOX absorber (7,000 ppmv) and fume sweeps of
the nitric acid storage tanks (9,000 ppmv).32 Storage tank NOX
fumes are routed to the boilers. Using the average NOX emission
rate and assuming Plant B operates 24 hours per day, the NOX
emission factor is calculated by dividing the annual NOX
emissions by the annual adipic acid production capacity. This
calculation results in a NOX emission factor of 4.7 kg of NOX per
metric ton of adipic acid produced (9.3 Ib/ton). It should be
noted that several off-gas streams from various sources are fed
into the thermal reduction unit for combustion at Plant B.
Therefore, determining the amount of NOV contributed by the NOV
Jt a
absorber and the tank fume sweeps is difficult.
The NOX emissions data for Plant c were taken from a report
on emissions from adipic acid plants (1976). No current NOV
X*
emissions data for Plant C were available from the plant or from
the State. The NOX emission rate (determined from a 1976 stack
test) from the thermal reduction unit is 112 pounds of NOX per
hour.30 The NOX emission factor was determined using the same
assumptions as used for Plant B and was calculated to be 1.7 kg
of NOX per metric ton (3.3 Ib/ton) of adipic acid produced.
The NOX concentration in the flue gas of the thermal
reduction unit was 1,500 ppm.30 Although NOX concentrations as
low as 500 ppm were reported to be achievable with this unit,
ceramic cracking in the unit resulted from operating at the high
temperatures required to produce that level of NOX
concentration.3 °
5.2.3 Other Control Technique
Allied-Signal, Inc., in Hopewell, Virginia, produces about
13,000 tons of adipic acid per year. The adipic acid is produced
as a byproduct of their caprolactam plant.30'35 This plant is
unique because it produces a small quantity of adipic acid
relative to the other three plants and because, unlike other
plants, the NOX absorber is not the main source of NOX emissions.
5-43
-------
Instead, the major sources of NOX emissions are the adipic acid
reactors and nitric acid storage tanks.35
Recent data for NOX emissions were not available. The NOX
emissions from the adipic acid reactors and the storage tanks are
recovered by suction and transferred to the caprolactam side for
use in that process.35 Likewise, the tail gas from the NOV
a
absorber is routed to the caprolactam process.30 Allied contends
that NOX emissions are low, although no emissions test data were
provided.3 5
5.2.4 Control Technique Performance Summary
Table 5-7 summarizes the NOX control data for extended
absorption and thermal reduction used in adipic acid
manufacturing. For each control technique, Table 5-7 presents
the level of achievable NOX control and the NOX control
efficiency (based on an uncontrolled emission factor of
53 Ib/ton).
5.3 REFERENCES
1. Review of New Source Performance Standards for Nitric Acid
Plants. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/8-84-011. April
1984. Ch. 4: pp. 1-9.
2. Keleti, C. (ed.). Production of Commercial-Grade Nitric
Acid. In: Nitric Acid and Fertilizer Nitrates. New York,
Marcel Dekker, Inc. 1985. pp. 79-84.
3. Telecon. Boyd, D., Weatherly, Inc., with Lazzo, D., Midwest
Research Institute. May 22, 1991. Nox controls for nitric
acid plants.
4. Reference 1, p. 4-4.
5. Nitric Acid Plant Inspection Guide. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. EPA-340/1-84-013. August 1984. pp. 25-55.
6. Weatherly, Inc. NOX Abatement Systems for Nitric Acid
Plants: Process Description. Engineering firm's
information on catalytic abatement and extended absorption.
5 pp.
7. Reference 1, p. 4-3.
5-44
-------
TABLE 5-7. SUMMARY OF NOX CONTROL TECHNIQUE
PERFORMANCE FOR ADIPIC ACID PLANTS
Control technique
Extended absorption
Thermal reduction
Emission factor
kg/metric ton
3.7
4.9 (1.7-8.4)
Ib/ton
7.4
9.8 (3.3-16.7)
Control efficiency,
percent*
86
81
*Based on an uncontrolled NOX emission factor of S3 Ib/ton.
''Based on recent reported data and data in the 1976 adipic acid study. Emission factor is the average of
available data. Range is given in parenthesis.
5-45
-------
8. Control Techniques for Nitrogen Oxides Emissions from
Stationary Sources: Revised 2nd Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-83-002. January 1983.
Ch. 3: pp. 47-53. Ch. 6: pp. 11-33.
9. Compilation of Air Pollutant Emission Factors: Volume 1:
Stationary Point and Area Sources. U. S. Environmental
Protection Agency. Research Triangle Park, NC. Publication
No. AP-42. September 1985. pp. 5.9.3-6.
10. Reference 1, p. 5-2.
11. Ohsol, E.O. Nitric Acid. In: Encyclopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1990. pp. 150-155.
12. Exhaust Control, Industrial. In: Kirk-Othmer Encyclopedia
of Chemical Technology. New York, John Wiley & Sons. 1981.
pp. 527-530.
13. Blackwood, T.R., and B.B. Crocker. Source Control—
Chemical. In: Handbook of Air Pollution Technology, S.
Calvert and H. M. Englund (eds.). New York, John Wiley &
Sons. 1984. p. 655.
14. N-Ren Corporation. Performance test for nitric acid plant.
Prepared for U. S. Environmental Protection Agency,
Region 6. St. Paul, MN. October 10, 1978. 84 pp.
15. Nitric Acid Plants: Summaries of Test Data. Nitric acid
NSPS Review Report docket reference No. A-83-34, II-A-3.
August 1971. 10 pp.
16. Memorandum from J. Eddinger, EPA/ISB, to K. Durkee, EPA/ISB.
April 6, 1983. Plant Visit—Columbia Nitrogen Corp.
17. Luck, F. and J. Roiron (Rhone-Poulenc). Selective Catalytic
Reduction of NOX Emitted by Nitric Acid Plants. Catalysis
Today. 4.: 205-218. 1989.
18. Telecon. S. Shoraka, MRI, with D. Durila, Englehard.
July 12, 1991. Discussion of SCR materials used in nitric
acid manufacturing plants.
19. Iskandar, R.S. NOX Removal by Selective Catalytic
Reduction, "SCR." Cormetech, Inc. California Clean Air and
New Technologies Conference. October 15-17, 1990. 12 pp.
20. Dittmar, H. Catalytic Reduction of NO., in Nitric Acid Tail
Gases. BASF AG. Federal Republic of Germany.
Environmental Symposium. Kissimmee, FL. October 24-26,
1984. 24 pp.
5-46
-------
21. Telecon. Anderson, D. P., First Chemical Corporation, with
Lazzo, D., Midwest Research Institute. July 29, 1991. NOX
control and control cost of SCR at nitric acid plants.
22. Reference 8, p. 6-23.
23. Reference 8, p. 6-20.
24. Reference 8, p. 6-15.
25. Telecon. Snipes, C., Tennessee Valley Authority, with
Lazzo, D., Midwest Research Institute. May 22, 1991. TVA
chilled absorption process.
26. Reference 8, p. 6-30.
27. Letter from Beck, W.B., E.I. DuPont de Nemours & Company to
Roberts, L.R., Texas Air Control Board. July 2, 1986. NOX
controls for Sabine adipic acid facility.
28. Letter from Plant A to Lazzo, D.W., Midwest Research
Institute. June 21, 1991. NOX control at adipic acid
plant A.
29. SRI International. Directory of Chemical Producers, United
States of America. Menlo Park, CA. 1991. p. 450.
