EPA-600/2-77-107m
November 1977
Environmental Protection Technology Series
SOURCE ASSESSMENT:
;YNTHETIC AMMONIA PRODUCTION
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Heatth Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield. Virginia 22161.
-------
MRODA-736
EPA-600/2-77-107m
November 1977
SOURCE ASSESSMENT:
SYNTHETIC AMMONIA PRODUCTION
by
G. D. Rawlings and R. B. Reznik
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
Contract No. 68-02-1874
ROAP No. 21AXM-071
Program Element No. 1AB015
EPA Task Officer: Ronald A. Venezia
Office of Energy, Minerals, and Industry
Industrial Environmental Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
The Industrial Environmental Research Laboratory (IERL) of the
U.S. Environmental Protection Agency (EPA) has the responsibility
for insuring that pollution control technology is available for
stationary sources to meet the requirements of the Clean Air Act,
the Federal Water Pollution Control Act, and solid waste legis-
lation. If control technology is unavailable, inadequate, or
uneconomical, then financial support is provided for the develop-
ment of the needed control techniques for industrial and extrac-
tive process industries. Approaches considered include:
process modifications, feedstock modifications, add-on control
devices, and complete process substitution. The scale of the
control technology programs ranges from bench- to full-scale
demonstration plants.
The Chemical Processes Branch of the Industrial Processes
Division of IERL has the responsibility to develop control tech-
nology for a large number (>500) of operations in the chemical
industries. As in any technical program, the first question to
answer is, "Where are the unsolved problems?" This is a deter-
mination which should not be made on superficial information;
consequently, each of the industries is being evaluated in detail
to determine if there is, in EPA's judgment, sufficient environ-
mental risk associated with the process to invest in the develop-
ment of control technology. This report contains the data
necessary to make that decision for the air emissions from
synthetic ammonia production.
Monsanto Research Corporation has contracted with EPA to investi-
gate the environmental impact of various industries which repre-
sent sources of pollution in accordance with EPA's responsibility
as outlined above. Dr. Robert C. Binning serves as Program
Manager in this overall program entitled "Source Assessment,"
which includes the investigation of sources in each of four cate-
gories: combustion, organic materials, inorganic materials, and
open sources. Dr. Dale A. Denny of the Industrial Processes
Division at Research Triangle Park serves as EPA Project Officer.
In this study of synthetic ammonia production, Dr. Ronald A.
Venezia served as EPA Task Leader.
11
-------
CONTENTS
Preface . . . . ii
Figures iv
Tables V
Abbreviations and Symbols vii
1. Introduction ............ 1
2. Summary 2
3. Source Description .6
General description „ 6
Process description ... ........ 9
4. Emissions 23
Emission characteristics . 23
Potential environmental effects 36
5. Control Technology . 48
Desulfurization ...... 48
Primary reformer ..... 48
Carbon dioxide removal system . ... 49
Condensate stripper .... 49
6. Growth and Nature of the Industry . 52
Present technology 52
Emerging technology . 52
Industry production trends . 57
References 59
Appendixes
A. Synthetic ammonia plants in the U.S. in 1976 64
B. Total mass of emissions . 67
Glossary ........... 72
Conversion Factors and Metric Prefixes 73
iii
-------
FIGURES
Number Page
1 Location of synthetic ammonia plants in the
United States 8
2 General process flow diagram of a typical
ammonia plant 10
3 Synthesis gas formation 13
4 Synthesis gas purification 15
5 Ammonia synthesis loop 20
6 Process condensate steam stripper 34
7 Source severity distributions for desulfuriza-
tion tank 42
8 Source severity distributions for primary
reformer 43
9 Source severity distributions for C02
regenerator 44
10 Source severity distributions for condensate
steam stripper 45
11 Modified process condensate strip system .... 50
12 Ammonia process based on partial oxidation of
heavy hydrocarbons (alternative A) 54
13 Ammonia process based on partial oxidation of
heavy hydrocarbons (alternative B) 54
14 Ammonia production based on Koppers-Totzek
coal gasification 55
15 Ammonia process based on Lurgi coal gasification. 56
16 Production of synthetic anhydrous ammonia .... 58
17 New and expanding ammonia plant locations .... 58
IV
-------
TABLES
Number Paqe
1 Emission Characteristics from Average Ammonia Plant
Producing 480 Metric Tons/Day 3
2 Summary of Mass of Emissions of Criteria Pollutants
from Ammonia Plants 4
3 Hydrocarbon Analysis of Typical Pipeline Grade
Natural Gas , 13
4 Typical Purge Gas Analysis 21
5 Plant Data for Emissions from Primary Reformer .... 26
6 Average Emission Factors for Emissions from Primary
Reformer Based on Plant Data 27
7 Emission Factors for Primary Reformer Based on
Combustion Data for Natural Gas and Distillate
Fuel Oil 28
8 Summary of Primary Reformer Emissions 29
9 Data from Material Balances and Design Calculations
Used to Establish Emission Factors 32
10 Emission Factors for Emissions from Regeneration of
the Carbon Dioxide Scrubbing Solution 32
11 Contaminants in Process Condensate 33
12 Mass Balance Around Condensate Steam Stripper-
Result of 65 Test Measurements 34
13 Emission Factors for Process Condensate Steam
Stripper „ 35
14 1976 Distribution of Ammonia Synthetis Plants by
Capacity 37
15 Stack Heights and Uncontrolled Emission Factors for
Representative Ammonia Plant 37
16 Values of F Used to Calculate Appropriate Source
Severities 38
17 Source Severity Values for Uncontrolled Emissions
From Representative Ammonia Plant 40
18 Range of Source Severities and Percentage of Plants
Having Severities Greater than 0.05 or 1.0 . . . . 41
v
-------
TABLES (continued)
Number Page
19 Contribution of Emissions from Ammonia Plants to
State Burdens 46
20 Affected Population for x/F > 0.05 and 1.0 47
VI
-------
ABBREVIATIONS AND SYMBOLS
AAQS --ambient air quality standard
e --natural logarithm base (=2.72)
F --hazard factor
h --stack height
Q --mass emission rate
S --source severity
to --instantaneous averaging time
t --averaging time
TLV —threshold limit value
u --national average wind speed
x --downwind dispersion distance from a source of
emissions
X --downwind ground level concentration at reference
_ coordinate with emission height of h
X --time-averaged ground level concentration of an
emission species
Xmax —instantaneous maximum ground level concentration
Xm=,v —time-averaged maximum ground level concentration
IUCI.A.
TT —3.14
vii
-------
SECTION 1
INTRODUCTION
In 1976, 90 synthetic ammonia plants located in 30 states pro-
duced 15.2 x 106 metric tons9 of synthetic anhydrous ammonia,
making it the fourth largest chemical manufactured in the U.S.
Texas and Louisiana together accounted for almost half of the
national production. Most of the ammonia was used as a direct
application fertilizer and in the manufacture of other fertilizer
products such as urea, ammonium nitrate, and ammonium phosphates.
A small amount of ammonia was used to produce nonfertilizer
materials.
Ammonia production involves not only the reaction of nitrogen
and hydrogen to form ammonia, but also the formation and purifi-
cation of the hydrogen needed in the synthesis. Catalytic steam
reforming of natural gas to produce hydrogen is the only manu-
facturing process studied in this report because 98% of the
ammonia in the U.S. is produced by this method at 84 of the 90
plants. The remaining six plants purchase hydrogen feedstock
from plants that produce hydrogen and chlorine by electrolysis
of sodium chloride.
This report discusses air emissions released during the manu-
facture of synthetic anhydrous ammonia. Emission points within
the manufacturing process are identified, types and quantities
of emission species are delineated, and characteristics of emis-
sions are discussed. The total mass of each emission species is
calculated. State and national emissions of criteria pollutants
[particulates, nitrogen oxides (NOX), sulfur oxides (SOX)» car-
bon monoxide (CO), and hydrocarbons] from ammonia plants are
compared to total state and national emissions from all station-
ary sources. Time-averaged maximum ground level concentrations
of emissions from a typical ammonia plant are compared to the
corresponding ambient air quality standards. Effects of present
and emerging control technology are also discussed.
1 metric ton = 106 grams; conversion factors and metric system
prefixes are presented at the end of this report.
-------
SECTION 2
SUMMARY
This report summarizes the assessment of air emissions released
during the production of synthetic anhydrous ammonia, Standard
Industrial Classification No. 2873131. It encompasses the re-
forming of the feedstock to produce hydrogen, synthesis gas puri-
fication, and ammonia synthesis. Catalytic steam reforming of
natural gas is the only process covered in this report because
98% of the ammonia in the U.S. is produced by that method.
In 1976, 15.2 x 10G metric tons of synthetic anhydrous ammonia
were produced in the U.S. Approximately 80% of this ammonia was
used as a direct application fertilizer and in the production of
other fertilizer products such as urea, ammonium nitrate, and
ammonium phosphates. The remaining ammonia was used to manu-
facture nonfertilizer materials such as ammonium nitrate for
explosives, urea for animal feeds and resins, nitric acid,
acrylonitrile, and amines.
Synthetic ammonia was produced in 30 states by 90 plants which
have a combined annual production capacity of 16.8 x 106 metric
tons. Ammonia plants are concentrated in areas with abundant
supplies of natural gas, such as along the Texas and Louisiana
coast, in California, and in the Central Plains states. Texas
and Louisiana accounted for 45% of national production in 1974.
An average ammonia plant has a capacity of 180 x 103 metric
tons/yr, has a daily production rate of 480 metric tons/day,
and is located in a county having a population density of
117 persons/km2.
In the United States 98% of the synthetic production of anhydrous
ammonia begins with the catalytic steam reforming of natural gas.
Natural gas is first desulfurized and then sent to the primary
reformer where methane is reformed into carbon monoxide and hy-
drogen. This process gas is sent to the secondary reformer where
it is mixed with enough compressed air to give a hydrogen-to-
nitrogen raole ratio of 3:1. A carbon monoxide shift reactor, a
carbon dioxide removal system, and a methanation reactor are then
used to remove all traces of carbon monoxide, carbon dioxide,
and methane from the synthesis gas. Synthesis gas is compressed
and cooled to -33°C for anhydrous ammonia production.
-------
Condensate generated by the carbon monoxide shift reactor, the
carbon dioxide removal system, and the methanation reactor is
steam-stripped to remove ammonia and methanol impurities. Steam
and stripped impurities are vented directly to the atmosphere.
Purge gas from the synthesis loop containing hydrogen, argon,
methane, and nitrogen impurities is vented to the primary re-
former for use as fuel in the radiant heat section.
Air emissions from an ammonia plant result from regeneration of
the desulfurization tank, from combustion occurring in the
radiant heat section of the primary reformer, from regeneration
of the carbon dioxide scrubbing solution, and from steam strip-
ping of process condensate. Emission species and emission
factors associated with these emission sources are shown in
Table 1. On the average, emissions from the regeneration of
the desulfurization tank are released for 10 hours but only
once every 30 days. Emissions from the three other sources are
continuous while the plant is on stream. The industry does not
employ controls on these emission points because no state or
federal standards are exceeded.
TABLE 1. EMISSION CHARACTERISTICS FROM AVERAGE
AMMONIA PLANT PRODUCING 480 METRIC TONS/DAY
' — • - ' • — "
Emission point
Desulfurization tank
Primary reformer
Burning natural gas
Burning fuel oil
Carbon dioxide regenerator
Condensate stripper
Emission species
Total sulfur
CO
Hydrocarbons
NOX
SOX
CO
Particulates
Hydrocarbon s
NOX
SOx
CO
Particulates
Hydrocarbons
Ammonia
CO
Carbon dioxide
Hydrocarbons
Monoethanolamine
Ammonia
Carbon dioxide
Methanol
Emission
factor, g/kg
0.0096b'C
6.9
3.6
2.7 ± 23%
0.0024
0.068
0.072
0.012
2.7
1.3
0.12
0.45
0.15
1.0
1.0
1,220
0.47
0.05
1.1 ± 4%
3.4 ± 60%
0.6 ± 2%
Source
severity
0.05d
0.30
32.4
4.1
<0.01
<0.01
0.03
0.01
4.1
0.44
<0.01
0.21
0.16
2.2
<0.01
0.25
0.54
0.33
3.2
<0.01
0.12
Affected
population,
persons
Oe
0
130
357
0
0
0
0
357
0
0
0
0
197
0
0
0
0
237
0
0
Intermittent source of emissions; desulfurization tank is regenerated on the average
once every 30 days for a 10-hr period.
Worst case condition assuming all sulfur entering the tank is emitted during
regeneration.
Normalized to a 24-hr emission factor.
Based on all sulfur being emitted as sulfur dioxide; if hydrogen sulfide is released,
severity is 0.21 and affected population is 0.
BZero affected population indicates that )(/F is always less than 1.0.
-------
In order to measure potential environmental effects from ammonia
plant emissions, the source severity, S, has been defined as the
ratio of the time-averaged maximum ground level pollutant con-
centration (x'max) to a hazard factor, F. For criteria pollutants,
F is equal to the primary ambient air quality standard. For non-
criteria emission species, it equals a reduced threshold limit
value (TLV®). Values of S are shown in Table 1. The highest
continuous source severities result from nitrogen oxide emissions
from the primary reformer (4.1), and ammonia emissions from the
regeneration of the carbon dioxide scrubbing solution (2.2) and
from the condensate steam stripper (3.2) .
The affected population is defined as the number of persons- -
living in the area around an ammonia plant_where the ratio of the
time-averaged ground level concentration (x) to the hazard factor
is greater than 1.0. The affected populations for the emission
sources from ammonia plants where this ratio exceeds 1.0 are
shown in Table 1. These values are based on the average county
population density for 90 ammonia plants of 117 persons/km2.
The mass of criteria pollutants emitted from ammonia plants by
spates was computed by multiplying the appropriate emission
factor by the amount of ammonia produced in each state. The
range in the mass of emissions for the 30 states which produce
ar.imonia are shown in Table 2. Total national emissions are also
reported. On a state and national basis, the mass of criteria
pollutants emitted from ammonia plants was compared to total
state and national emissions from all stationary sources. Only
emissions of criteria pollutants can be compared because a compre-
hensive data base is only available for these materials. Percent
contributions of ammonia plant emissions to these burdens are
given in Table 2.