30. Screening Study to Determine Need for Standards of
Performance for New Adipic Acid Plants: Final Report.
GCA/Technology Division. Bedford, MA.
Publication No. GCA-TR-76-16-G. July 1976.
31. APTI Course 415: Control of Gaseous Emissions. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA 450/2-81-005. December 1981.
Chapter 3.
32. Letter from Plant B to Neuffer, B., EPA/ISB. June 18, 1991.
NOX control at adipic acid Plant B.
33. Reference 13, p. 668.
34. Telecon. Neuffer, B., EPA/ISB with Plant C. April 10,
1991. NOX control at adipic acid Plant C.
35. Telecon. Lazzo, D.W., MRI, with Gaillard, R., Allied-Signal
Inc. April 16, 1991. NOX control at adipic acid plant.
5-47
-------
6.0 CONTROL COSTS
This chapter presents capital and annual costs and cost
effectiveness for the NOX control techniques used in nitric and
adipic acid manufacturing plants. Section 6.1 presents costs for
NOX control techniques used in nitric acid plants. The costs are
presented for the following controls: (1) extended absorption,
(2) NSCR, and (3) SCR. Section 6.2 presents costs for NOX
control techniques used in adipic acid plants. These costs are
for (1) extended absorption and (2) thermal reduction.
Three model plant sizes were used to develop costs for the
nitric acid plant NOX control techniques. These model plant
sizes are 181, 454, and 907 metric tons/d (200, 500, and
1,000 tons/d) of nitric acid production (100 percent basis).
These three sizes cover the range of most nitric acid plants in
the United States. Actual plant sizes were used to develop costs
for the adipic acid plant NOX-control techniques.
The capital cost of a control system includes the purchased
equipment costs, direct installation costs, and indirect
installation costs. Purchased equipment costs are those costs
related to purchasing the control equipment. Direct installation
costs include costs for foundations and supports, erecting and
handling the equipment, electrical work, piping, insulation, and
painting. Indirect installation costs include engineering,
contractor's fees, construction expenses, and a contingency fee.1
Annual costs represent the cost of owning and operating the
control system. The total annual cost consists of direct costs,
indirect costs, and recovery credits. Direct costs vary with the
quantity of exhaust gas processed by the control system and
include raw materials, utilities, waste treatment and disposal,
6-1
-------
maintenance materials, replacement parts, and operating,
supervisory, and maintenance labor. Indirect costs are fixed
regardless of the quantity of exhaust gas processed by the
control system and include overhead, administrative charges,
property taxes, insurance, and capital recovery. Direct and
indirect costs are offset by recovery credits, taken for
materials or energy recovered by the control system, which may be
sold, recycled to the system, or reused elsewhere at the site.1
Cost effectiveness is the cost of controlling NOX emissions
by dividing the annual control cost by the quantity of NOX
removed from the exhaust gas stream. Units of cost effectiveness
are given in dollars per ton of NOX removed ($/ton). Annual NOX
emission reduction levels were developed assuming an uncontrolled
emission level of 10 kg per metric ton (20 Ib/ton), which is
equivalent to a NOX concentration of 1,500 ppm (typical for
modern pressure plants), and a controlled emission level based on
the average control efficiency of each control technique.
6.1 COSTS OF CONTROL TECHNIQUES USED IN NITRIC ACID PLANTS
This section presents costs for NOV control systems used in
a
nitric acid plants. Three control systems are analyzed:
(1) extended absorption, (2) NSCR, and (3) SCR. Capital and
annual costs and cost effectiveness are presented for three model
plant sizes: 181, 454, and 907 metric tons/d (200, 500, and
1,000 tons/d) of nitric acid production (100 percent basis). The
cost estimates for extended absorption and NSCR are taken from
the 1984 NSPS review report. Cost estimates for SCR are based on
cost information obtained from an SCR vendor and a U.S. nitric
acid plant that uses SCR for NOX control.
6.1.1 Extended Absorption
This section presents capital and annual costs associated
with using extended absorption to control NOX emissions from
nitric acid plants. The extended absorption control system
costed in this chapter consists of a secondary absorber and a
6-2
-------
closed-loop, chilled-water system for recovering additional
nitric acid. This system is described in detail in Chapter 5.
6.1.1.1 Capital Costs. Table 6-1 shows the capital costs
for an extended absorption system estimated for each of the three
model plant sizes. The extended absorber is a bubble tray column
with 39 trays, regardless of absorber size. The chilled-water
cooling system for the extended absorber consists of a chiller,
compressor, condenser, chilled water tank, and the necessary
pumps and piping.2 Estimates of the capital cost are based on
published cost data.3'4
The purchased equipment cost of each system component was
estimated along with installation, labor, and materials costs to
obtain the total direct costs. This cost includes all the
necessary auxiliaries, such as foundations, insulation, and
ladders. The indirect costs were calculated by multiplying the
total direct costs by the factor shown for each indirect cost
component. All of these costs and factors were taken from
References 3 and 4 and escalated to January 1991 dollars using
the Chemical Engineering (CE) Plant Cost Index.
6.1.1.2 Annual Costs. Table 6-2 shows the annual costs for
an extended absorption system estimated for each of the three
model plant sizes. The annual costs include the direct operating
costs for the pumps, water chiller, and the extended absorber.
Utilities and direct operating labor costs are based on the
following estimates:2
Annual cost element
Water, 106 gallons
Electricity, 106 kW-hr
Labor , man-hr
Plant size, metric tons/d (tons/d)
181 (200)
26
1.2
2,130
454 (500)
72
3.02
3,200
907 (1,000)
130
6.5
4,330
Indirect operating costs are based on percentage factors applied
to direct operating costs and capital costs.
6-3
-------
TABLE 6-1. CAPITAL COST SUMMARY FOR NITRIC ACID PLANTS USING
EXTENDED ABSORPTION FOR NOX CONTROL
(Costs, $1,000)
(January 1991 dollars)
Description
Plant size, metric tons/d (tons/d)
181 (200)
A. Direct costs
1. Absorber tower*
2. Pumps and drives"
3. Chilled water system0
4. Piping, valves, and fittings'*
5. Electrical6
6. Instrumentation'
Total direct costs (TDC)
377
88
23
86
50
50
674
454 (500)
907 (1,000)
637
114
46
211
84
84
1,176
933
218
80
333
124
124
1,812
B. Indirect costs
1. Contractor's fee (696 of TDC)S
2. Engineering (10% of TDC)«
3. Construction expense (8% of TDC)8
Total indirect costs (TIC)
C. Contingency (10% of TOC and TIQS
Total indirect costs (TIC)
Total capital cost (TDC + TIC +
contingency)
40
67
54
161
84
161
919
71
118
94
283
146
283
1,600
109
181
145
435
225
435
2,470
"Reference 3, pp. 768, 769, 770, 772.
Reference 3, pp. 555, 557, 558.
Reference 4, pp. 265, 278.
Reference 3, pp. 529, 530.
Reference 3, p. 171.
^Reference 3, p. 170.
^Reference 3, p. 164.
6-4
-------
TABLE 6-2. ANNUAL COST SUMMARY FOR NITRIC ACID PLANTS USING
EXTENDED ABSORPTION FOR NOX CONTROL
(Costs, $1,000)
(January 1991 dollars)
Description
Plant size, metric tons/d (tons/d)
181 (200)
454 (500)
907 (1,000)
A. Direct operating costs
1. Utilities
a. Water ($0.74/1, 000 gal)
b. Electricity ($0.06/kWh)
19
72
53
181
96
390
2. Operating labor
a. Direct ($22/man-hr)
b. Supervision (20% of direct labor)
47
9
70
14
95
19
3. Maintenance and supplies
(4% x capital cost)
a. Labor and material
b. Supplies
37
64
99
B. Indirect operating costs
1. Overhead
a. Plant (50% x A2 and A3 above)
b. Payroll (20% x A2 above)
47
11
74
17
107
23
2. Fixed costs
a. Capital recovery
(13.5% x capital cost)
b. Insurance, taxes, and G&A (4% x capital
cost)
C. Subtotal
D. Credit for recovered acid
E. Net annualized cost (C-D)
124
37
403
201
202
217
64
754
504
250
334
99
1,260
1,010
257
6-5
-------
The recovery credit for recovered nitric acid is highly
sensitive to the quantity and quality of the recovered acid.
Furthermore, although nitric acid prices are quoted in the
Chemical Marketing Reporter, these prices are not directly
applicable because many nitric acid plants are captive facilities
(acid is produced for in-house use, rather than for market use).
The value of the recovered acid was calculated based on the
following assumptions:
1. Acid production increases by 1.6 percent; and
2. The increased production is a weak acid having a value
of $175 per ton.5
6.1.1.3 Cost Effectiveness. Table 6-3 shows the cost
effectiveness for the three model plants using extended
absorption for NOX control. Cost effectiveness ranges from
$83/metric ton ($76/ton) for a 907-metric tons/d (1,000-tons/d)
plant to $327/metric ton ($297/ton) for a 181-metric tons/d
(200-tons/d) plant. The data show that cost effectiveness
improves (i.e., $/ton of NOX removed decreases) as plant size
increases. This improved cost effectiveness is attributed to the
nitric acid recovery credit. As Table 6-2 shows, as plant size
increases, the acid recovery credit increases at a higher rate
than the direct and indirect operating costs for each plant,
resulting in increasingly lower net annual costs. It should be
noted, however, that the amount of acid recovery credit is
sensitive to the recovery efficiency at each plant and to the
value of the recovered acid.2 In general, the cost of using
extended absorption for NOX control decreases (on a $/ton basis)
as plant size increases.
6.1.2 Nonselective Catalytic Reduction
This section presents capital and annual costs associated
with using NSCR to control NOX emissions from nitric acid plants.
Although nonselective reduction of tail gas pollutants is
generally considered a part of the process (because of the
recovery of heat), it is generally recognized that some portion
of the system constitutes air pollution control. A detailed
description of an NSCR unit and its operation are provided in
6-6
-------
TABLE 6-3. COST EFFECTIVENESS FOR MODEL PLANTS USING EXTENDED
ABSORPTION FOR NOX CONTROL
(January 1991 dollars)
Plant size,
metric tons/d
(tons/d)
181 (200)
454 (500)
907 (1,000)
Annual cost,
$l,000/yr
202
250
257
NOX removed,
metric tons/yr
(tons/yr)
617 (679)
1,550 (1,700)
3,090 (3,400)
COSt
effectiveness ,
$ /metric ton
NOV ($/ton NOV)
327 (297)
161 (147)
83 (76)
6-7
-------
Chapter 5. For costing purposes, it is assumed that the
catalytic treatment unit, the catalyst, the short run of pipe on
either side of the unit for the gases, and the fuel lines
comprise the air pollution control system.
6.1.2.1 Capital Costs. Because of the proprietary nature
of the cost information, no current detailed capital cost data
for an NSCR unit could be obtained. Therefore, capital costs are
based on cost data in the 1984 NSPS review report. In that
report, the capital costs of an NSCR unit are based on a turnkey
price of $2.3 million (January 1983) which includes the cost of
the catalytic unit and the catalyst. The capital costs are
determined for the model plants by applying the Six-Tenths Power
Rule to this cost. Escalating to January 1991 dollars using the
CE Plant Cost Index, capital costs for an NSCR system are as
follows:2
Plant size, metric tons
(tons/d)
181 (200)
454 (500)
907 (1,000)
Capital cost, $106/d
(January 1991 dollars)
1.07
1.86
2.82
6.1.2.2 Annual Costs. Table 6-4 shows the annual costs for
an NSCR system estimated for the three model plant sizes. The
direct costs consist of the fuel (natural gas assumed) used in
the catalytic reduction unit, operating and maintenance labor,
and supplies.
Effective fuel use is reduced by postoxidation heat
recovery. A unit that treats 30.1 m3/s (64,000 standard cubic
feet per minute [scfm]) of tail gas consumes about 1,240 m3
(45,000 ft3) of natural gas per hour. The heat content of this
quantity of natural gas is about 45.6 gigajoules (GJ) (43 million
Btu), of which 23.5 GJ (22.2 million Btu), or 52 percent, is
recovered downstream. Consequently, the net energy requirement
is about 5.74 megajoules (MJ) per 28.3 m3 (5.42 thousand
6-8
-------
TABLE 6-4. ANNUAL COST SUMMARY FOR NITRIC ACID PLANTS USING
NONSELECTIVE CATALYTIC REDUCTION FOR NOX CONTROL
(Costs, $1,000)
(January 1991 dollars)
Description
Plant size, metric tons/d (tons/d)
181 (200)
454 (500)
907 (1,000)
A. Direct operating costs
1. Utilities
a. Natural gas (net of recovered heat) at
$4.12/MMBtu
216
546
1,080
2. Operating labor
a. Direct ($22/man-hr)
b. Supervision (20% of direct labor)
3. Maintenance and supplies (4% x capital cost)
a. Labor and material
b. Supplies
16
3
43
16
3
74
16
3
113
B. Indirect operating costs
1. Overhead
a. Plant (50% x A2 and A3 above)
b. Payroll (20% x A2 above)
31
4
47
4
66
4
2. Fixed costs
a. Capital recovery (13.5% x capital cost)
b. Insurance, taxes, and G&A (4% x capital cost)
C. Total
145
43
501
251
74
1,010
381
113
1,780
6-9
-------
Btu/1,000 scf) of tail gas.2 Utilities and direct operating
labor costs are based on the following:
Annual cost element
Natural gas, 106 Btu
Labor, man-hr
Plant size, metric tons/d (tons/d)
181 (200)
52,500
733
454 (500)
130,000
733
907 (1,000)
263,000
733
Direct operating labor is estimated at 0.5 man-hr per shift,
regardless of the unit size. As with the extended absorption
system, maintenance and supplies are estimated at 4 percent of
the capital cost (including the average cost of catalyst
replacement). Reportedly, the catalyst must be replaced every 3
to 8 years at a cost of about $517,000 for a plant producing
816 metric tons/d (900 tons/d).2 Therefore, the estimated
average annual cost of catalyst replacement (5-year life) at the
model plants is:2
Plant size, metric tons (tons/d)
181 (200)
454 (500)
907 (1,000)
Capital cost, $106/d
(January 1991 dollars)
20.7
53.2
104.9
Estimates of indirect operating costs are based on
percentage factors applied to direct operating costs and capital
costs.