~~ TABLE 2. SUMMARY OF MASS OF EMISSIONS OF CRITERIA
POLLUTANTS FROM AMMONIA PLANTS
Mass of emissions,
Contribution to total
state or national
Emission
species
Particulate
NOX
SOX
CO
Hydrocarbons:
metric
State
6
113
9
40
46
to
to
to
to
to
441
8,810
716
3,144
3,562
tons/yr
National
2,051
40,954
3,325
14,616
16,557
emission
State
<0
<0
<0
<0
<0
.1
.1
.1
.1
.1
to
to
to
to
6.
0.
7.
1.
3
24
0
4
burden
/ %
National
0.
0.
0.
0.
0.
01
44
01
02
10
Includes monoethanolamine and methanol.
-------
The trend in the ammonia industry is toward large capacity,
>1,000 metric tons/day plants. With the construction of new
plants and demand for ammonia, annual production should increase
by 4% to 8% through 1980. Industry emissions should also
increase at this rate since no new emission controls are planned,
-------
SECTION 3
SOURCE DESCRIPTION
GENERAL DESCRIPTION
The Bureau of the Census, U.S. Department of Commerce reported
that 15.2 x 106 metric tons of synthetic anhydrous ammonia were
produced in the U.S. in 1976, with approximately 88% produced
for fertilizer use (1-14). Ammonia is used as a direct appli-
cation fertilizer and also as feedstock in the production of
other nitrogen fertilizers such as urea, ammonium nitrate, and
ammonium phosphates. Remaining ammonia is used to produce such
nonfertilizer materials as ammonium nitrate for explosives, urea
for animal feeds and resins, nitric acid, acrylonitrile, and
amines (15). It is believed that a portion (^10%) of the ammonia
(1) Current Industrial Reports, Inorganic Chemicals 1976. Pub-
lication No. M28A(76)-14, U.S. Department of Commerce,
Washington, D.C., August 1977. 30 pp.
(2) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products June 1977. Publication No. M28B(77)-6,
U.S. Department of Commerce, Washington, D.C., August 1977.
7 pp.
(3) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products May 1977. Publication No. M28B(77)-5,
U.S. Department of Commerce, Washington, D.C., July 1977.
7 pp.
(4) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products April 1977. Publication No, M28B(77)-
4, U.S. Department of Commerce, Washington, D.C., June 1977.
7 pp.
(5) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products March 1977. Publication No.
M28B(77)-3, U.S. Department of Commerce, Washington, D.C.,
May 1977. 6 pp.
(6) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products February 1977. Publication No.
M28B(77)-2, U.S. Department of Commerce, Washington, D.C.,
April 1977. 6 pp.
(continued)
-------
reported as produced for fertilizer use is actually consumed in
compounds such as urea and ammonium nitrate that are destined for
nonfertilzier end use. This occurs because the major use for
both urea and ammonium nitrate is as fertilzier. Consequently,
only '^80% of ammonia produced is believed to be consumed in
fertilizers.
Locations of the 90 synthetic ammonia plants in the U.S. are
shown in Figure 1 (16). A description of each plant is given in
(7) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products January 1977. Publication No.
M28B(76)-12, U.S. Department of Commerce, Washington, B.C.,
February 1977. 6 pp.
(9) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products November 1976. Publication No.
M28B(76)-11, U.S. Department of Commerce, Washington, D.C.,
January 1977. 6 pp.
(10) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products October 1976. Publication No.
M28B(76)-10, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
(11) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products September 1976. Publication No.
M28B(76)-9, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
(12) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products August 1976. Publication No. M28B(76)-
8, U.S. Department of Commerce, Washington, D.C., September
1976. 6 pp.
(13) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products July 1976. Publication No. M28B(76)-7,
U.S. Department of Commerce, Washington, D.C., September
1976. 6 pp.
(14) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products June 1976. Publication No. M28B(76)-6,
U.S. Department of Commerce, Washington, D.C., August 1976.
6 pp.
(15) Carbone, W. E., and 0. F. Fissore. Ammonia. In: Kirk-
Othmer Encyclopedia of Chemical Technology, Second Edition,
Volume 2. John Wiley & Sons, Inc., New York, New York,
1963. pp. 258-312.
(16) Hargett, N. World Fertilizer Capacity - Computer Printout.
Tennessee Valley Authority, Muscle Shoals, Alabama, 1976.
7 pp.
-------
Appendix A. These plants have a combined annual production
capacity of approximately 16.8 x 106 metric tons.
Figure 1. Location of synthetic ammonia
plants in the United States (16).
Anhydrous ammonia is synthesized by reacting hydrogen with
nitrogen at a molar ratio of 3:1, then compressing the gas and
cooling it to -33°C. Nitrogen is obtained from the air, but
hydrogen must be produced from a feedstock by one of the fol-
lowing processes: 1) catalytic steam reforming of natural gas
(methane) or naphtha, 2) partial oxidation of heavier hydro-
carbons (petroleum oil or distillates), 3) maximum cryogenic
recovery from petroleum refinery gases or other cracking opera-
tions, 4) gasification of coal or coke, or 5) brine electrolysis
cells at chlorine plants.
It is estimated that 75% to 80% of world ammonia production
utilizes hydrogen produced from catalytic steam reforming
operations, with approximately 60% to 65% of this production
-------
based on natural gas feedstock (17). In the U.S., 84 of the
90 ammonia plants use the catalytic steam reforming process and
natural gas feedstock for producing hydrogen. The other six
plants, representing less than 2% of the ammonia production,
obtain their feedstock hydrogen from electrolysis cells in
chlorine-caustic soda plants. Only those plants that use cata-
lytic steam reforming of natural gas were studied in detail in
this assessment.
PROCESS DESCRIPTION
Six process steps are required to produce synthetic ammonia by
the catalytic steam reforming method:
1. natural gas desulfurization
2. catalytic steam reforming
3. carbon monoxide shift
4. carbon dioxide removal
5. methanation
6. ammonia synthesis
The first, third, fourth, and fifth steps are designed to remove
impurities such as sulfur, CO, C02, and water from the feedstock,
hydrogen, and synthesis gas streams. In the second step, hydro-
gen is manufactured and nitrogen is introduced into the process.
The sixth step produces anhydrous ammonia from the synthesis
gas. While all ammonia plants use this basic process, process
details such as pressures, temperatures, and quantities of feed-
stock, vary from plant to plant.
The volume of natural gas required depends on its heating value
and varies from 35 to 47 GJ/metric ton of ammonia produced, with
an average value of approximately 37 GJ/metric ton (18) . For
natural gas with a heating value of 39 MJ/m3, approximately
900 m3 to 1,200 m3 are required per metric ton of ammonia produced
(17) Buividas, L. J., J. A. Finneran, and 0. J. Quartulli.
Alternate Ammonia Feedstocks. In: Ammonia Plant Safety,
Vol. 17, Chemical Engineering Progress Technical Manual.
American Institute of Chemical Engineers, New York,
New York, 1975. pp. 4-18.
(18) Strelzoff, S. Partial Oxidation for Syngas and Fuel.
Hydrocarbon Processing, 53 (12):79-87, 1974.
-------
An average of 65% to 75% of the natural gas is used as chemical
feedstock, while 25% to 35% is used as fuel for the radiant
heat section of the primary reformer and for steam production
(17, 19). Approximately 40% of the plants have had to use
No. 2 fuel oil to heat the primary reformer because of
natural gas curtailments in the winter.
A genera], process flow diagram of a typical synthetic ammonia
plant using the catalytic steam reforming process is shown
in Figure: 2. The six process steps are described in detail in
the following sections.
NATURAL GAS
FEEDSTOCK
DESULFURIZATION
EMISSIONS
EMISSIONS DURING
'REGENERATION OF TANK
EMISSIONS
EMISSIONS
C02 SOLUTION
REGENERATION
STEAM EFFLUENT _
STEAM
PURGE GAS VENTED
— TO PRIMARY REFORMER
FOR FUEL
Figure 2. General process flow diagram of a
typical ammonia plant.
(19) Finneran, J. A., L. J. Buividas, and N. Walen. Advanced
Ammonia Technology. Hydrocarbon Processing, 51(4):127-130,
1972.
10
-------
Natural Gas Desulfurization
Sulfur content of natural gas feedstock must be reduced to as low
a level as is economically possible to prevent poisoning the
nickel catalyst in the primary reformer. Sulfur is present as
hydrogen sulfide, 70%; mercaptans, 10%; monosulfides (such as
dimethylsulfide, CH3-S-CH3) 12%; and disulfides (such as dimethyl-
disulfide, CH3-S-S-CH3) 8% (20). Total sulfur concentration in
pipeline grade natural gas ranges from 229 yg/m3 to 915 yg/m3,
with an average value of 450 yg/m3. This concentration must be
reduced to <280 yg/m3.
Over 95% of the ammonia plants use a regenerable activated carbon
fortified with a metallic additive, such as copper, for feedstock
desulfurization. Remaining plants use a zinc oxide bed which is
replaced. Ammonia plants using activated carbon for desulfuriza-
tion employ a dual-tank system so that one tank is always on-
stream while the other is being regenerated. The tank design
factor is 200,000 m3 of natural gas per cubic meter of carbon
(21). Tank sizes depend on the process flow rate and the length
of time the unit remains on stream. Pressures in the tank range
from 3.4 MPa to 4.1 MPa at temperatures of 38°C to 425°C, depend-
ing on the particular plant design (22).
Sulfur compounds such as hydrogen sulfide are removed from the
feedstock by reaction with the metallic oxide in the carbon to
form a metallic sulfide. A typical reaction is:
CuO + H2S »- CuS + H2O (1)
The activated carbon tank is used until the elemental sulfur
buildup reaches 13 to 25 weight percent of the carbon. When the
concentration in the exit gas reaches about 0.2 ppm, the feed-
stock flow is diverted to a second carbon tank, while the first
is regenerated. Regeneration is accomplished by passing super-
heated steam (230°C to 290°C) through the bed at a rate of about
900 kg/hr. The carbon bed is then heated to 230°C for
(20) Personal communication with Dayton Power and Light Company,
Dayton, Ohio, July 1976.
(21) Green, R. V. Synthetic Nitrogen Products. In: Riegel's
Handbook of Industrial Chemistry, Seventh Edition. J. A.
Kent, ed. Van Nostrand Reinhold Company, New York, New
York, 1974. pp. 7'5-122.
(22) Haney, G., and K. Wright. Media for Removing Sulfur from
Natural Gas. In-: Ammonia Plant Safety, Vol. 12, Chemical
Engineering Progress Technical Manual. American Institute
of Chemical Engineers, New York, New York, 1970. pp. 50-54.
11
-------
8 to 10 hours. During this period, additional air or oxygen may
be added with the steam to increase the oxygen concentration to
^5,000 ppm (23, 24). Free oxygen reacts with the metal sulfide
to form a metal oxide and elemental sulfur, for example:
2 CuS + 02 — ^2CuO + 2S (2)
Elemental, sulfur is trapped in the pores of the carbon bed.
Calgon Corporation, a major manufacturer of activated carbon
units, reports that all of the elemental sulfur remains on the
carbon surface, and is not removed by the regeneration steam
(25, 26) . However, because the vapor pressure of sulfur9 at
230°C to 290°C is 0.8 kPa to 6.0 kPa, some sulfur must be lost
by vaporization.
Activated carbon will also adsorb hydrocarbons. CI-GI+ hydrocar-
bons are weakly adsorbed, while C5-Cy hydrocarbons are more
strongly adsorbed. Shorter chain hydrocarbons are gradually
replaced by longer chain compounds during the desulfurization
cycle. Hydrocarbon loading can be as much as 5% to 10% of the
weight of the carbon in the bed (23, 25). During regeneration,
portions of the Cj-C^ and C^-C-j compounds are desorbed and
vented with the steam. A hydrocarbon analysis of typical pipe-
line grade natural gas is shown in Table 3.
Catalytic Steam Reforming
Natural gas leaves the desulfurization tank containing less than
0.2 ppm sulfur. Sweetened natural gas is mixed with process
Vapor pressures of sulfur at 230°C and 290°C were obtained by
interpolation of data given in Reference 27.
(23) Personal communication with P. D. Langston, Calgon Corpora-
tion, Pittsburgh, Pennsylvania, 5 September 1975.
(24) Personal communication with F. R. Bossi, Calgon Corpora-
tion, Pittsburgh, Pennsylvania, 29 September 1975.
(25) Personal communication with W. Lovett, Calgon Corporation,
Pittsburgh, Pennsylvania, 29 September 1975.
(26) Air Purification with Granular Activated Carbon, Brochure
23-55, Calgon Corporation, Pittsburgh, Pennsylvania. 1975.
p. 24.
(27) Liley, P. E., and W. R. Gambill. Physical and Chemical
Data. In: Chemical Engineers' Handbook, Fifth Edition,
Section 3, R. H. Perry and C. H. Chilton, eds. McGraw-
Hill Book Company, New York, New York, 1973. p. 48.
12
-------
TABLE 3. HYDROCARBON ANALYSIS OF TYPICAL
PIPELINE GRADE NATURAL GAS (20)
Compound
Concentration,
mole %
Methane
Ethane
Propane
iso-Butane
n-Butane
iso-Pentane
n-Pentane
Hexane
Nitrogen
Carbon dioxide
95.398
2.666
0.143
0.030
0.029
0.012
0.008
0.015
0.323
1.376
steam and preheated to ^540°C in the second coil in the waste
heat removal section of the primary reformer, Figure 3. The
steam-to-gas ratio normally ranges from 3.5 to 4.0 moles steam
per mole carbon. A lower ratio could be used, but the higher
ratio improves conversion and helps supply the steam needed in
the carbon monoxide conversion step. Moreover, an excess of
steam in the reformer prevents carbon formation on the catalyst,
STEAM DRUM
WATER
STEAM
EMISSIONS
DESULFURIZATION
TANKS
V
STACK
NATURAL GAS-
COMPRESSOR
PRIMARY REFORMER
Y
A
\
TO STEAM
DRUM
SECONDARY
REFORMER
PURGEGAS
TOCO
SHIR
REACTOR
Figure 3. Synthesis gas formation.