6.1.2.3 Cost Effectiveness. Table 6-5 shows the cost
effectiveness for the three model plants using NSCR for NOX
control. Cost effectiveness ranges from $639 per metric ton
($581 per ton) of NOX removed in a 907-metric tons/d
(1,000-tons/d) plant to $904/metric ton ($823/ton) of NOX removed
in a 181-metric tons/d (200-tons/d) plant. In comparison with
the cost-effectiveness data for extended absorption (Table 6-3),
NSCR is considerably less cost effective. This effect can be
6-10
-------
TABLE 6-5. COST EFFECTIVENESS FOR MODEL PLANTS USING
NONSELECTIVE CATALYTIC REDUCTION FOR NOX CONTROL
(January 1991 dollars)
Plant size,
metric tons/d
(tons/d)
181 (200)
454 (500)
907 (1,000)
Annual cost,
$l,000/yr
501
1,015
1,778
NOX removed,
metric tons/yr
(tons/yr)
637 (701)
1,600 (1,760)
3,190 (3,510)
Cost
effectiveness ,
$ /metric ton
NOV ($/ton NOV)
786 (715)
634 (580)
557 (507)
6-11
-------
attributed to higher utilities costs for NSCR and the lack of any
recovery credit.
6.1.3 Selective Catalytic Reduction
This section presents the costs associated with using SCR to
control NOX emissions from nitric acid plants. Capital costs are
presented in Section 6.1.3.1; annual costs are presented in
Section 6.1.3.2; and Section 6.1.3.3 presents cost effectiveness.
6.1.3.1 Capital Costs. Table 6-6 shows the capital costs
for an SCR system estimated for each of the three model plant
sizes. The estimated costs were provided by a catalyst
manufacturer (Engelhard Corporation) based on the following
information:6
Plant size,
metric tons (tons/d)
181 (200)
454 (500)
907 (1,000)
Stack flow rate, scfm
15,000
34,000
60,000
Nitric acid concentration: 58 percent nitric acid
Absorber tail gas NOX content: (1,500 ppm) equal amounts of
NO and NO2
Absorber tail gas O2 content: 3 percent O2
Temperature (inlet of SCR): 355°C (671°F)
Pressure (inlet of SCR): 612 kPa (90 psi)
Ammonia slip: 10 ppm
Control efficiency: 86 percent reduction (based on
reduction to 209 ppm)
Catalyst: vanadia-titania over honeycomb substrate
The total capital cost of the SCR system depends on the
design and requirements of the system. Capital cost variability
is attributed to three system components: the grid, blowers, and
instrumentation. Depending on the size of the catalyst vessel,
an injection "grid" may be required to ensure an even
distribution of ammonia across the face of the catalyst. A grid
is usually required for large SCR units. The cost for a grid
6-12
-------
TABLE 6-6. CAPITAL COST SUMMARY FOR NITRIC ACID PLANTS USING
SELECTIVE CATALYTIC REDUCTION FOR NOX CONTROL
(Costs, $l,000)a
(January 1991 dollars)
Description
Plant size, metric tons/d (tons/d)
181 (200)
454 (500)
907 (1,000)
A. Direct Costs
1. Catalyst vessel
2. Catalyst capital cost
3. Gridb
4. Blowers0
5. Instrumentation1'
Total direct costs (TDC)
35
45
0-30
0-30
40-200
120-340
50
100
0-30
0-30
40-200
190-410
65
190
0-30
0-30
40-200
295-515
B. Indirect Costs
1. Contractor's fee (6% TDC)
2. Engineering (10% TDC)
3. Construction (8% TDC)
Total indirect costs (TIC)
C. Contingency (10% TDC and TIC)
Total capital investment (TCI) = (TDC + TIC
+ contingency)
Average TCI
7.2-20.4
12-34
9.6-27.2
28.8-81.6
14.9-42.2
164-464
314
11.4-24.6
19-41
15.2-32.8
45.6-98.4
259-559
409
17.7-30.9
29.5-51.5
23.6-41.2
70.8-123.6
36.6-63.9
402-703
553
aBased on cost estimates provided by SCR vendor.
''Based on size of grid required. In some cases, no grid is required ($0).
°Based on temperature requirements of blower. In some cases, no blower is required.
Dependent on sophistication of instrumentation.
6-13
-------
ranges as high as $30,000, depending on injection system
requirements.6
Blowers may be required if air is used as a carrier for the
ammonia. The blowers are used to overcome the pressure within
the catalyst vessel. The temperature of the carrier air affects
the cost of the blowers. If recirculated flue gas is used, high-
temperature-resistant blowers are required, consequently
increasing the cost. Capital cost for the blowers can range as
high as $30,000, depending on the type of blower used.6 The need
for blowers can be eliminated if pressurized steam is used as the
carrier.7
Instrumentation is used to monitor unconverted NOV and/or
a
ammonia slip in the exhaust stream. The cost of the
instrumentation varies from $40,000 to $200,000 depending on the
degree of sophistication. Degree of sophistication ranges from
simple gas flow meters to equipment capable of data acquisition
and trend analysis.6
Capital costs were also provided by First Chemical
Corporation in Pascagoula, Mississippi. First Chemical is a new
nitric acid plant (producing 250 tons/d) that conducted
compliance testing in April 1991. The SCR system was purchased
and installed as part of a turnkey package; therefore, no SCR
component costs could be determined directly.** However, First
Chemical provided an estimate of the capital costs of the SCR
system. The capital costs (reported in October 1989 dollars)
were escalated to January 1991 dollars using the CE Plant Cost
Index and are as follows:8
Description
Catalytic vessel and catalyst
Pumps, piping, electrical
Instrumentation
Installation
Total capital investment (TCI)
Cost, $1,000
500
8
15
25
548
6-14
-------
First Chemical was contacted to determine the type of SCR
catalyst in use. Although the catalytic material was not known,
the catalyst substrate was reported to be a pellet type.9
6.1.3.2 Annual Costs. Table 6-7 shows the annual costs for
an SCR system estimated for the three model plant sizes based on
cost estimates provided by Engelhard. The cost factors and
estimating procedure are based on guidelines for annual costs of
catalytic incinerators from the OAQPS Control Cost Manual.1
Annual anhydrous ammonia costs ranged from $100,000 to $550,000
depending on the plant size. Using aqueous ammonia will reduce
the per-tank cost, but the annual cost will increase due to the
required increase in ammonia consumption.6 Capital recovery cost
is based on the average total capital investment for each model
plant size. Total annual costs based on estimates from the
catalyst manufacturer range from $188,000 for the 181 kg per
metric ton per day (200 ton/d) plant to $714,000 for the 907 kg
per metric ton per day (1,000 ton/d) plant.