13
-------
A mixture of steam and gas enters the vertically supported
primary reformer tubes, which are filled with a nickel-based
reforming catalyst. The endothermic reforming reaction (Equa-
tion 2) requires heat input of 227 kJ/mole:
CH^ + H20 » CO + 3H2 (3)
Radiant heat for the reforming reaction is supplied by firing
natural gas and purge gas (from the synthesis loop) on the out-
side of the reformer tubes. Approximately 70% of the methane
is converted to hydrogen and carbon monoxide in the primary re-
former. Reformed gas contains about 8% methane (dry basis).
Gas exits the primary reformer tubes at ^730°C to %820°C. Flue
gas leaves the radiant section at ^960°C. Excess heat in the
flue gas is removed by waste heat coils and preheaters. Flue
gas leaves the reformer stack at 230°C.
Process gas is sent to the secondary reformer, where it is mixed
with air that has been compressed in a centrifugal compressor to
^3.4 MPa and preheated to about 540°C in heat exchangers in the
primary reformer. Sufficient air is added to .produce a final
synthesis gas having a hydrogen-to-nitrogen mole ratio of 3:1.
Heat necessary for completion of the reforming reactions is
supplied by combustion of the two gases as they mix in the top
of the secondary reformer. Reaction gases pass over a bed of
nickel-based reforming catalyst similar to that used in the
primary reformer. Gas temperature at the exit of the secondary
reformer is 950°C to 1,000°C which is reduced to ^360°C in a
waste heat boiler. Sufficient heat is recovered to produce from
50% to 100% of the 10.3 MPa steam required in the plant.
Carbon Monoxide Shift
After cooling, the secondary reformer effluent gas (12.0% carbon
monoxide and 8.4% carbon dioxide on a dry basis) enters a high
temperature (330°C to 550°C) CO shift converter, shown in Fig-
ure 4, which is filled with a chromium oxide,promoted - iron
oxide shift catalyst. Conversion of carbon monoxide to carbon
dioxide and hydrogen is necessary for economical use of the raw
synthesis gases. The following reaction takes place in the
carbon monoxide shift converter:
CO + H20 —> C02 + H2 (4)
Heat from the exothermic reaction (41 kJ/mole of CO) is used to
produce steam. The temperature of the exit gas from the con-
verter is ^425°C.
14
-------
FROM
SECONDARY REFORMER
HIGH TEMP.
SHIFT
CONVERTER
EMISSIONS
KNOCKOUT ( )
DRUM
LOW TEMP.
SHIFT
CONVERTER
co2
C02 SOLUTION ABSORBER
REGENERATOR f
METHANATOR
C
TO AMMONIA
SYNTHESIS LOOP
PROCESS CONDENSATE
TO STEAM STRIPPER
PROCESS CONDENSATE
TO STEAM STRIPPER
Figure 4. Synthesis gas purification.
Shift gas is cooled to %200°C in a heat exchanger. In some
plants (^85%), the gas is then passed through a bed of zinc
oxide to remove any residual sulfur contaminants that would
poison the low temperature shift catalyst. In other plants,
excess low temperature shift catalyst is added to insure that
the unit will operate through its expected lifetime (28).
The low temperature shift converter is filled with a copper
oxide/zinc oxide catalyst which is highly active between 200°C
and 260°C. Final shift gas (0.3% CO on a dry basis) is cooled
from ^210°C to %110°C, and enters the bottom of the carbon
dioxide absorption system. Unreacted steam is condensed and
separated from the gas in a knockout drum.
A 544-metric ton/day ammonia plant produces 7.89 x 10~3 m3/s of
condensed steam (process condensate) . A 900-metric ton/day plant
produces 1.39 x 10~2 m3/s of condensate. This water contains
(28) Personal communication with J. C. Barber, Tennessee Valley
Authority, Muscle Shoals, Alabama, June 1976.
15
-------
approximately 600 ppm to 1,200 ppm ammonia, 200 ppm to 1,000 ppm
methanol, and 200 ppm to 2,800 ppm carbon dioxide (29-32).
Ammonia in the process condensate is formed in the high tempera-
ture shift converter. It is present as ammonium bicarbonate be-
cause the condensate is saturated with carbon dioxide. Methanol
is formed in the low temperature shift converter. Condensate
also contains small amounts (<1 ppm) of sodium, iron, copper, zinc,
calcium, and aluminum, which enter the process stream through
contact with catalyst, internal refractory, vessel walls, and
piping (29, 33).
Process condensate is sent to a stripper to remove volatile
gases such as ammonia, methanol, and carbon dioxide. The con-
centration of ammonia in the effluent is reduced to 20 ppm, the
methanol to 20 ppm, and the carbon dioxide to 40 ppm with 96 kg
to 240 kg of steam per cubic meter of condensate (29). Ion ex-
change units or molecular sieves are then used to further purify
the condensate if it is to be recycled to the boilers. Steam
and volatile gases are vented to the atmosphere. Trace metals
remaining in the process condensate are removed by the ion
exchange unit.
(29) Quartulli, 0. J. Stop Wastes: Reuse Process Condensate.
Hydrocarbon Processing, 54 (10):94-99, 1975.
(30) Romero, C. J., F. Yocum, J. H. Mayes, and D. A. Brown.
Treatment of Ammonia Plant Process Condensate Effluent.
EPA-600/2-77-200, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, September 1977.
85 pp.
(31) Spangler, H. D. Repurification of Process Condensate. In;
Ammonia Plant Safety, Vol. 17, Chemical Engineering Progress
Technical Manual„ American Institute of Chemical Engineers,
New York, New York, 1975. pp. 85-86.
(32) Quartulli, 0. J. Review of Methods for Handling Ammonia
Plant Process Condensate. Presented at the Fertilizer
Institute Manufacturing Environmental Seminar, New Orleans,
Louisiana, January 14-16, 1976. 20 pp.
(33) Finneran, J. A., and P. H. Whelchel. Recovery and Reuse of
Aqueous Effluent from a Modern Ammonia Plant. In: Ammonia
Plant Safety, Vol. 13, Chemical Engineering Progress Tech-
nical Manual. American Institute of Chemical Engineers,
New York, New York, 1971. pp. 29-32.
16
-------
Carbon Dioxide Removal
The final shift gas contains C02 which must be removed. (About
1.22 metric tons (34) of carbon dioxide are produced per metric
ton of ammonia.) Removal of carbon dioxide depends on its acid-
gas character; it tends to form carbonic acid in water:
C02 + H20 ^—^ H2C03 (5)
Carbonic acid can be absorbed by solutions of amines, for
example:
2NH2CH2CH2OH + H2C03 » (NH2CH2CH2)2C03 + 2H20 (6)
or by solutions of alkaline salts, such as:
K2C03 + H2C03 > 2KHC03 (7)
to form carbonates. These carbonates decompose into C02 and the
amine or salt on heating, regenerating the absorption solution
(Figure 4). Carbon dioxide scrubbing systems used in the U.S.
today employ either monoethanolamine or hot potassium carbonate
as the scrubbing medium (21, 34).
Monoethanolamine Scrubbing1—
The classical method for removing carbon dioxide, used by 80% of
the ammonia plants, is absorption in monoethanolamine (MEA). Gas
containing carbon dioxide is passed upward through an absorption
tower countercurrent to a 15% to 30% solution of MEA in water
fortified with effective corrosion inhibitors (21). The amine
solution, after absorbing the carbon dioxide, is first preheated,
then regenerated in a reactivating tower. Carbon dioxide is
removed from the solution first by steam stripping and then by
heating. Two kilograms of steam per kilogram of carbon dioxide
are used for regeneration (35). Carbon dioxide gas (98.5% C02)
is then vented to the atmosphere or used for chemical feedstock
in other parts of the plant complex. After being cooled in a
heat exchanger and solution cooler, regenerated MEA is pumped
back to the absorber tower. This process permits carbon dioxide
removal at atmospheric pressure.
(34) Strelzoff, S. Choosing the Optimum CO2-Removal System.
Chemical Engineering, 82 (19):115-120, 1975.
(35) Hahn, A. V. The Petrochemical Industry - Market and
Economics. McGraw-Hill Book Company, New York, New York,
1970. pp. 19-38.
17
-------
Hot Potassium Carbonate Scrubbing —
A comparatively new process for carbon dioxide removal uses hot
potassium carbonate (K2C03) as a scrubbing medium and is based
on the reaction:
K2C03 + C02 + H20 - »• 2KHC03 (8)
The scrubbing system is "hot" because the gas enters at 120°C.
Normally, the solution from the regenerator is not cooled before
it returns to the absorber. In this system, the cost of heat
exchangers can be eliminated and the solubility of hydrogen and
nitrogen is decreased.
Because the capacity of hot carbonate solutions (25% to 30%) for
absorbing carbon dioxide is between that of water and MEA solu-
tions, carbon dioxide is removed in two stages operating at
different pressures. Higher pressure is required for cleanup,
whereas removal of the bulk of the carbon dioxide at lower pres-
sure reduces compression costs. Converted synthesis gas at
about 120 °C and 1.5 MPa enters the bottom of the absorber and
passes up through the packing, countercurrent to a 40% carbonate
solution. This stage, reduces the carbon dioxide to about 3%.
In the top section of the absorber, a 40% carbonate solution at
about 90 "C and 5.2 MPa to 5.5 MPa scrubs the residual carbon
dioxide from the gas and reduces the carbon dioxide content to
<1%. As the solution absorbs carbon dioxide, some carbonate
becomes bicarbonate. Spent solution from the absorbers is
reactivated in a two-stage regenerator. The rich carbonate
solution flashes as it enters the regenerator, producing carbon
dioxide and water vapor. This vapor is condensed and approxi-
mately 80% is returned to the regenerator. The carbon dioxide
has a purity of at least 98.5% and is either vented to the
atmosphere or used in urea manufacture.
Flashed carbonate solution is stripped further by reboiler-
generated, steam. The solution leaving the top section of this
regenerator contains <1% C02 and is used as the absorbent in the
lower portion (low-pressure section) of the absorber tower.
A leaner carbonate solution is available at the bottom of the
regenerator section, and it is used as the absorbent in the top
(higher pressure)' section of the absorber, after being cooled
to
Similar potassium carbonate systems have been developed: the
Fluor solvent process which employs propylene carbonate, the
(36) Haase, D. J. New Solvent Cuts Costs of Carbon Monoxide
Recovery. Chemical Engineering, 82 (16) :52-54, 1975.
18
-------
Giammarco-Vetrocoke hot carbonate process which employs solution
activators such as arsenic trioxide, and the Catacarb process
which activates hot carbonate with an organic additive (21, 34,
36) .
Methanation
In commercial practice, all carbon dioxide absorption methods
leave a small amount of carbon dioxide (usually <1.0%) in the
synthesis gas which must be removed because it is a poison to
most ammonia synthesis catalysts. Residual C02 is removed by
catalytic methanation (Figure 4) which is conducted over a
nickel catalyst (nickel oxide on alumina) at temperatures of
400°C to 600°C and pressures up to 3 MPa according to the
following reactions:
CO + 3H2 - »> Ch4 + H20 (9)
C02 + H2 - >• CO + H20 (10)
C02 + 4H2 - > CRk + 2H20 (11)
The methanation reaction (Equation 11) is the reverse of cata-
lytic steam reforming of methane. Methane formation is favored
by the use of lower temperatures and the removal of excess
water.
Exit gas from the methanator which contains <20 ppm total carbon
oxides is cooled to 38 °C. Condensate is removed from the syn-
thesis gas in a process condensate drum. Final synthesis gas
at 38 °C and ^2.5 MPa has a 3:1 mole ratio of hydrogen to nitro-
gen and contains less than one percent methane and argon.
Ammonia Synthesis
Many variations of the original Haber process are used commerci-
ally. Most important of these are the modified Haber-Bosch,
Claude, Casale, Fauser, and Mont Cenis processes. Fundamentally,
these processes are the same: nitrogen is fixed with hydrogen
as ammonia in the presence of a catalyst. They vary, however,
in the arrangement and construction of equipment, the composi-
tion of catalysts, and the temperature and pressure used.
The first step in the synthesis process is to compress the syn-
thesis gas from the methanator (Figure 5) . Within the last 15
years, the application of centrifugal compressors to handle
synthesis gas at pressures ranging from 13.8 MPa to 68.9 MPa
has revolutionized the ammonia industry (21, 37). Centrifugal
(37) Looking at Ammonia Technology. Chemical and Process
Engineering, 51(10):5, 1970.
19
-------
compressors require a minimum gas volume at the entrance of any
given stage which can be met only at high capacities and low
synthesis loop pressures. These units have several advantages
over reciprocal compressors: 1) they are cheaper; 2) they re-
quire less frequent and cheaper maintenance; 3) they require no
lubrication and thus eliminate oil entrainment into the syn-
thesis loop, which can foul the catalyst and contaminate the
product; and 4) they can be turbine driven by the steam generated
in the high-pressure steam reforming unit that precedes the
synthesis loop, and still supply exhaust steam at a level suf-
ficient to regenerate the carbon dioxide absorption solution.
Thus, centrifugal compressors involve lower investments and
maintenance costs (36, 38). However, plants using centrifugal
compresscrs consume more natural gas for producing the steam to
drive the compressors, than do plants equipped with reciprocating
compressors (38).
SYNTHESIS GAS
FROM METHANATOR
LIQUID-VAPOR
SEPARATOR
PURGE GAS TO
PRIMARY REFORMER
FOR FUEL
SYNTHETIC
ANHYDROUS
AMMONIA
Figure 5. Ammonia synthesis loop.
(38) Scheel, L. F. Refrigeration: Centrifugal or Recip? Hydro-
carbon Processing, 48 (3) :123-129, 1969.
20
-------
Centrifugal compressors are driven by high-pressure, high tem-
perature steam which is generated in the plant by recovering heat
from the secondary reformer process gas and the primary reformer
flue gases. Fresh synthesis gas from the methanator is com-
pressed to 13.8 MPa to 34.5 MPa and is mixed with recycled syn-
thesis gas and is cooled to 0°C.
Condensed ammonia is separated from the unconverted synthesis
gas in a liquid-vapor separator and sent to a let-down separator.
Unconverted gas is compressed and preheated to ^180°C before
entering the synthesis converter. This gas contains about 13%
inerts and 3.7% ammonia with a hydrogen-to-nitrogen mole ratio
of 3:1.
Synthesis gas enters the converter and is radially dispersed
through the triply promoted iron oxide (FesOiJ synthesis cata-
lyst. Exit gas from the converter contains ^15% ammonia and
%14% inerts and is cooled from ^370°C to ^38°C. Ammonia which
condenses is separated in a primary separator. A small portion
of the overhead gas is taken as a purge to prevent the buildup
of inerts such as argon in the circulating gas system (21).
Purge gas is cooled to -23°C to minimize the loss of ammonia
and used as fuel in the primary reformer (Table 4) (39). A
recently developed cryogenic unit recovers the hydrogen in the
purge gases from the synthesis loop (40). Sufficient hydrogen
can be recovered by this process to produce an additional 5% to
6% ammonia from the feedstock.
TABLE 4. TYPICAL PURGE GAS ANALYSIS (39)
Component Mole %
Hydrogen 60
Nitrogen 20
Argon 3.5
Methane 16.5
TOTAL 100.0
(39) Haslam, A. A., and W. H. Isalski. Hydrogen from Ammonia
Plant Purge Gas. In: Ammonia Plant Safety, Vol. 17,
Chemical Engineering Progress Technical Manual. American
Institute of Chemical Engineers, New York, New York, 1975.
pp. 80-84.
(40) Ammonia Plants Seek Routes to Better Gas Mileage. Chemical
Week, 116(8):29, 1975.
21
-------
Liquid ammonia from the primary, secondary, and purge separators
collects in the let-down separator where the ammonia is flashed
to 0.1 MPa at -33°C to remove impurities such as argon from the
liquid. The flash vapor is condensed in the let-down chiller
where anhydrous ammonia is drawn off and stored in a low temper-
ature (-28°C) atmospheric storage tank or piped to other loca-
tions within the plant.
22
-------
SECTION 4
EMISSIONS
EMISSION CHARACTERISTICS
At a typical synthetic ammonia plant, four process steps are
responsible for air emissions: 1) regeneration of desulfuriza-
tion tanks, 2) primary reforming, 3) regeneration of carbon
dioxide scrubbing solution, and 4) steam stripping of process
condensate.
In addition to emissions from process vents and stacks, potential
fugitive emission sources include leaking seals from ammonia com-
pressors and pumps, ammonia storage tank vents, pressure relief
valves, and ammonia spillages. The following sections charac-
terize the emission species associated with each emission point.
Desulfurization
More than 95% of the ammonia plants in the U.S. use activated
carbon fortified with metallic oxide additives for feedstock
desulfurization (25, 41). Each plant is equipped with a dual-
tank system so that one tank is always on stream while the other
is being regenerated. Regeneration of this activated carbon
desulfurization tank may cause emissions of sulfur oxides (SOX)
and Hydrogen sulfide (H2S), depending on the amount of oxygen
in the regeneration steam. Regeneration also results in hydro-
carbon and carbon monoxide (CO) emissions.
The remaining 5% of the ammonia plants use a tank filled with
zinc oxide for feedstock desulfurization. This tank is not an
emission source because it is replaced instead of regenerated
at the plant site.
The desulfurization tank is regenerated once every 30 days by
passing steam through the bed for an average of 8 hr to 10 hr.
Metallic sulfide is converted to metallic oxide and elemental
sulfur. Calgon Corporation reported that the sulfur remains
on the carbon surface and is not vented to the atmosphere
with the regeneration steam (23-25). However, sulfur has an
(41) Personal communication with N. Walen, The M. W. Kellogg
Company, Houston, Texas, 9 September 1975.
23
-------
appreciable vapor pressure at the regeneration temperatures
(0.8 kPa at 230°C to 6.0 kPa at 290°C), and some sulfur must be
lost by vaporization. In addition, sulfur may react to form H2S
or S02. Because source data are not available, a worst-case
emission factor for sulfur emissions has been calculated in
order to estimate potential environmental effects.
On this basis all sulfur that is adsorbed on the carbon bed is
assumed '-o be vented to the atmosphere during regeneration.
The quantity of sulfur in the carbon bed can be determined from
the sulfur concentration in the feedstock, the feedstock process
rate, the size of the carbon bed, and the age of the carbon tank.
Total sulfur concentrations of pipeline grade natural gas range
from 229 ug/m3 to 915 ug/m3 and average 450 ug/m3 (18, 19, 41).
The carbon bed reduces the sulfur concentration of the exiting
process gas below 280 ug/m3.
Ammonia plants consume 35 GJ to 47 GJ of natural gas per metric
ton of ammonia produced, with an average value of 37 GJ/metric
ton (18, 21). About 75% of this natural gas (26 GJ to 35 GJ)
is desulfurized and used for feedstock, while 25% (8.7 GJ to
11.6 GJ) is used as fuel for the radiant heat section of the
primary reformers. For natural gas with a heating value of
39 MJ/m3, the quantity of feedstock natural gas used to produce
one metric ton of ammonia ranges from 700 m3 to 900 m3, with an
average value of 710 m3.
Desulfurization tank size varies considerably among ammonia
plants and depends on the sulfur concentration of the feedstock,
the production rate, and the desired length of time between
regenerations. Consequently, the tanks at one plant may need
regeneration only once a year, while those at another plant
require regeneration every 5 days. On the average, desulfuriza-
tion tanks are regenerated every 30 days. For design considera-
tions, 1 ir3 of metal-impregnated activated carbon will remove
all of the sulfur species in 2 x 105 m3 of natural gas (21).
An average ammonia plant produces 480 metric tons/day (see
Section 4, page 36). Therefore, on the average this plant
desulfurizes 3.41 x 105 m3/day of natural gas and, for a feed-
stock containing 450 ug/m3 of sulfur, collects 153 g/day of
sulfur in the desulfurization tank. For a 30-day regeneration
schedule, --he tank must hold 51 m3 of carbon and collect a total
of 4.6 kg of sulfur.
Sulfur vapor pressures were obtained by interpulation of data
given in Reference 27.
24
-------
Therefore, a maximum of 4.6 kg of sulfur will be emitted to
the' atmosphere during the 10-hr regeneration period which occurs
once every 30 days. This results in an emission rate of 0.128
g/s over the 10-hr period or 0.053 g/s normalized over a 24-hr
period. For an average ammonia plant producing 480 metric tons
of ammonia per day, the corresponding emission factors are 23
mg/kg of ammonia produced over a 10-hr period and 9.6 mg/kg of
ammonia produced over a 24-hr period. If all of this sulfur is
emitted in the form of S02 or H2S, the corresponding 24-hr emis-
sion factors are 0.019 g/kg or 0.010 g/kg, respectively.
The carbon bed also collects quantities of longer chain hydro-
carbons which are present in the feedstock. Upon regeneration
of the carbon bed with steam, hydrocarbons and carbon monoxide
are emitted. Based on one set of source test measurements in
the public files at the Texas Air Control Board, the emission
factors for total hydrocarbons (measured as methane9) and for
carbon monoxide are 3.6 g/kg and 6.9 g/kg of ammonia produced,
respectively.
Vent heights for the desulfurization tank range throughout the
industry from 6 m to 12 m, with an average height of 10 m.
Primary Reformer
Natural gas and purge gases are fired at 1,000°C to 1,200°C using
10% to 20% excess air in the radiant heat section of the primary
reformer (19, 39). Hot combustion gases are cooled by a series
of four to six waste heat boilers and heat exchangers before
they are vented via a stack to the atmosphere. Final stack gas
temperatures are ^230°C.
Emission species from this source consist of natural gas or fuel
oil combustion products (NOX, CO, SOX, hydrocarbons, and partic-
ulates). Natural gas is currently the primary fuel, but natural
gas shortages have forced about 50% of the ammonia plants to use
No. 2 fuel oil to fire the primary reformer during the winter
months (approximately 4 months).
Data collected from the literature (30), from public files at
the Texas Air Control Board, and from the Louisiana Air Control
Commission on emissions from the primary reformer were used to
establish emission factors as shown in Table 5. All four plants
were 900 metric ton/day units. Emission factors from plants A
and B were established from a material balance and engineering
estimates. Emission factors at plant C were the result of sam-
pling the primary reformer stack gases. Except where noted,
all emission factors are based on natural gas fuel.
aThe test method for hydrocarbons reports total hydrocarbons in
terms of methane equivalents; individual species are not
determined.
25
-------
TABLE 5. PLANT DATA FOR EMISSIONS FROM PRIMARY REFORMER
Plant
Type of data
Emission
species
Emission factor,
g/kg
Material balance
NO
x
4.38
B Material balance NOX
Particulate
SOX
CO
C Stack test NOX
Hydrocarbons
CO
Particulate
S02c'd
De Stack test NOX
NOX
NOX
NOX
NOX
NOX
NOX
NOX
5.95
0.15
0.005
0.003
2.03
0.10
0.14
0.51
0.49
2.71
4.04
2.83
4.23
2.76
1.99
2.57
2.72
Public files at Texas Air Control Board, Austin,
Texas, 1976.
Public files at Louisiana Air Control Commission,
New Orleans, Louisiana, 1976.
Material balance estimates for burning fuel oil in
reformer.
Assuming all sulfur in fuel oil is converted to sulfur
dioxide.
Eight stack tests collected over a nine-month period
from a plant representative of industry (30).
i
Emission factors from plant D come from recent stack test data
at a 900 metric ton/day ammonia plant (30). These numbers were
generated in a study designed to evaluate the effect of adding
the overhead gases from a condensate steam stripper to the
primary reformer stack gases. Measurements reported in Table 5
were collected when no overhead gases were added.
26
-------
Emission factors for criteria pollutants from the primary •
reformer, shown in Table 6, were determined by-averaging the
appropriate values .in Table 5. The uncertainty value associated
with the NOX emission factor was calculated from the Student "t"
test at a 95% confidence limit.
TABLE 6. AVERAGE EMISSION FACTORS FOR EMISSIONS
FROM PRIMARY REFORMER BASED ON PLANT DATA
Emission
species
NOX
CO
SOX
Hydrocarbons
Particulate
Emission
Natural gas
2.7 ± 23%3
0.07
0.005
0.1
0.15
factor, g/kg
Fuel oil
N.A.b
N.A.
0.49
N.A.
0.51
Average based on test measurements only.
Not available.
In order to supplement the data in Table 6, emission factors were
also calculated from data (42) on the combustion of natural gas
and fuel oil (see Table 7). Because emissions from the primary
reformer result from the combustion of fuels used to heat process
gas in the reformer, these calculated values are believed to be a
good estimate of the actual emission factors. The emissions data
in Reference 42 are reported in terms of fuel consumed and must be
converted to the proper units for comparison with Table 6. The
conversion was' based on energy requirements of 8.7 GJ to 11.6 GJ
for every metric ton of ammonia production; i.e., 200 m3 to
300 m3 of natural gas with a heating value of 39 MJ/m3 are
burned in the primary reformer per metric ton of ammonia produced
(18, 21). The corresponding combustion rate for distillate fuel
oil with a heating value of 39 GJ/m3 is 0.2 m3 to 0.3 m3 of fuel
oil per metric ton of ammonia. The combustion rate for each fuel
type was then multiplied by the appropriate emission factors (42)
to arrive at the calculated emission factors in Table 7.
(42) Compilation of Air Pollutant Emission Factors, Second
Edition. Publication No. AP-42, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, April
1973. pp. 1.3-1 to 1.3-4 and 1.4-1 to 1.4-3.
27
-------
TABLE 7. EMISSION FACTORS FOR PRIMARY REFORMER
BASED ON COMBUSTION DATA FOR NATURAL
GAS AND DISTILLATE FUEL OIL
Emission species
Reported emission
factor (42),
of fuel
Calculated emission factor
for primary reformer,
g/kg of product
Natural gas
Particulate
SOX
NOX
CO
Hydrocarbons
(as
290 a
9.63
1,920 to 3,700
270
48
0.058 to 0.087
0.0019 to 0.0029
0.384 to 0.576
0.054 to 0.081
0.0096 to 0.0144
Distillate fuel oil
Particulcite
Sulfur dioxide
Sulfur trioxide
NOX
CO
Hydrocarbons
Aldehydes
1.8
17 sb
0.25 Sb
4.8 to 9.6
0.5
0.35
0.25
0.36 to 0.54
1.02 to 1.53C
0.015 to 0.022C
0.96 to 2.88
0.10 to 0.15
0.070 to 0.105
0.050 to 0.075
a 5
For natural gas containing 460 yg/m3 of sulfur.
S is the percent sulfur in the fuel oil.
Based on fuel oil containing 0.3% sulfur.
A general comparison of Tables 5, 6, and 7 indicates that emis-
sion factors based on stack test measurements are higher than
the corresponding calculated values in Table 7 (i.e., for NOX,
CO, and hydrocarbons), and that emissions from the combustion of
fuel oil are higher than those for natural gas. The following
specific observations and conclusions can be made:
• Measured and calculated NOx emission factors demonstrate
that combustion conditions in the primary reformer are
more severe than those in a typical industrial boiler.
Because nitrogen oxides are formed by the combination of
atmospheric nitrogen and oxygen at elevated temperatures,
more are produced as the temperature, residence time, or
excess oxygen increases. I
• As a result, the actual emissions from the combustion of
fuel oil are expected to be higher than those given in
Table 7. However, they will not be higher by a factor
of five to seven, as in the case of natural gas, because
fuel oil combustion normally produces higher flame tem-
peratures than natural gas.
28
-------
• Sulfur oxide emissions result from the oxidation of sulfur
in the fuel. Consequently, emission factors can be calcula-
ted from a material balance once the fuel sulfur content is
known.
• The greatest effect in changing from natural gas to fuel
oil will be the increase in SOX emissions. These may
increase by a factor of 100 to 1,000 or more, depending on
the percent sulfur in the fuel.