Annual costs for an SCR system were also estimated based on
information obtained from First Chemical Corporation. The annual
costs for SCR used in a 250 ton/d nitric acid plant are as
follows:8
Description
A. Direct operating costs
1. Anhydrous ammonia
2. Maintenance and supplies (4 percent of TCI)
3. Catalyst replacement [CRF (5 yr,
10 percent) x catalyst cost]
B. Indirect operating costs
1. Overhead plant (60 percent of A2)
2. Administrative (2 percent of TCI)
3. Insurance (1 percent of TCI)
4. Property taxes (1 percent of TCI)
5. Capital recovery {CFR (10 yr, 10 percent) x
[TCI - (1.08 x cat. cap. cost)]}
Total annual cost (TAG)
Cost,
$1,000
44.7
22
73.9
13.2
11
5.5
5.5
76.2
252
6-15
-------
TABLE 6-7. ANNUAL COST SUMMARY FOR NITRIC ACID PLANTS
USING SELECTIVE CATALYTIC REDUCTION FOR
NOX CONTROL
(Costs, $l,000)a
(January 1991 Dollars)
Description
Plant size, metric tons/d (tons/d)
181 (200)
454 (500)
907 (1,000)
A. Direct Operating Costs
1. Anhydrous ammonia
2. Maintenance and supplies (4% of TCI)
3. Catalyst replacement (CRF [5 yr, 10%] x
catalyst cost)
100
13
11.9
325
16
26.4
550
22
50.1
B. Indirect Operating Costs
1. Overhead Plant (60% of A2)
2. Administration (2% of TCI)
3. Insurance (1% of TCI)
4. Property taxes (1 % of TCI)
5. Capital recovery {CFR (10 yr, 10%) x [TCI
- (1.08 x cat. cap. cost]}
Total annual cost (TAG)
7.8
6.3
3.1
3.1
43.2
188
9.6
8.2
4.1
4.1
49
442
13.2
11.1
5.5
5.5
56.6
714
aBased on cost estimates provided by SCR vendor.
6-16
-------
Annual ammonia cost was estimated based on a pure ammonia
injection rate of 9 scfm and a unit cost of $400 per ton of
anhydrous ammonia. The unit cost of ammonia is an average of
costs that were obtained from three sources.10'11'12 The
catalyst cost was estimated to be 56 percent of the combined cost
of the catalyst and catalyst vessel. This factor (56 percent)
was based on catalyst costs for a similarly sized plant.6
The estimated total annual cost of the SCR in operation at
First Chemical (250 ton/d) is $252,000. The estimated total
annual cost of the SCR unit for a 200 ton/d plant based on costs
supplied by the catalyst vendor is $188,000. Comparing these two
annual costs (relative to respective plant size), it is evident
that the vendor-estimated costs are in line with actual annual
costs.
6.1.3.3 Cost Effectiveness. Table 6-8 shows the cost
effectiveness for the three model plants and the actual plant.
The cost effectiveness ranges from $255/metric ton ($232/ton) of
NOX removed in a 907 metric tons/d (1,000 tons/d) plant to
$336/metric ton ($305/ton) of NOX removed in a 181 metric tons/d
(200 tons/d) plant. These cost effectiveness estimates are based
on cost information supplied by Engelhard (SCR catalyst vendor)
and indicate the cost (on a $/ton-of-NOx-removed basis) of
reducing NOX emissions from an uncontrolled level of 20 Ib/ton
down to 3.0 Ib/ton.6 This reduction represents an 86 percent NOV
A.
control efficiency.
Cost effectiveness based on information obtained from a
250 tons/d nitric acid plant using SCR (First Chemical) is
estimated to be $318/metric ton ($289/ton) of NOX removed. This
cost effectiveness is based on a 97.2 percent reduction
efficiency.8
6.2 COSTS OF CONTROL TECHNIQUES USED IN ADIPIC ACID PLANTS
This section presents costs for NOV control systems used in
H
adipic acid plants. Two NOX control systems are analyzed:
(1) extended absorption and (2) thermal reduction. Cost
information was requested from all three adipic acid plants that
use these controls. However, detailed cost data were not
6-17
-------
TABLE 6-8. COST EFFECTIVENESS FOR NITRIC ACID PLANTS USING
SELECTIVE CATALYTIC REDUCTION FOR NOX CONTROL
(January 1991 dollars)
Plant size,
metric tons/d
(tons/d)
181 (200)
454 (500)
907 (1,000)
227 (250)
Annual cost,
$l,000/yr
501a
l,015a
l,778a
252°
NOX removed,
metric tons/yr
(tons/yr)
637 (701)b
1,600 (l,760)b
3,190 (3,510)b
794 (873)d
Cost
effectiveness ,
$/metric ton
NOV ($/ton NOV)
786 (715)
634 (580)
557 (507)
318 (289)
^Reference 7.
"Based on 86 percent control efficiency.
^Reference 8.
dBased on 97.2 percent control efficiency.
6-18
-------
provided. Available capital and annual costs and cost
effectiveness for extended adsorption and thermal reduction are
presented in Sections 6.2.1 and 6.2.2, respectively.
6.2.1 Extended Absorption
This section presents the costs associated with using
extended absorption to control NOX emissions from adipic acid
plants. Capital costs are presented in Section 6.2.1.1; annual
costs are presented in Section 6.2.1.2; and Section 6.2.1.3
presents cost effectiveness for extended absorption.
6.2.1.1 Capital Costs. The capital costs for extended
absorption are based on cost data obtained from the single plant
that uses this control technique. Plant A reported a total
capital investment of $2.5 million (1986 dollars) for its single-
tower extended absorption system.13 No details on the components
of the capital costs were provided by Plant A. This type of
system is described in detail in Chapter 5. In this case, the
extended absorber was installed as a retrofit control device.
The capital cost of this extended absorption system is
$2.83 million.13
6.2.1.2 Annual Costs. Table 6-9 presents the estimated
annual costs for an extended absorption system used for NOV
a
control in an adipic acid plant. The procedure used to estimate
the annual costs closely follows the annual cost estimating
procedure used for extended absorption systems in nitric acid
plants. An operating cost of $25,000, reported by Plant A,
includes maintenance and utilities. Operating labor costs,
usually included in the direct operating costs, were reported to
be "minimal" by Plant A.13 Therefore, operating labor cost was
assumed to be zero. The credit for recovered nitric acid was
determined by estimating the quantity of nitric acid recovered
based on flow rates from a larger plant. Following the acid
recovery credit procedure for nitric acid plants:2
1. Nitric acid recovery increases by 1.6 percent; and
2. The nitric acid recovered has a value of $175 per ton.3
The price of nitric acid ($175/ton) is for acid with a 60 percent
concentration. Nitric acid recovered in the adipic acid
6-19
-------
TABLE 6-9. ANNUAL COSTS FOR AN ADIPIC ACID PLANT USING EXTENDED
ABSORPTION FOR NOX CONTROL
(January 1991 dollars)
Plant A (190,000 tons/yr)
Description
A.
B.
Direct operating costs3
Utilities and maintenance
Cost, $1,000
25
Indirect operating costs
1 . Overhead
a. Plant — 50% of maintenance
6.25
2. Fixed costs
a. Capital recovery (13.5 percent x
capital cost)
b. Insurance, taxes, and G&A (4% x
capital cost)
C.
D.
E.
Subtotal
Credit for recovered acid*3
Net annual ized cost
382
113
526
(101)
425
aBased on reported annual cost of $25,000 for maintenance and
utilities.
bBased on the following:
1. Estimated production of 300 tons/d of 20 percent nitric
acid;
2. 1.6 percent increase in nitric acid recovery;
3. Market price of $175/ton of 60 percent nitric acid; and
4. Operating 359 d/yr.
6-20
-------
production process has a concentration of only 20 percent.
Consequently, the price used to calculate the acid recovery
credit is one-third of the quoted material price, or
approximately $58 per ton of nitric acid recovered.
The estimated annual cost for extended absorption, before
the acid recovery credit, is $526,000. Including the credit for
recovered nitric acid ($101,000), the net annual cost for
extended absorption used for NOX control in a 173,000 metric
ton/yr (190,000 ton/yr) adipic acid plant is $425,000.