• Conditions that favor an increase in NOX emissions also
favor a decrease in CO and hydrocarbon emissions. In this
respect the test results from Plant C in Table 5 are sur-
prising. Emission factors for NOX, CO, and hydrocarbons
are greater than those in Table 7. One possible explanation
is higher flame temperatures along with some regions of poor
fuel/air mixing. Another possibility is error in the test
measurements. Under good combustion conditions (i.e., ade-
quate fuel/air mixing to ensure complete combustion) CO and
hydrocarbon emission factors are expected to be equal or
lower than those shown in Table 7.
A summary of primary reformer emissions takes these factors into
consideration and is presented in Table 8.
TABLE 8. SUMMARY OF PRIMARY REFORMER EMISSIONS
Emission Emission factor,
species g/kg
Natural gas
Particulate 0.072
SOX 0.0024
NOX 2.7 ± 23%
CO 0.068
Hydrocarbons 0.012
Fuel oil
Particulate 0.45
SOX 1.3
NOX 2.7
CO 0.12
Hydrocarbons 0.15
29
-------
The following are conclusions regarding emissions from the
primary reformer:
NOX - For natural gas combustion the average emission
factor is 2.7 g/kg, based on actual test measurements.
This same value is taken as an estimate for combustion
with fuel oil. Table 7 indicates that NOX levels may
be higher when fuel oil is burned in the primary re-
former, but no test data are available to determine the
actual amount. The increase shown in Table 7 is from
two to seven times the NOX emissions for natural gas.
SOx - None of the data are from source tests. Average
values taken from the material balance calculations
reported in Table 7 are 2.4 mg/kg for natural gas, and
1.3 g/kg for fuel oil.
Particulate - Average material balance data from
Table 7 are 72 mg/kg for natural gas and 0.45 g/kg
for fuel oil.
Particulate - Average material balance data from
Table 7 are 72 mg/kg for natural gas and 0.45 g/kg
for fuel oil.
CO cind hydrocarbons - Emission factors for combustion
with fuel oil as taken from Table 7 are 0.12 g/kg for
CO and 0.15 g/kg for hydrocarbons (including aldehydes).
It is difficult to arrive at final CO and hydrocarbon
emission factors for natural gas combustion without
knowing more about the test measurements at Plant C.
Becc.use natural gas is burned in the primary reformer
with 25% excess air, and because NOX data indicate severe
combustion conditions are present, it is believed the data
from Plant C are higher than the average primary reformer.
Therefore average emission factors obtained from Table 7
are 68 mg/kg for CO and 12 mg/kg for hydrocarbons.
Primary reformer stacks vary in height from 10 m to 70 m.
Average stack height and standard deviation for the stack
heights at 16 ammonia plants is 28.6 m ± 14.9 m.a Stack gas
temperatures range from 150°C to 370°C with an average value
and standard deviation for 16 samples of 226°C ± 58°C.
Regeneration of the Carbon Dioxide Scrubbing Solution
Carbon dioxide is removed from the process gas by absorption in
an amine solution or a carbonate solution. Approximately 80% of
the ammonia plants use a MEA scrubbing solution and the remain-
der use a hot potassium carbonate solution. As the solution
becomes saturated with carbon dioxide, it must be regenerated
resulting in air emissions.
aObtained from public data on file at Texas Air Control Board
and Louisiana Air Control Commission.
30
-------
Carbon dioxide is separated from the scrubbing solution by steam
stripping and then by heating. Stripped gas .containing at least
98.5% carbon dioxide is either vented to the. atmosphere or used
as chemical feedstock for a urea plant. Approximately 30% of the
ammonia plants, employing the MEA system pipe the carbon dioxide
gas to a urea plant. In addition to carbon dioxide, vented gas
contains smaller quantities of water, methane, ammonia, carbon
monoxide, and MEA.
Approximately 20% of the ammonia plants use hot potassium car-
bonate solutions to scrub carbon dioxide from process gas. Sat-
urated solution is regenerated by flashing. Overhead vapors
contain carbon dioxide and water which is condensed from the
vapor and recycled to the regenerator. Carbon dioxide is either
vented to the atmosphere or used as a feedstock for producing
urea. Approximately 50% of the ammonia plants with hot potassium
carbonate systems sell the carbon dioxide gas as chemical feed-
stock. In addition to the carbon dioxide which is vented to the
atmosphere, the gas contains smaller quantities of water, meth-
ane, ammonia, and carbon monoxide.
Gas vented from the regenerator contains a minimum of 98.5% car-
bon dioxide and 1.0% water. Local air pollution control author-
ities have not required ammonia plants to sample this source.
Source test data on these emissions are unavailable. Worst-case
emission factors can be determined from material balances and
unit design calculations.
Data collected from public information files at the Texas Air
Control Board and the Louisiana Air Control Commission in June
1976 concerning emissions from this source are shown in Table 9.
Emissions data from plants A, C, D, E, F, and G were obtained
from material balances and engineering estimates. Data from
plant B are based on unit design calculations.
During regeneration of the C02 scrubbing solution, 1.22 metric
tons of C02 are released per metric ton of ammonia produced (34).
Emission factors developed from data in Table 9 and shown in
Table 10 were calculated by averaging the appropriate values.
Corresponding values from both types of system for total hydro-
carbons, ammonia, and CO were averaged together because data for
establishing separate emission factors for both types of plants
were unavailable, and no definite trend in emission factors was
discernible.
Data from 10 plants indicate that the stack height on the carbon
dioxide solution regenerator ranges from 10 m to 65 m, with an
average value and standard deviation of 27.9 m ± 15.5 m. Stack
gas temperature ranges from 37°C to 146°C, with an average value
and standard deviation of 64°C ± 14°C.
31
-------
TABLE 9. DATA FROM MATERIAL BALANCES AND DESIGN
CALCULATIONS USED TO ESTABLISH EMISSION FACTORS
Emission factor
Plant
Emission species
kg/hr
g/kg
With hot potassium
carboncite system
A
Hydrocarbons
NH3
Hydrocarbons
CO
25.4
37.2
< 6.4
<25
0.65
1.0
<0.1
<0.2
With monoethanolamine system
c
D
E
F
(3
Monoethanolamine
Hydrocarbons
Monoethanolamine
CO
Hydrocarbons
Monoethanolamine
Monoethanolamine
0.59
6.9
6.8
11.8
5.0
0.23
0.14
0.015
0.26
0.15
1.0
0.85
0.02
0.005
Based on unit design calculations for 1,360-metric ton/day
plant.
TABLE 10. EMISSION FACTORS FOR EMISSIONS FROM REGENERATION
OF THE CARBON DIOXIDE SCRUBBING SOLUTION
Emission species Emission factor, g/kg
Hydrocarbons
Ammonia
Monoethanolamine'
Carbon dioxide
CO
0.47
1.0
0.05
1,220
1.0
Only for plants with monoethanolamine
absorption system.
Condensate Stripper
Process condensate forms upon cooling the synthesis gas after low
temperature shift conversion. This water contains quantities of
ammonia, methanol, carbon dioxide, and trace metals. In the
past, ammonia plants discharged the water which contains a
32
-------
high nutrient value that might cause eutrophication in the
receiving stream. The EPA has set effluent standards which
require industry to reduce ammonia discharges in effluent to
<50 kg/day (30). Ammonia plants now use process condensate
steam strippers to remove ammonia from water. Instead of dis-
charging ammonia to the water, plants now discharge it to the
atmosphere. The steam stripping process results in air emissions
of ammonia and methanol. Trace metals are discharged with the
stripper bottoms which are discharged from the plant to a receiv-
ing stream or recycled to the boilers.
Table 11 shows the range in concentrations and average values for
the contaminants in the process condensate from a plant that
reforms natural gas and uses a high- and low-temperature shift
converter (29-32).
TABLE 11. CONTAMINANTS IN PROCESS CONDENSATE (29-32)
Concentration, ppm
Component Range Average
Ammonia
Carbon dioxide
Methanol
Total trace
metals
600
200
200
0.4
to
to
to
to
1
2
1
0
,200
,800
,000
.8
1,000
1,500
500
0.
7
Process condensate also contains trace amounts of metals such as
iron, copper, sodium, zinc, calcium, and aluminum. These can
originate either from carryover or from the leaching operations
in the various catalytic services: desulfurization, reforming,
and shift conversion (29, 33).
Theoretical reaction equations indicate that amines and lower
molecular weight alcohols might also be present in the conden-
sate, but several tests for these species by industry repre-
sentatives proved negative (30, 43).
A 544-metric ton/day unit produces 7.89 x 10"3 m3/s of process
condensate, a 900-metric ton/day unit produces 1.39 x 10~2 m3/s,
and a 1,360-metric ton/day unit produces up to 2.20 x 10~2 m3/s
(29, 32). The quantity of low-pressure, low-temperature (100°C
to 115°C) steam used to strip the impurities from the condensate
ranges from 96 kg/m3 to 240 kg/m3 of condensate treatedfr with an
average value of 144 kg/m3 (29, 41).
(43) Personal communication with J. H. Mayes, Gulf South Research
Institute, Baton Rouge, Louisiana, 1976.
33
-------
For conventional stripping operations where an atmospheric vented
steam stripper is used, ammonia concentration in the condensate
can be reduced to ^20 ppm to 50 ppm, methanol to ^10 ppm to
50 ppm, and carbon dioxide to 40 ppm. (29, 31, 32, 43).
The only available data concerning emissions from the condensate
steam stripper are reported in a recent study by Gulf South
Research Institute where different techniques were evaluated for
handling steam stripper overheads at a typical ammonia plant
(30). Complete material balances were calculated around a 10 m
tall steam stripper at a 900-metric ton/day ammonia plant
generating 1.26 x 10~2 m3/s of process condensate.
The condensate stripper system is shown in Figure 6, and the mass
balance resulting from 65 measurements is shown in Table 12.
PROCESS
CONDENSATE
OVERHEAD
EFFLUENT
•STEAM
Figure 6. Process condensate steam stripper,
TABLE 12. MASS BALANCE AROUND CONDENSATE STEAM STRIPPER-
RESULT OF 65 TEST MEASUREMENTS (30)a
Stream
Stream
flow rates,
kg/hr
Mass flow rate, kg/hr
Carbon
Ammonia Methanol dioxide
Process condensate
Steam
Effluent
Overhead
80
7
81
8
,500
,980
,200
,680
39.
0
0.
41.
2
57
2
21.
0
0.
22.
1
28
7
127
0
0
127
Mass entering stripper does not exactly equal mass exiting
since these values are averages from 65 test measurements.
34
-------
In order to calculate emission factors for ammonia, methanol,
and carbon dioxide emitted in the steam stripper overheads,
individual source test measurements (30) were used. A total of
65 source test measurements were used to calculate the ammonia
and methanol emission factors, and 7 measurements were used for
the carbon dioxide emission factor. Results are shown in
Table 13. Uncertainty values were calculated using the Student
"t" test for 95% confidence limits. Average height of the steam
stripper vent is 24 m.
TABLE 13. EMISSION FACTORS FOP. PROCESS
CONDENSATE STEAM STRIPPER (30)
Emission Emission factor,
species g/kg
Ammonia 1.1 ± 4%
Methanol 0.60 ± 2%
Carbon dioxide 3.4 ± 60%
For comparison, emission factors can be calculated from the
average species concentration in the process condensate (from
Table 11), based on 7.89 x 10~3 m3/s of process condensate that
is generated at a 544 metric ton/day ammonia plant. Process
condensate with 1,000 ppm ammonia and 95% removal efficiency
results in an ammonia emission factor of 1.18 g/kg of ammonia
produced. A methanol emission factor of 0.59 g/kg of ammonia
produced is obtained from an average methanol concentration of
500 ppm and steam stripper removal efficiency of 95%. For
1,500 ppm of C02 in the process condensate, the resulting emis-
sion factor is 1.78 g/kg of ammonia produced. These values are
within 10% of the results in Table 13 for ammonia and methanol,
and within the 6.0% uncertainty given in Table 13 for COj.
Fugitive Emissions
Sources of intermittent fugitive ammonia emissions include
ammonia compressors and pumps, ammonia storage tanks and their
pressure relief valves, and ammonia spillages which occur during
loading operations. The only continuous source of fugitive
ammonia emissions is the building which houses the ammonia com-
pressors. Ammonia may escape from various seals into the room
and then escape from the doors, windows, and ventilation
system.
35
-------
No source test data exist for fugitive emissions. It should be
noted that ammonia leakages within the plant are held to a
minimum ro avoid product loss, and also because government
regulations limit the allowable concentration of ammonia in the
workplace atmosphere to 18 mg/m3, the threshold limit value (TLV)
(44).
POTENTIAL ENVIRONMENTAL EFFECTS
Emissions released during the manufacture of ammonia enter the
atmosphere and are dispersed through the environment. This
section evaluates potential environmental effects from air emis-
sions in terms of the effect of an average plant and also in
terms of the impact of the entire industry.
Representative Ammonia Plant
In the United States, 84 of the 90 synthetic ammonia plants use
the catalytic steam reforming process and natural gas feedstock.
The other six plants, representing less than 2% of the ammonia
production, obtain their hydrogen from brine electrolysis plants.
Only the plants which use steam reforming of natural gas are
studied in detail in this report.
Ammonia plants range in capacity from 8,000 to 655,000 metric
tons/yr. The distribution of plant capacities is shown in Table
Table 14. The influence of the large ammonia plants (>200 x 103
metric tons/yr) is illustrated in Table 14 because they dominate
the industry with 62% of the production capacity. Average plant
capacity is 186 x 103 metric tons/yr. Based on national produc-
tion figures the average plant has an annual production rate of
170 x 103 metric tons. Average daily production rate, assuming
351 days per year operation (2 weeks downtime for schedule main-
tenance) , is 480 metric tons/day. The trend in the ammonia
industry is towards the larger plants, >900 metric tons/day.
Locations of the 90 ammonia plants are presented in Appendix A,
which also shows the population densities of the counties where
the ammonia plants are located. Population densities range from
0.5 to 1,103 persons/km2. Average county population density is
117 persons/km2.