6.2.1.3 Cost Effectiveness. The cost effectiveness of
extended absorption was calculated by dividing the annual cost by
the quantity of NOX removed. The data are as follows:
Plant size,
metric tons/d
(tons/d)
173,000
(190,000)
Annual cost,
$l,000/yr
425
NOX removed,
metric tons/yr
(tons/yr)
3,940
(4,330)
Cost
effectiveness ,
$/metric ton
NOV ($/ton NOV)
108
(98)
The NOX reduction presented above was calculated based on an
uncontrolled NOV emission factor of 26.5 kg/metric ton
JL
(53 Ib/ton) and a controlled NOX emission factor of 3.7 kg/metric
ton (7.4 Ib/ton). It should be noted that cost effectiveness is
highly sensitive to the quality and quantity of nitric acid
recovered as well as fluctuation in market price.
6.2.2 Thermal Reduction
This section presents the costs associated with using
thermal reduction to control NOX emissions from adipic acid
plants. Sections 6.2.2.1 and 6.2.2.2 present the capital and
annual costs, respectively. Cost effectiveness is presented in
Section 6.2.2.3.
6.2.2.1 Capital Costs. Capital costs are based on reported
cost data from the two adipic acid plants using thermal reduction
for NOX control. Plant B reported the current (1991) total
replacement cost of its thermal reduction system, which consists
6-21
-------
of four boilers.14 Plant C reported the cost of their thermal
reduction system, a furnace, in 1990 dollars.15 The cost for
Plant C was escalated to January 1991 dollars using the CE Plant
Cost Index. The capital costs of the respective thermal
reduction units for Plants B and C are as follows:
Plant
B
C
Production capacity,
tons/yr
350,000
300,000
Capital cost, $106
(January 1991 dollars)
8.00
7.05
6.2.2.2 Annual Costs. Table 6-10 shows the annual costs
for a thermal reduction system estimated for Plants B and C. The
cost factors and estimates are based on guidelines for annual
costs of thermal incinerators from the OAQPS Control Cost
Manual.16 Plant B reported an annual cost of $2.05 million for
natural gas based on a natural gas price of $2.09 per thousand
standard cubic feet of gas.14 Natural gas consumption for
Plant C was estimated using the consumption rate reported by
Plant B and scaling that rate for Plant C based on the production
capacities of each plant. Annual costs for Plant C were
subsequently estimated using the estimated natural gas
consumption rate. Total annual costs for Plants B and C are
estimated to be $3.72 and $3.24 million per year, respectively.
Thermal reduction units generate heat through combustion.
Heat from these units is usually recovered as steam for use
elsewhere at the facility. The thermal reduction unit at Plant C
produces 50,000 Ib/hr of steam.15 It should be noted that the
annual costs estimated in Table 6-10 do not include a credit for
the recovered heat. No data are available to determine the
amount of such a heat recovery credit, although total plant
annual cost would be reduced.
6.2.2.3 Cost Effectiveness. Table 6-11 shows the cost
presents the cost effectiveness for thermal reduction units used
at two adipic acid plants. The cost effectiveness for the two
plants is $509/metric ton ($462/ton) of NOX removed for the
6-22
-------
TABLE 6-10.
ANNUAL COSTS FOR ADIPIC ACID PLANTS USING THERMAL
REDUCTION FOR NO^ CONTROL3
(January 1991 dollars)
Description
Costs, $1,000
PlantB
(350,000 tons/yr)
Plant C
(300,000 tons/yr)
A. Direct operating costs
1. Operating labor
a. Operator ($13.6/man-hr)(0.5 hr/shift)b
b. Supervisor (15% of operator)
7.45
1.12
7.45
1.12
2. Maintenance
a. Labor ($15/man-hr)(0.5 hr/shift)b
b. Material (100% of maintenance labor)
8.21
8.21
8.21
8.21
3. Utilities
Natural gas
2,050
1,760C
B. Indirect operating costs
1. Overhead (60% of Al + Al)
2. Administrative [2% of total capital investment (TCI)]
3. Insurance (1% of TCI)
4. Property taxes (1 % of TCI)
5. Capital recovery [CRF (10 yr, 10%) x TCI]
Total annual cost (rounded)
15
160
80
80
1,300
3,720
15
141
70.5
70.5
1,150
3,240
aCosts calculated using reported cost data in conjunction with OAOPS Control Cost Manual format.
"Based on operating time of 24 hr/d; 365 d/yr.
cNatural gas consumption scaled from Plant B consumption based on plant capacities for Plants B and C.
6-23
-------
TABLE 6-11. COST EFFECTIVENESS FOR ADIPIC ACID PLANTS USING
THERMAL REDUCTION FOR NOX CONTROL
(January 1991 dollars)
Plant size,
metric tons/d
(tons/d)
273,000
(300,000)
318,000
(350,000)
Annual cost,
$l,000/yr
3,240
3,720
NOX removed,
metric tons/yr
(tons/yr)
6,370 (7,010)
7,430 (8,170)
Cost
effectiveness ,
$ /metric ton
NOV ($/ton NOV)
509 (462)
501 (455)
aBased on reduction NOX emission from 53 Ib/ton (uncontrolled
emission factor) to 6.3 Ib/ton (average controlled emission
factor.
6-24
-------
300,000 ton/yr plant and $501/metric ton ($455/ton) of NOX
removed for the 350,000 ton/yr plant. These cost effectiveness
figures are based on an uncontrolled NOX emission factor of
26.5 kg/metric ton (53 Ib/ton) to a controlled NOX emission
factor of 3.2 kg/metric ton (6.3 Ib/ton). Comparing the cost
effectiveness for thermal reduction with that for extended
absorption ($108/metric ton [$98/ton]), it is clear that thermal
reduction reduces NOX emissions at a much higher cost. The
higher cost of NOX removal for thermal reduction can be partly
attributed to the cost of the fuel. However, credit for heat
recovery would improve the cost effectiveness of thermal
reduction.
6.3 REFERENCES
1. OAQPS Cost Control Manual: 4th Edition. U. S.
Environmental Protection Agency. Research Triangle Park,
NC. Publication No. EPA-450/3-90-006. January 1990.
Ch. 2: pp. 5-9.
2. Review of New Source Performance Standards for Nitric Acid
Plants. U. S. Environmental Protection Agency. Research
Triangle Park, NC. Publication No. EPA-450/8-84-011. April
1984. Ch. 6.
3. Peters, M. and K. Timmerhaus. Plant Design and Economics
for Chemical Engineers. 3rd Edition. McGraw-Hill, New
York, NY. 1980.
4. Means, R. Building Construction Cost Data. 1983.
5. Nitric Acid. In: Chemical Marketing Reporter. New York,
Schnell Publishing Company, Inc. 1991.
6. Letter from Adams, G. B., Engelhard Corporation to Lazzo,
D., Midwest Research Institute. August 15, 1991. Costs for
an SCR system for model plants.
7. Telecon. Lazzo, D., Midwest Research Institute, with Adams,
G. B., Engelhard Corporation. August 19, 1991. SCR used in
nitric acid plants.
8. Telecon. Lazzo, D., Midwest Research Institute, with
Anderson, D. P., First Chemical Corporation., July 29, 1991.
Control performance and costs for an SCR system.
6-25
-------
9. Telecon. Lazzo, D., Midwest Research Institute, with Marks,
D., First Chemical corporation. August 20, 1991. SCR used
at First Chemical.