A summary of the emission factors previously developed is shown
in Table 15. Currently there are no air pollution control
devices associated with these emission sources; therefore, these
values are uncontrolled emission factors. Stack heights of the
four emission sources are also listed.
(44) TLVs© Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1976. American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, 1976. 97 pp.
36
-------
TABLE 14. 1976 DISTRIBUTION OF AMMONIA
SYNTHESIS PLANTS BY CAPACITY (16)
Annual capacity,
10 3 metric ton
<10
>10 to 50
>50 to 100
>100 to 150
>150 to 200
>200 to 300
>300 to 400
>400 to 500
>500
TOTAL
Number of
plants
1
13
15
17
15
6
16
4
3
90
Percent of
total plants
1.1
14.4
16.7
18.9
16.7
6.7
17.8
4.4
3.3
100.0
Annual capacity,3
10 3 metric ton
8
408
1,184
2,101
2,707
1,529
5,257
1,769
1,793
16,756
Percent of
total capacity
<0.1
2.4
7.1
12.5
16.2
9.1
31.4
10.6
10.7
100.0
Summed from Appendix A
TABLE 15. STACK HEIGHTS AND UNCONTROLLED EMISSION
FACTORS FOR REPRESENTATIVE AMMONIA PLANT
stack
height,
Emission point m
a
Desulfurization tank 10
Primary reformer 28.6
Natural gas fuel
Number 2 fuel oil
Carbon dioxide regenerator 27.9
Condensate steam stripper 24
Emission factor, gAg
Ammonia N6X
ob
0 2.7 ± 23%
0 2.7
1.0 0.
1.1 ±4% 0
SOx CO Particulate
0.019° 6.9 0
0.0024 0.068 0.072
1.3 0.12 0.45
0 1.0 0
00 0
Total
hydrocarbons
3.6
0.012
0.15
0.47
0
Other
0
0
0
Monoethanol-
amine : ° 0.05
Carbon
dioxide: 1,200
Carbon
dioxide: 3.4 ±
Methanol: 0.60 ±
60%
2%
Intermittent source of emissions; desulfurization tanks are regenerated once every 30 days for 10-hour period.
Zero indicates species not emitted from this source.
Assuming all sulfur is released as S(>2; if H2S is emitted the emission .factor would be 0.010 gA9-
Only at plants with monoethanolamine absorption system.
Source Severity
In order to evaluate the relative significance of emissions from
synthetic ammonia plants, a source severity, S, is defined as:
where x
S =
X
max
(12)
time-averaged maximum ground level concentration
of an emission species
max
F = hazard factor
37
-------
For the criteria pollutants NOX, SOX, CO, hydrocarbons and parti-
culates, F is the primary ambient air quality standard. For
other emission species F is defined by a reduced threshold limit
value (TLV):
F = TLV
x
1
100
(13)
where 8/24 normalizes the TLV to a 24-hr exposure and 1/100 is a
safety factor.
Values of F are reported in Table 16. Equation 13 was not used
to find the hazard factor for C02; instead F was defined as the
average ambient concentration of 630 mg/m3. The ratio Xmax/^
compares an ambient air concentration with a standard concentra-
tion where incipient deleterious effects may begin; it is thus a
measure of environmental severity.
TABLE 16. VALUES OF F USED TO CALCULATE
APPROPRIATE SOURCE SEVERITIES
Primary ambient
Emission air quality TLV (44) •
species standard, mg/m3 mg/m3
Particulate
SOX
Hydrocarbon
CO
Ammonia
Hydrogen sulfide
Monoethanolamine
Carbon dioxide
Methanol
0.26
0.365
0.16b
40
none
none
none
none
none
N.A.3
N.A.
N.A.
N.A.
18
15
6.0
9,000
260
F, |
yg/m3
260
365
160
4,000
60
50
20
6.3 x 10sC
87
£)..»
0.35
0.35
0.50
0.60
0.35
0.35
0.35
0.35
0.35
Not applicable.
3There is no primary ambient air quality standard for hydrocarbons.
The value of 160 yg/m3 used for hydrocarbons in this report is a
recomme;nded guideline for meeting the primary ambient air quality
standard for photochemical oxidants.
•t
'Average ambient C02 concentration.
38
-------
Values of x are calculated from the equations suggested by
Turner (45)?ax
V = V
Amax Amax
^o
t
(14)
where
- 2 Q
max
max
t
Q
h
IT
e
u
ireuh2
instantaneous (i.e., 3-min average)
maximum ground level concentration
instantaneous averaging time, 3 min
averaging time, min
emission rate, g/s
stack height, m
3.14
2.72
wind speed, m/s
For criteria pollutants, the averaging time, t, is the same as
that for the corresponding ambient air quality standard. For
noncriteria emission species, t is 1,440 minutes (24 hours)
(Table 16).
Since NOX has a primary ambient air quality standard with a 1-yr
averaging time, Equation 14 for xmax cannot be used. Instead,
Equation 5.13 in Reference 45 was used for neutral atmospheric
conditions to develop the following source severity equation
(46):
NO,
315 Q
h2.1
(15)
Using Equations 12 through 15 and the data from Tables 15 and 16
values of x and s were calculated for a representative syn-
thetic ammonia plant based on a production rate of 480 met-
ric tons/day. Results are shown in Table 17. Source severity
(45) Turner, D. B. Workbook of Atmospheric Dispersion Estimates,
Public Health Service Publication No. 999-AP-26, U.S.
Department of Health, Education, and Welfare. Cincinnati,
Ohio, 1969. 62 pp.
(46) Reznik, R. B. Source Assessment: Flat Glass Manufacturing
Plants. EPA-600/2-76-032b, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1976.
147 pp.
39
-------
values for sulfur emissions from regenerating the desulfuriza-
tion tank were individually calculated, assuming all sulfur was
emitted in one of two forms: S02 or hydrogen sulfide. Sulfur
emission factors based on a 24-hr average were used because the
sulfur severity equations are derived for 24-hr averages.
TABLE 17. SOURCE SEVERITY VALUES FOR UNCONTROLLED
EMISSIONS FROM REPRESENTATIVE AMMONIA PLANT
Sourc
Emission point Ammonia NOX SOx CO Pai
Desulfurizatian tank 0
Primary refor.ner
Natural gas fuel 0
Number 2 fui;l oil 0
Carbon dioxide regenerator 2.2
Condensate steam stripper 3.2
aAssuming all sulfur is emitted as
Assuming all sulfur is emitted as
0 0.0533 0.30
4.1 <0.001 <0.001
4.1 0.44 <0.001
0 0 0.006
00 0
S02.
hydrogen sulfide.
e severity
ticulate
0
0.034
0.21
0
0
(S)
Total
hydrocarbon
32.4
0.013
0.16
0.54
0
Other
Hydrogen
sulfide:
0
0
Monoethanol
amine : 0
Carbon
dioxide:
Carbon
dioxide:
Methanol :
0.21b
-
.33
0.25
<0.001
0.12
The source severity values in Table 17 were calculated for a
single plant representative of the ammonia industry. In order
to illustrate the potential environmental impact of emissions
from each plant in the entire industry, severity ranges for
each species and each emission point were calculated and are
presented in Table 18. Sulfur emissions from the desulfuriza-
tion tank were calculated by assuming all the sulfur was emitted
either as sulfur dioxide or as hydrogen sulfide, normalized to
a 24-hr emission factor.
Actual stack heights from the 90 ammonia plants for each emis-
sion point are not available; therefore the average stack
heights previously determined for a representative plant were
used in calculations. Average emission factors were employed
for the sa.me reason. As a result, the range in severity values
reflects the range in individual plant production rates. The
minimum severity value is for the plant with the smallest pro-
duction rate and the maximum severity value corresponds to the
plant with the highest production rate.
Table 18 also shows the percentage of plants having a severity
exceeding D.05 and 1.0. At severity value of 1.0, the estimated
average maximum ground level concentration for a given emission
species exceeds the corresponding ambient air quality standard or
reduced TLV. For those emission points where the severity
exceeds 1.0 at some plants, the entire severity distributions are
presented in Figures 7 through 10. Source severity values were
calculated from the emission factors and average plant stack
heights in Table 15 and individual plant capacity data
40
-------
TABLE 18. RANGE OF SOURCE SEVERITIES AND PERCENTAGE OF
PLANTS HAVING SEVERITIES GREATER THAN 0.05 OR 1.0
Source severity
Emission point
Desulfurization tank
Primary reformer
Natural gas
Fuel oil
Carbon dioxide regenerator
Condensate steam stripper
Species
S02
Hydrogen sulfide
CO
Hydrocarbon
NOX
sox „ .
CO
Hydrocarbon
Particulate
NOX
sox
CO
Hydrocarbon
Particulate
Ammonia
CO
Carbon dioxide
Hydrocarbon
Monoethanolamine
Ammonia
Methanol
Carbon dioxide
Minimum
0.002
0.009
0.016
1.3.9
0.18
0.00003
0.00002
0.0006
0.0015
0.18
0.019
0.00003
0.007
0.009
0.094
0.0002
0.011
0.023
0.014
0.14
0.005
0.00004
Maximum
0.187
0.73
1.35
114.0
14.5
0,. 04)3
0.001
0.046
0.12
14.5
1.55
0.002
0.58
0.75
7.68
0.002
0.89
1.91
1.15
11.41
0.43
0.003
Percentage of jplants
S > 0.05
44
88
93
100
100
0
0
0
13
100
96
0
82
86
100
0
89
97
93
100
79
0
S > 1.0
0
0
3
100
86
0
0
0
0
86
7
0
0
0
74
0
0
14
2
83
0
0
-------
o
LU
£
OO
00
23
LL.
o
o
cc
0.02
0.05 0.10 0.20
SOURCE SEVERITY
0.50
1.0
2.0
§ 100
-------
to
Z
<
to
to
2!
to
O
UJ
0.1
0.2
0.5
1.0 2.0
SOURCE SEVERITY
5.0
to
2
<
to
to
UJ
25
to
<
D_
5
o
a:
0.02
0.05
0.1 0.2
SOURCE SEVERITY
0.5
1.0
2.0
Figure 8. Source severity distributions for primary
reformer using oil and gas fuel (emission
factor calculation given in Appendix B).
43
-------
as i.o
SOURCE SEVERITY
5.0
0.01 0.02
0.05 0.1 0.2
SOURCE SEVERITY
a 02
0.05 0.1 0.2
SOURCE SEVERITY
0.5
Figure 9. Source severity distributions
for C02 regenerator.
44
-------
1.0 2.0
SOURCE SEVERITY
5.0
10
20
Figure 10.
Source severity distributions
for condensate steam stripper.
Appendix A. Daily production rates were found by assuming 351
days of operation per year and an annual production rate equal
to 90.5% of capacity (the national average).
Total Industry Emissions
Another measure of the environmental impact from the production
of ammonia is provided by the total amount of industry emis-
sions. These can be compared to total state and national emis-
sions from all sources to find the emission burden due to the
ammonia industry. State and national emission burdens for the
criteria pollutants (NOX, SOX, CO, particulates and hydrocar-
bons) are given in Table 19. Appendix B gives a detailed
description of how the emission burdens were calculated, a table
of the total emissions of each criteria pollutant by the ammonia
industry on a state-by-state basis, and state-by-state listings
of total emissions of criteria pollutants from all sources (47,
48) .
(47) 1972 National Emissions Report. EPA-450/2-74-012, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, June 1974. 422 pp.
(48) Eimutis, E. C., and R. P. Quill. Source Assessment:
State-by-State Listing of Criteria Pollutant Emissions.
EPA-600/2-77-107b, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, June 1977. 138 pp
45
-------
TABLE 19. CONTRIBUTION OF EMISSIONS FROM
AMMONIA PLANTS TO STATE BURDENS
State
Alabana
Alaska
Arizona
Arkansas
California
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Louisiana
Mississippi
Missouri
Nebraska
New Mexico
New York
North Carolina
Ohio
Oklahoma
Oregon
Pennsylvania
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wyoming
United States
Particulate
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.01
Percent
NOX
0.22
6.3
0.31
1.2
0.54
0.16
0.30
3.9
<0.10
<0.10
2.4
1.3
3.2
3.6
0.19
2.8
0.60
<0.10
0.16
0.29
1.8
0.47
0.14
0.38
0.90
0.18
0.51
0.41
0.27
0.54
0.44
of state burden
SOX
<0.10
0.10
<0.10
0.10
<0.10
<0.10
<0.10
0.11
<0.10
<0.10
<0.10
<0.10
0.22
0.18
<0.10
0.13
<0.10
<0.10
<0.10
<0.10
0.24
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.01
I CO
<0.10
7.0
<0.10
0.25
<0.10
<0.10
<0.10
0.19
<0.10
<0.10
2.5
0.62
0.42
0.53
0.14
2.0
2.1
<0.10
<0.10
<0.10
0.74
<0.10
<0.10
0.18
0.13
<0.10
0.15
<0.10
<0.10
1.6
0.02
:
Hydrocarbons
0.10
1.4
<0.10
0.22
<0.10
<0.10
<0.10
0.39
<0.10
<0.10
0.49
0.22
0.35
0.46
<0.10
0.52
0.24
<0.10
<0.10
0.10
0.35'
<0.10
<0.10
0.17
0.13
<0.10
0.11
0.14
0.23
0.17
0.10
The contribution to total state emissions exceeds 1.0% for NOX
in nine states, for CO in five states, and for hydrocarbons
in one state. On a national basis only NOX emissions exceed
0.1%.
Affected Population
Dispersion equations predict that the average ground level con-
centration, x, varies with the distance, x, downwind from a
source. For elevated sources, x is zero at the source (where
46
-------
x = 0), increases to some maximum value, Xmax' as x increases,
and then falls back to zero as x approaches infinity. There-
fore, a plot of xYF vs x will have the following appearance:
DISTANCE FROM SOURCE '
Affected population is defined as the number of nonplant persons
around a representative ammonia plant exposed to a x/F ratio
greater than 1.0. The mathematical derivation of the affected
population calculation is presented in Reference 46. Affected
population values for emissions from a representative ammonia
plant, located in a county with a population density of
117 persons/km2, are shown in Table 20. Also presented is the
population exposed to x/F > 0.05.
TABLE 20. AFFECTED POPULATION FOR X/F > °-05 AND 1-°
Affected population, persons
Emission point
Desulfurization tank
Primary reformer
Burning natural gas
Burning fuel oil
Carbon dioxide regenerator
Condensate stripper
Emission species
3
Total sulfur
CO
Hydrocarbons
NOx
SOx
CO
Particulates
Hydrocarbons
NOx
SOx
CO
Particulates
Hydrocarbons
Ammonia
CO
Carbon dioxide
Hydrocarbons
Honoethanolamine
Ammonia
Carbon dioxide
Hethanol
y
ob
0
130
357
0
0
0
0
357
0
0
0
0
197
0
0
0
0
237
0
0
£> 0.05
8
42
7,764
104,627
0
0
0
0
104,627
1,076
0
479
225
5,626
0
0
869
736
6,220
0
168
As 302; if emissions are H2S, the affected population is 0 for x/F >
and 51 for xA > 0.05.
Zero affected population indicates
around a representative plant.
always less than 1.0 or 0.05
47
-------
SECTION 5
CONTROL TECHNOLOGY
To date, add-on air pollution control devices are not used at
synthetic ammonia plants because their emissions are below cur-
rent stcite emission standards. Process modifications have been
implemented which reduce air emissions and improve the utiliza-
tion of raw materials and energy. Control techniques currently
being used and those proposed for the future are discussed in the
following sections.
DESULFURIZATION
During regeneration of the desulfurization tank, stripping steam
is vented directly to the atmosphere. Air pollution control
devices are not used on this emission source because the regener-
ation lasts only about 10 hours and occurs from once per week to
once per year depending on the plant, with an industry average of
once every 30 days.
PRIMARY REFORMER
Natural, gas is fired in the radiant heat section of the primary
reformer to supply the energy needed to reform the feedstock.
Flue gas containing combustion products first passes over a
series of heat exchangers and is then vented to the atmosphere.
No air pollution control devices are being used for this emission
source because natural gas is considered a clean fuel. Nitrogen
oxide and CO emissions from this source are below state emission
standards.
Burning of the purge gas in the primary reformer is the most
significant process modification used at all ammonia plants to
reduce air pollution and recover energy. (In the past, the purge
gas was vented to the atmosphere.) Purge gas is vented from the
synthesis process to prevent the buildup of inerts such as argon.
Purge gas contains about 60% hydrogen, 20% nitrogen, 3.5% argon,
and 16.5% methane (39).
If present NOX emission standards are made more stringent, it
might be difficult for ammonia plants to meet them because of a
lack of proven NOX emissions control technology. The most
promising applicable control system is the ammonia injection
48
-------
technique (49). At temperatures ranging from 250°C to 450°C, 1.0
moles to 1.3 moles of ammonia per mole of NOx are injected into
the flue gas and, aided by a catalyst, 90+% of the nitrogen
oxides are reduced to nitrogen and water. However, if No. 2 fuel
oil is used in the reformer, the increased particulates and
sulfur would poison the catalyst needed in the ammonia injection
control, technique. Other research is considering ammonia injec-
tion into the combustion chamber.
CARBON DIOXIDE REMOVAL SYSTEM
Carbon dioxide gas liberated during regeneration of the scrubbing
solution is either vented directly to the atmosphere (67% of
plants) or used as a chemical feedstock (33% of plants) for other
processes at the plant complex. Since 98.5% of the gas is carbon
dioxide and 1.0% is water vapor, air pollution control devices
are not installed on this process.
CONDENSATE STRIPPER
To comply with effluent standards, the ammonia production process
includes a condensate stripper that reduces the concentration of
ammonia and methanol in the wastewater stream. The simple con-
densate stripper, used at 95% of the plants, passes low-pressure
steam countercurrent to the condensate in a packed or tray tower.
Stripped condensate is either discharged into a receiving stream
or used as cooling water makeup or as boiler feedwater. Uncon-
densed steam and 95% to 99% of the ammonia and methanol are then
vented directly to the atmosphere. Currently, add-on control
devices are not used at these plants to reduce emissions of
ammonia and methanol.
The few ammonia plants without condensate strippers are currently
researching alternative procedures for handling process conden-
sate because the simple condensate stripper, though it performs
satisfactorily, has four major disadvantages: 1) it does not
permit complete reuse of condensate; 2) it requires a substantial
amount of steam; 3) it has no provision for waste heat recovery;
and 4) it does not eliminate air emissions of ammonia and meth-
anol (15). An alternative procedure for recovering process
condensate and reducing air emissions is described below.
The process shown in Figure 11 can be used for improving overall
performance and eliminating atmospheric pollution (32). This
scheme has a modified overhead reflux and product recovery sys-
tem, and it provides for the use of low level heat in conjunction
with the operation of the carbon dioxide scrubbing system
regenerator. All process condensate is preheated from the hotter
(49) Ushio, S. Japan's NOX Cleanup Routes. Chemical Engi-
neering, 82(15):70-71, 1975.
49
-------
TO AND FROM
co2
REGENERATOR
PREHEATED
NATURAL GAS
VENT
CONDEN
STRIP
PROCESS
SATE
PER
•*__
JL
J"L
%
^ 1 r. *l 1 "
LOW-
—1 PRESSURE
1" x-k STEAM
(SC
TO PRIMARY
REFORMER AS
FEEDSTOCK
SUPERHEATED
REFORMING
STEAM
T
DESUPERHEATER
STRIPPED
CONDENSATE
REFLUX SYSTEM
ENTRAPMENT
(TO STRIPPER)
Figure 11. Modified process condensate strip system (32).
stripper effluent. Stripper overhead is reboiled in a reflux
condenser to minimize the quantity of water in the overhead and
to recover clean condensate. Liquid product taken from the
overhead of the condensate stripper is recycled by injection into
the reforming facility. Superheated reforming steam heats the
overhead product and the resulting gas is combined with preheated
reforming feedstock. All contaminants - carbon dioxide, ammonia,
and methanol - are reformable and should cause no problems in the
catalyst tubes. Condensate stripper bottoms may be used for
cooling tower makeup or may be returned to the boiler feedwater
facility for further treatment.
In a similar but much simpler scheme researched by GSRI (43), the
overhead is simply heated to M75°C, and a reciprocating pump is
used to pressurize the stream to 3.7 MPa. This stream is then
combined with preheated reforming feedstock and sent to the
primary reformer. Results indicate that this process is not
economically feasible because reinjecting the overhead reduces
production and disrupts the delicate steam balance at the ammonia
plant (30).
In the same study (30) four other techniques for disposing of
stripper overheads were investigated:
• Direct discharge to the atmosphere.
• Precipitation of ammonia with magnesium phosphate to
produce a magnesium ammonium phosphate.
50
-------
• Adsorption of ammonia on a vanadium pentoxide catalyst.
• Injection of overheads into the primary reformer stack
along with the combustion gases.
The direct discharge of stripper overheads to the atmosphere was
obviously the least expensive disposal technique, but certainly
did not reduce emissions of ammonia and methanol to the environ-
ment. GSRI found that the precipitation with magnesium phosphate
was impractical because of the larger capital cost and minimum
cost-benefit expectations. Adsorption by vanadium pentoxide was
not fully developed as a viable process and the cost-benefit
ratio was the poorest of the five systems GSRI investigated. The
fifth process, injecting the overheads into the primary reformer
stack, not only had the best cost-benefit ratio, but according to
test measurements, it also reduced ammonia and methanol emissions
to the environment by 59.3% and 74.7%, respectively (30). In
terms of potential environmental effects, this technique decreased
the source severity for ammonia and methanol emissions from the
condensate stripper by 95% and 97%, respectively. The reduction
in severity was greater than the reduction in the amount of mater-
ial released because the stack height for the primary reformer is
28.6 m versus 10 m for the condensate stripper exhaust. However,
although both ammonia and methanol emissions decreased, nitrogen
oxide emissions from the primary reformer increased by 41%. The
source severity for NOX also increased by 41%.
Stack test data in Reference 30 imply that approximately 50% of
the ammonia in the stripper overhead reacts with the NOX in the
reformer stack gases to produce N2 and water. Remaining ammonia
is converted to NOX. Other research studies indicate that the
reduction of ammonia to form N2 and water requires higher temper-
atures than those in the reformer and also requires a catalyst
(49) .
51
-------
SECTION 6
GROWTH AND NATURE OF THE INDUSTRY
PRESENT TECHNOLOGY
In the U.S. 98% of the synthetic ammonia is produced by steam
reforming natural gas. Six plants representing <2% of domestic
production use hydrogen feedstock obtained from salt water elec-
trolysis plants. Other foreign processes for ammonia production
include;: use of naphtha and other light hydrocarbons for feed-
stock, partial oxidation of heavier hydrocarbons (petroleum oil
for distillates), cryogenic recovery of hydrogen from petroleum
refinery gases, and gasification of coal or coke. These produc-
tion techniques are not used in the U.S. because of their high
capital and operating costs compared to the low prices and avail-
ability of natural gas in this country.
Within the last 15 years a dramatic change in the ammonia indus-
try was brought about by the replacement of reciprocal compres-
sors with newly designed centrifugal compressors (37). Centrif-
ugal compressors are designed to handle synthesis gas at pres-
sures ranging from 14 MPa to 70 MPa, resulting in much higher
production capacities, lower synthesis loop pressures, and re-
duced investment and operating costs (38). As a result, older
design plants were rapidly replaced by new plants with larger
capacity. In 1955, a plant producing 270 metric tons/day was
considered large (21). Today, 53% of the ammonia is produced by
plants with a production capacity >900 metric tons/day. A large
plant today produces 1,300 to 2,000 metric tons/day.
EMERGING TECHNOLOGY
As long as natural gas is available and cheaper than oil, plants
in the U.S. will continue to use the catalytic steam reforming
process. However, because of the uncertainty in natural gas
supply and prices, several companies are considering other pro-
duction techniques, such as the use of naphtha for feedstock,
partial oxidation of heavier hydrocarbons, and coal gasification.
Because of natural gas curtailments, ^50% of the plants are
beginning to fire No. 2 fuel oil in the radiant heat section
52
-------
of the primary reformer during the winter months (28, 43, 50).
If gas supplies continue to dwindle, this practice probably will
be extended to full annual operation, resulting in a two orders
of magnitude increase in sulfur oxide emissions and a threefold
increase in particulate emissions from the primary reformer.
In terms of feedstock substitution, estimates indicate that con-
version from natural gas to naphtha would cost 50% to 100% of the
original expense of a new plant (40, 51). For example, a new
900-metric ton/day ammonia plant built to use naphtha would cost
15% more than one built to operate on natural gas because addi-
tional desulfurization and carbon removal equipment would have
to be installed to treat the dirtier feedstock (40).
For situations where light hydrocarbon feedstocks are expensive
or unavailable, the partial oxidation of heavier hydrocarbons is
one process alternative. This process generates raw synthesis
gas noncatalytically at relatively high temperature and pressure,
and uses high-purity oxygen in the combustion step. As shown in
Figures 12 and 13, a partial oxidation plant requires an air
separation plant and additional carbon removal and recycle, de-
sulfurization, and other facilities (17). The increased invest-
ment for a partial oxidation plant over a steam reforming plant
ranges from 20% to 50% (17, 51).
Prior to World War II, practically all ammonia plants outside of
the U.S. used coal as a source of synthesis gas. Coal-based
plants slowly gave way to natural gas and liquid hydrocarbons,
primarily because of the lower investment and favorable price
structure of these feeds. However, with the current energy
supply problem and prospects of rising costs for both natural
gas and petroleum, the use of coal has reentered the picture and
is again being given serious consideration. Several German pro-
cesses such as Lurgi, Koppers-Totzek, and Winkler have been used
for many years to produce synthesis gas for steam generation and
heating and as a feedstock for ammonia, methanol, and synthetic
liquid fuels. These three processes differ somewhat in mechani-
cal design and operating conditions.
The Koppers-Totzek (K-T) process reacts coal, steam, and oxygen
in an entrained bed at atmospheric pressure (Figure 14). Because
synthesis gas leaves the gasifier at relatively high temperatures
(50) Sloan, C. R., and A. S. McHone. The Effect of the Energy
Crisis on Ammonia Producers. In: Ammonia Plant Safety,
Vol. 15, Chemical Engineering Progress Technical Manual,
American Institute of Chemical Engineers, New York,
New York, 1973. pp. 91-95.
(51) Strelzoff, S. Make Ammonia from Coal. Hydrocarbon Pro-
cessing, 53(10):133-135, 1974.
53
-------
HEAVY
OIL STEAM
AIR
en
Figure 12.
CARBON
REMOVAL
STEAM
H2S,-
COS, (
«•
•
:o2
HEAVY
OIL STEAM
! j
GASIFIER
STEAM
GENERATION
i
SULFUR
REMOVAL
i
HIGH TEMP. /
LOW TEMP.
CO SHIFT
1
co2
REMOVAL
(
•*•
A
R
AIR
SEPARATION
METHANATION
i
SYNTHESIS
GAS
COMPRESSION
J
NH3
SYNTHESIS
•PURGE
Ammonia process based on partial
oxidation of heavy hydrocarbons
(alternative A) (17, 51).
Figure 13.
NH,
Ammonia process based on partial
oxidation of heavy hydrocarbons
(alternative B) (17, 51).
-------
(>1,800°C), the gasifier effluent contains no hydrocarbons higher
in molecular weight than methane. Methane content is <0.2% and
the CO/H2 content is ^85%, making it an ideal gas mixture for
ammonia synthesis (52). Outside the U.S., 49 K-T gasifiers are
used to produce synthesis gas for ammonia production.
COAL STEAM
STEAM -
ASH-
COS
AIR
AIR
SEPARATION
METHANOL
SCRUBBING
LIQUID N2
SCRUBBING
SYNTHESIS
-co.
NH,
Figure 14.
Ammonia production based on Koppers-Totzek
coal gasification (52).