10. Permit Application Processing and Calculations by South
Coast Air Quality Management District for proposed SCR
control of gas turbine at Saint John's Hospital and Health
Center, Santa Monica, CA. May 23, 1989.
11. Letter and attachments from Henegan, D., Norton Company, to
Snyder, R., MRI. March 28, 1991. Response to SCR
questionnaire.
12. Champagne, D. SCR Cost-Effective for Small Gas Turbines.
Cogeneration. January-February 1988. pp. 26-29.
13. Letter from Plant A to Lazzo, D. W., Midwest Research
Institute. June 21, 1991. NOX control at adipic acid
Plant A.
14. Letter from Plant B to Neuffer, B., U. S. Environmental
Protection Agency, Industrial Studies Branch. June 18,
1991. NOV control at adipic acid Plant B.
Ji
15. Telecon. Neuffer, B., U. S. Environmental Protection
Agency, Industrial studies Branch, with Plant C. April 10,
1991. NOX control at adipic acid Plant C.
16. Reference 1, Chapter 3: pp. 51-58.
6-26
-------
7.0 ENVIRONMENTAL AND ENERGY IMPACTS
This chapter presents the environmental and energy impacts
of control techniques (described in Chapter 5.0) used to control
NOX emissions from nitric and adipic acid manufacturing plants.
The impacts of these control techniques on air pollution, solid
waste disposal, water pollution, and energy consumption are
discussed. Section 7.1 discusses impacts for nitric acid
manufacturing plants; Section 7.2 discusses impacts for adipic
acid manufacturing plants; and Section 7.3 presents references
used in this chapter.
7.1 NITRIC ACID MANUFACTURING
The control techniques used to reduce NOX emissions from
nitric acid manufacturing plants include extended absorption,
NSCR, and SCR. Section 7.1.1 presents air pollution impacts;
Section 7.1.2 presents solid waste disposal impacts; and
Section 7.1.3 presents energy consumption impacts for each of
these control techniques. Wastewater impacts are not discussed
because liquid effluent waste is not generated by any of the
control techniques.
7.1.1 Air Pollution
7.1.1.1 NOX Emissions. Estimates of NOX emission
reductions achievable through the application of extended
absorption, NSCR, and SCR for the three model plants were
presented in Chapter 6 and are shown in Table 7-1. For each of
the three model plants, the uncontrolled level and controlled NOX
emissions, emission reduction, and percent reduction are
presented.
For this analysis, the amount of NOX removed represents a
reduction from an uncontrolled level of 10 kg/metric ton (kg/ton)
7-1
-------
TABLE 7-1. NOX EMISSIONS FROM NITRIC ACID MANUFACTURING PLANTS
Plant size,
tons/d
200
500
1,000
250
Emissions
Emission
reduction
% reduction
Emissions
Emission
reduction
% reduction
Emissions
Emission
reduction
% reduction
Emissions
Emission
reduction
% reduction
Uncontrolled
NOX emissions,
tons/yr
718
1,800
3,590
Controlled NOX emissions, tons/yr
Extended
absorption
190
3,400
94.6
NCSR
39
679
94.6
100
1,700
94.6
80
3,510
97.7
SCR
17
701
97.7
40
1,760
97.7
25
873
97.2
Note: SCR information based on data provided by First Chemical Corporation.
7-2
-------
(20 Ib/ton), which is equivalent to a NOX concentration of
1,500 ppm (typical for modern pressure plants), to a controlled
level based on average control efficiencies (shown in Table 5-5)
achievable with each of the three control technologies. Nitrogen
oxide emissions are reduced from the uncontrolled level by
94.6 percent for extended absorption, by 97.7 percent for NSCR,
and by 70.8 percent for SCR. The data on NOX emissions from
plants with SCR units are from European plants where less
stringent standards are imposed. The SCR is used to reduce NOX
emissions to required levels only.
7.1.1.2 Emissions Trade-Offs.
7.1.1.2.1 CO and HC emissions from NSCR. Using NSCR to
control NOX emissions increases HC and CO emissions. Fuel is
added in the NSCR unit to react with NOX and oxygen to form
elemental nitrogen, water, and carbon dioxide. Fuel must be
added in excess of stoichiometry to ensure completion of the NOX
reduction reactions.1 However, as the ratio of fuel to oxygen
increases, HC and CO emissions also increase because of
incomplete combustion caused by the fuel-rich conditions in the
unit. The quantity of these emissions is site-specific and
varies with different plant operating parameters.
7.1.1.2.2 NH3 emissions from SCR. The SCR process reduces
NOX emissions by injecting NH3 into the flue gas to react with
NOX to form elemental nitrogen and water. The NH3/NOX ratio
affects the NOX removal efficiency of this unit. Higher ratios
increase amounts of NOX removed but also increase the probability
of unreacted ammonia's passing through the catalyst unit into the
atmosphere (known as "ammonia slip"). Figure 5-5 illustrates NOX
removal efficiency and NH3 slip as a function of NH3/NOX mole
ratio. Gas residence time in the catalyst unit can also have an
impact on the amount of NH3 slip. As the residence time of the
flue gas within the unit decreases, NOX removal efficiency also
decreases, thereby increasing the amount of unreacted NH3.
Figure 5-6 illustrates NOX removal efficiency and NH3 slip as a
function of area velocity.
7-3
-------
7.1.2 Solid Waste Disposal
Catalytic materials used in reduction units typically have a
3- to 8-year life expectancy for NSCR units and a 5- to 10-year
life expectancy for SCR units.2/3 When the catalyst no longer
functions as designed, the catalyst materials will need to be
disposed of.
The catalyst materials predominantly used in NSCR are
platinum and mixtures of platinum and rhodium. Base metal oxides
(e.g., titania/vanadia), precious metal oxides (e.g., platinum/
rhodium, palladium), and zeolites can be used as catalyst
materials in SCR units. Titania/vandia is the catalyst material
most commonly used at nitric acid manufacturing plants with SCR.4
This material is considered hazardous and therefore must be
treated and disposed of as such. Disposal problems are not
encountered with the other materials because they are not
identified as hazardous wastes.
7.1.3 Energy Consumption
Additional electrical energy is required over the
uncontrolled level for extended absorption and SCR, while
additional fuel energy is required for NSCR. These energy
impacts are described below.
Extended absorption requires additional electrical energy to
operate the pumps used to maintain the absorber inlet gas
pressure at the required level of at least 730 kPa. For both
single- and double-tower extended absorption systems, additional
electrical energy is also required to operate a closed-loop
refrigeration system used to cool water in the "extended" portion
of the tower. The extent of the increase in electricity usage is
specific to each nitric acid manufacturing plant. This increase
in electricity usage is presented in Table 7-2 for each of the
three model plants.2
For SCR systems, additional electrical energy is required to
operate ammonia pumps and ventilation fans. This energy
requirement is believed to be minimal and therefore was not
included in this analysis.
7-4
-------
TABLE 7-2. ANNUAL ELECTRICITY REQUIREMENTS FOR EXTENDED
ABSORPTION AND ANNUAL FUEL REQUIREMENTS FOR NSCR
Model plant size,
metric tons/d
(tons/d)
181 (200)
454 (500)
907 (1,000)
Extended absorption
electricity usage
106 MJ/yr
4.3
10.9
23.4
10 6 kwh/yr
1.2
3.02
6.5
NSCR net fuel
requirements
106
MJ/yr
55.3
140
277
106
Btu/yr
52,500
132,500
262,500
7-5
-------
The NSCR process requires additional fuel energy but at the
same time generates a significant amount of heat, which can be
recovered in a waste heat boiler and a tail gas expander. This
recovered heat can supply the energy for process compression
needs with additional steam available for export.5 The amount of
energy recovered in this process lessens the impact of the
additional fuel requirements by reducing the effective fuel use.