In the Lurgi process, coal is gasified at ^2 MPa to 3 MPa using
oxygen and steam in a fixed bed (Figure 15). Because the gasifi-
cation temperature (560°C to 620°C) is in the intermediate range
and operating pressure is high, the content of methane (10%) and
carbon dioxide (28%) in the crude gas is considerably greater
than that in the conventional reforming and partial oxidation
processing (17). Methane can be removed by liquid nitrogen
scrubbing in a downstream processing operation and subsequently
reprocessed for the additional production of hydrogen and carbon
monoxide. High-pressure, low temperature gasifier operation
(52) Rothman, S. N., and M. E. Frank. Opportunities in Ammonia
from Coal. In: Ammonia Plant Safety, Vol. 17, Chemical
Engineering Progress Technical Manual, American Institute
of Chemical Engineers, New York, New York, 1975. pp. 19-23.
55
-------
results in the production of other components in the crude gas
such as tars, naphtha, phenols, and light oils; they must be
removed from the gas stream and can possibly be recovered as
salable product (53).
COAL STEAM
Figure 15. Ammonia process based on Lurgi coal gasification.
The Winkler gasifier falls between the Koppers-Totzek and Lurgi
processes, turning out methane but no liquid byproducts. This
process is an atmospheric fluidized bed in which the gasifying
media are oxygen and steam. The fluid bed operates at 800°C to
1,000°C, and most of the ash is carried over with the product
gas. All size ranges of coal can be used in the Winkler process,
but not that which is strongly coking. Oxygen consumption is
intermediate between that of the fixed-bed Lurgi and the
entrained Koppers-Totzek (53).
(53) Perry, H. Coal Conversion Technology,
neering, 81 (14):88-93, 1974.
56
Chemical Engi-
-------
Other coal gasification processes currently being developed that
could potentially be used to produce synthesis gas for ammonia
plants include the Hygas, C02-Acceptor, Synthane, Bigas, Cogas,
Union Carbide-Battelle, and Hydrane processes (54). However, it
is anticipated that coal gasification will not be used to pro-
duce ammonia until after 1980 (53). At current feedstock and
construction prices, a coal gasification-based ammonia plant
would cost 200% to 300% more than a new steam reforming plant
(17, 18, 52, 54).
INDUSTRY PRODUCTION TRENDS
The trend in synthetic ammonia production from 1960 to the pres-
ent and projected to 1980 is shown in Figure 16. The dramatic
upturn in production during the 1960's was a result of new devel-
opments in centrifugal compressor plants (21). Production rates
leveled off in early 1970 because no new plants were constructed.
Demand for ammonia now exceeds present capacity. Several new
plants are being built and older plants are being expanded (55).
Figure 17 shows the locations of new and expanding plants.
As a result of the increased capacity, it is estimated that
production of ammonia will increase at a rate of 4% to 8% per
year for the next 5 years (55, 56). This will result in a cor-
responding increase of 22% to 47% in emissions from 1975 to 1980.
(54) lammartino, N. R. Coal Chemicals are Making a Comeback.
Chemical Engineering, 82 (18):57-59, 1975.
(55) Wett, T. Outlook for Ammonia Seen Rosy Through the 1970's.
The Oil and Gas Journal, 73 (29):21-23, 1975.
(56) Ammonia Capacity Projections Down in 1975. Nitrogen,
98:6-7, November/December 1975.
57
-------
ZO.OOO
15,000
10.000
I
5,000
PROJECTED/
1960 1965 1970 1975 1980
PRODUCTION YEAR
Figure 16. Production of synthetic anhydrous ammonia (16)
Figure 17. New and expanding ammonia plant locations (55)
58
-------
REFERENCES
1. Current Industrial Reports, Inorganic Chemicals 1976. Pub-
lication No. M28A(76)-14, U.S. Department of Commerce,
Washington, D.C., August 1977. 30 pp.
2. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products June 1977. Publication No. M28B(77)-6,
U.S. Department of Commerce, Washington, D.C., August 1977.
7 pp.
3. Current industrial Reports, Inorganic Fertilizer Materials
and Related Products May 1977. Publication No. M28B(77)-5,
U.S. Department of Commerce, Washington, D.C., July 1977.
7 pp.
4. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products April 1977. Publication No. M28B(77)-
4, U.S. Department of Commerce, Washington, D.C., June 1977.
7 pp.
5. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products March 1977. Publication No. M28B(77)-
3, U.S. Department of Commerce, Washington, D.C., May 1977.
6 pp.
6. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products February 1977. Publication No. .-.: .
M28B(77)-2, U.S. Department of Commerce, Washington, D.C.,
April 1977. 6 pp.
7. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products January 1977. Publication No.
M28B(77)-1, U.S. Department of Commerce, Washington, D.C.,
March 1977. 6 pp.
8. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products December 1976. Publication No.
M28B(76)-12, U.S. Department of Commerce, Washington, D.C.,
February 1977. 6 pp.
9. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products November 1976. Publication No.
M28B(76)-11, U.S. Department of Commerce, Washington, D.C.,
January 1977. 6 pp.
59
-------
10. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products October 1976. Publication No.
M28B(76)-10, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
11. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products September 1976. Publication No.
M28B(76)-9, U.S. Department of Commerce, Washington, D.C.,
November 1976.
12. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products August 1976. Publication No. M28B(76)-
8, U.S. Department of Commerce, Washington, D.C., October
1976. 6 pp.
13. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products July 1976. Publication No. M28B(76)-7,
U.S. Department of Commerce, Washington, D.C., September
1976. 6 pp.
14. Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products June 1976. Publication No. M28B(76)-6,
U.S. Department of Commerce, Washington, D,C., August 1976.
6 pp.
15. Carbone, W. E., and 0. F. Fissore. Ammonia. In: Kirk-
Othmer Encyclopedia of Chemical Technology, Second Edition,
Volume 2. John Wiley & Sons, Inc., New York, New York, 1963,
pp. 258-312.
16. Hairgett, N. World Fertilizer Capacity - Computer Printout.
Tennessee Valley Authority, Muscle Shoals, Alabama, 1976.
7 pp.
17. Bui'.vidas, L. J. , J. A. Finneran, and 0. J. Quartulli.
Alternate Ammonia Feedstocks. In: Ammonia Plant Safety,
Vol. 17, Chemical Engineering Progress Technical Manual.
American Institute of Chemical Engineers, New York,
New York, 1975. pp. 4-18.
18. Strelzoff, S. Partial Oxidation for Syngas and Fuel.
Hydrocarbon Processing, 53 (12):79-87, 1974.
19. Fi;nneran, J. A., L. J. Buividas, and N. Walen. Advanced
Ammonia Technology. Hydrocarbon Processing, 51(4):127-130,
1972.
20. Personal communication with Dayton Power and Light Company,
Dayton, Ohio, July 1976.
60
-------
21. Green, R. V. Synthetic Nitrogen Products. In: Riegel's
Handbook of Industrial Chemistry, Seventh Edition.
J. A. Kent, ed. Van Nostrand Reinhold Company, New York,
New York, 1974. pp. 75-122.
22. Haney, G., and K. Wright. Media for Removing Sulfur from
Natural Gas. In: Ammonia Plant Safety, Vol. 12, Chemical
Engineering Progress Technical Manual. American Institute
of Chemical Engineers, New York, New York, 1970. pp. 50-54.
23. Personal, communication with P. D. Langston, Calgon Corpora-
tion, Pittsburgh, Pennsylvania, 5 September 1975.
24. Personal communication with F. R. Bossi, Calgon Corpora-
tion, Pittsburgh, Pennsylvania, 29 September 1975.
25. Personal communication with W. Lovett, Calgon Corporation,
Pittsburgh, Pennsylvania, 29 September 1975.
26. Air Purification with Granular Activated Carbon, Brochure
23-55, Calgon Corporation, Pittsburgh, Pennsylvania. 1975.
p. 24.
27. Liley, P.E., and W. R. Gambill. Physical and Chemical Data.
In: Chemical Engineers' Handbook, Fifth Edition, Section 3,
R. H. Perry and C. H. Chilton, eds. McGraw-Hill Book Com-
pany, New York, New York, 1973. p. 48.
28. Personal communication with J. C. Barber, Tennessee Valley
Authority, Muscle Shoals, Alabama, June 1976.
29. Quartulli, 0. J. Stop Wastes: Reuse Process Condensate.
Hydrocarbon Processing, 54(10):94-99, 1975.
30. Romero, C. J., F. Yocum, J. H. Mayes, and D. A. Brown.
Treatment of Ammonia Plant Process Condensate Effluent.
EPA-600/2-77-200, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, September 1977.
85 pp.
31. Spangler, H. D. Repurification of Process Condensate. In:
Ammonia Plant Safety, Vol. 17, Chemical Engineering Progress
Technical Manual. American Institute of Chemical Engineers,
New York, New York, 1975. pp. 85-86.
32. Quartulli, 0. J. Review of Methods for Handling Ammonia
Plant Process Condensate. Presented at the Fertilizer
Institute Manufacturing Environmental Seminar, New Orleans,
Louisiana, January 14-16, 1976. 20 pp.
61
-------
33. Finneran, J. A., and P. H. Whelchel. Recovery and Reuse of
Aqueous Effluent from a Modern Ammonia Plant. In: Ammonia
Plant Safety, Vol. 13, Chemical Engineering Progress Tech-
nical Manual. American Institute of Chemical Engineers,
New York, New York, 1971. pp. 29-32.
34. Strelzoff, S. Choosing the Optimum CC>2-Removal System.
Che.mical Engineering, 82 (19) : 115-120, 1975.
35. Hahn, A. V. The Petrochemical Industry - Market and
Economics. McGraw-Hill Book Company, New York, New York,
1970. pp. 19-38.
36. Haase, D. J. New Solvent Cuts Costs of Carbon Monoxide
Recovery. Chemical Engineering, 82 (16):52-54, 1975.
37. Looking at Ammonia Technology. Chemical and Process
Engineering, 51(10):5, 1970.
38. Scheel, L. F. Refrigeration: Centrifugal or Recip? Hydro-
carbon Processing, 48 (3):123-129, 1969.
39. Haslam, A. A., and W. H. Isalski. Hydrogen from Ammonia
Plant Purge Gas. In: Ammonia Plant Safety, Vol. 17,
Chemical Engineering Progress Technical Manual. American
Institute of Chemical Engineers, New York, New York, 1975.
pp. 80-84.
40. Amir.onia Plants Seek Routes to Better Gas Mileage. Chemical
Week, 116(8):29, 1975.
41. Personal communication with N. Walen, The M. W. Kellogg
Coirpany, Houston, Texas, 9 September 1975.
42. Compilation of Air Pollutant Emission Factors, Second
Edition. Publication No. AP-42, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, April
1973. pp. 1.3-1 to 1.3-4 and 1.4-1 to 1.4-3.
43. Personal communication with J. H. Mayes, Gulf South Research
Institute, Baton Rouge, Louisiana, 1976.
44. TLVs® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Chs.nges for 1976. American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, 1976. 97 pp.
45. Turner, D. B. Workbook of Atmospheric Dispersion Esti-
mates. Publication Health Service Publication No.
999-AP-26, U.S. Department of Health, Education, and Wel-
fare. Cincinnati, Ohio, 1969. 62 pp.
62
-------
46. Reznik, R. B. Source Assessment: Flat Glass Manufacturing
Plants. EPA-600/2-76-032b, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1976.
147 pp.
47. 1972 National Emissions Report. EPA-450/2-74-012, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, June 1974. 422 pp.
48. Eimutis, E. C., and R. P. Quill. Source Assessment: State-
by-State Listing of Criteria Pollutant Emissions. EPA-600/
2-77-107b, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, July 1977. 138 pp.
49. Ushio, S. Japan's NOX Cleanup Routes. Chemical Engi-
neering, 82(15):70-71, 1975.
50. Sloan, C. R., and A. S. McHone. The Effect of the Energy
Crisis on Ammonia Producers. In: Ammonia Plant Safety,
Vol. 15, Chemical Engineering Progress Technical Manual,
American Institute of Chemical Engineers, New York,
New York, 1973. pp. 91-95.
51. Strelzoff, S. Make Ammonia from Coal. Hydrocarbon Proc-
essing, 53(10):133-135, 1974.
52. Rothman, S. N., and M. E. Frank. Opportunities in Ammonia
from Coal. In: Ammonia Plant Safety, Vol. 17, Chemical
Engineering Progress Technical Manual, American Institute
of Chemical Engineers, New York, New York, 1975. pp. 19-23,
53. Perry, H. Coal Conversion Technology. Chemical Engi-
neering, 81(14):88-93, 1974.
54. lammartino, N. R. Coal Chemicals Are Making a Comeback.
Chemical Engineering, 82 (18):57-59, 1975.
55. Wett, T. Outlook for Ammonia Seen Rosy Through the 1970's.
The Oil and Gas Journal, 73(29):21-23, 1975.
56. Ammonia Capacity Projections Down in 1975. Nitrogen,
98:6-7, November/December 1975.
57. Standard for Metric Practice. ANSI/ASTM Designation E380-
76e, IEEE Std 268-1976, American Society for Testing and
Materials, Philadelphia, Pennsylvania, February 1976.
37 pp.
63
-------
APPENDIX A
SYNTHETIC AMMONIA PLANTS IN THE U.S. IN 1976
Table A--1 describes the 90 synthetic ammonia plants in operation
in the U.S. in 1976, listing the company name and location, an
estimate of annual plant capacity, and the population density of
the county where the plant is located.
64
-------
TABLE A-l.
SYNTHETIC AMMONIA PLANTS IN THE
UNITED STATES IN 1976 (16)
Company
Air Products and Chemicals, Inc.
Allied chemical Corp., Agricultural Division
American Cyanamid Co . , Agricultural Division
Apache Powder Co.
Beker Industries, Inc.
Borden Inc., Chemical Division
Camex, Inc.
Central Farmers Fertilizer Co.
Colorado Oil and Gas Corp., Wycon Chemical Co., Subsidiary
Columbia Nitrogen Corp.
Commercial Solvents Corp.
Cooperative Farm chemicals Assoc.
Diamond shamrock Corp.
The Dow Chemical Co.
E. I. Du Pont de Nemours and Co., Inc.
Explosives Dept.
Industrial and Biochemicals Dept.