The additional energy requirements and the energy recovery
options are site-specific to each nitric acid manufacturing
plant. As discussed in Section 6.1.2.2, the net fuel
requirements for each of the three model plants are presented in
Table 7-2.2
7.2 ADIPIC ACID MANUFACTURING
The control techniques used to reduce NOX emissions from
adipic acid manufacturing plants include extended absorption and
thermal reduction. Section 7.2.1 presents air pollution impacts
and Section 7.2.2 presents energy consumption impacts for each of
these control techniques. Solid waste disposal and wastewater
impacts are not discussed because these wastes are not generated
by either of the control techniques.
7.2.1 Air Pollution
7.2.1.1 NOX Emissions. Estimates of NOX emission
reductions achievable through applying extended absorption
(Plant A) and thermal reduction (Plants B and C) are presented
in Table 7-3. For each plant, the uncontrolled and controlled
emissions, emission reduction, and percent reduction are
presented.
Uncontrolled NOX emissions are based on an emission factor
of 26.5 kg/metric ton (53 Ib/ton) of adipic acid produced. This
is a typical level for uncontrolled adipic acid manufacturing
plants. Controlled NOX emissions are based on an emission factor
of 3.7 kg/metric ton (7.4 Ib/ton) of adipic acid produced for
plants using extended absorption (Plant A) and on 3.2 kg/metric
ton (6.3 Ib/ton) of adipic acid produced for plants using thermal
reduction (Plants B and C). Nitrogen oxide emissions from adipic
7-6
-------
TABLE 7-3. NOV EMISSIONS FROM ADIPIC ACID MANUFACTURING PLANTS
Plant size, tons/yr
A. 190,000
B. 350,000
C. 300,000
Emissions
Emissions reduction
% reduction
Emissions
Emissions reduction
% reduction
Emissions
Emissions reduction
% reduction
Uncontrolled NOX
emissions, tons/yr
5,040
9,280
7,950
Controlled NOX emissions, tons/yr
Extended absorption
703
4,340
86.1
—
•-
Thermal reduction
—
1,720
7,560
81.5
1,470
6,480
81.5
7-7
-------
acid manufacturing plants are reduced by 86.1 percent using
extended absorption and by 81.5 percent using thermal reduction.
7.2.1.2 CO and HC Emissions From Thermal Reduction.
Depending on combustion conditions in the thermal reduction unit,
this method of controlling NOX emissions may produce CO and HC
emissions. Fuel is added in the thermal reduction unit to react
with oxygen and NOX (in the absorber tail gas) to produce
elemental nitrogen, carbon dioxide and water. Adding fuel in
excess of stoichiometric amounts will ensure complete NOX
reduction reactions. However, this excess fuel in the thermal
reduction unit can result in incomplete combustion and,
consequently, CO and HC emissions. Data are not available to
quantify the amount of increased CO and HC emissions for the
three plants discussed in Section 7.2.1.1.
7.2.2 Energy Consumption
Additional electrical energy is required over the
uncontrolled level for extended absorption, while additional fuel
energy is required for thermal reduction.
Extended absorption requires additional electrical energy to
operate the pumps used to maintain the absorber inlet gas
pressure at a required level. The extent of this increase in
electricity usage is specific to each individual plant. These
requirements are not known for the plant discussed in previous
sections.
The thermal reduction process requires additional fuel
energy over the uncontrolled level but at the same time generates
a significant amount of heat, which can be recovered. Two adipic
acid manufacturing plants (Plants B and C) that utilize thermal
reduction for NOX control produce steam with the heat generated
from the control system.6/7 Plant C (300,000 tons/yr) produces
approximately 50,000 Ib/hr of steam from the thermal reduction
unit.7 Plant B (350,000 tons/yr) consumes approximately
983 MMft3/yr of natural gas.6 Data are not available to quantify
the amount of heat recovered at this plant.
7-8
-------
7.3 REFERENCES FOR CHAPTER 7
1. Nitric Acid Plant Inspection Guide. U. S. Environmental
Protection Agency. Research Triangle Park, N.C.
Publication No. EPA-340/1-84-013. August 1984.
pp. 25-55.
2. Review of New Source Performance Standards for Nitric
Acid Plants. U. S. Environmental Protection Agency.
Research Triangle Park, N.C. Publication
No. EPA-450/8-84-011. April 1984. Ch. 6.
3. Iskandar, R.S. NOX Removal by Selective Catalytic
Reduction, "SCR." Cormetech, Inc. California Clean Air
and New Technologies Conference. October 15-17, 1990.
12 pp.
4. Telecon. Shoraka, S., MRI, with Durila, M., Engelhard.
July 12, 1991. Discussion of SCR materials used in
nitric acid manufacturing plants.
5. Ohsol, E.O. Nitric Acid. In: Encyclopedia of Chemical
Processing and Design, J. J. McKetta and W. A. Cunningham
(eds.). New York, Marcel Dekker, Inc. 1990. pp. 150-155.
6. Letter from Plant B to Neuffer, B., U. S. Environmental
Protection Agency, ISB. June 18, 1991. NOX control at
adipic acid manufacturing Plant B.
7. Telecon. Neuffer, B., EPA/ISB, with Plant C. April 10, 1991,
NOX control at adipic acid manufacturing Plant C.
7-9
-------
TECHNICAL REPORT DATA
f Please read !rtstnje::ons en tlic ret crse before completing)
1 REPORT NO. |2.
EPA-450/3-91-026 [
4 TITLE AND SUBTITLE
Alternative Control Techniques Document — Control of
Nitrogen Oxides Emissions from Nitric and Adipic Acid
Manufacturing Plants
7. AUTHOR(S)
David W. Lazzo
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
401 Harrison Oaks Boulevard, Suite 350
Gary, NC 27513-2412
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Emission Standards Division (MD-13)
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5 REPORT DATE
December 1931
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-4379
13. TYPE OF REPORT AND PERIOD COVEF
14. SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
£PA Work Assignment Manager: William J. Neuffer (919) 541-5435
16. ABSTRACT
This Alternative Control Techniques document describes available control
techniques for reducing NOX emission levels from nitric and adipic acid
manufacturing plants. This document contains information on the formation of
NOx and uncontrolled NOx emissions from nitric and adipic acid plants. The
following NOx control techniques for nitric acid plants are discussed:
extended absorption, nonselective catalytic reduction (NSCR), and selective
catalytic reduction (SCR). The following NOx control techniques for adipic
acid plants are discussed: extended absorption and thermal reduction. For
each control technique, achievable controlled NOx emission levels, capital and
annual costs, cost effectiveness, and environmental and energy impacts are
presented.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Nitric Acid Manufacturing
Adipic Acid Manufacturing
NDx Emissions
Control Techniques for NOx Emissions
Extended Absorption
Nonselective Catalytic Reduction (NSCR)
f elective Catalytic Reduction (SCR)
Qfnis orNOx Emission Control
18. DISTRIBUTION STATEMENT
b.lDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report)
20 SECURITY CLASS (This page)
c. COSATl f-'ieid/Croup
21. NO. OF PAGES
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |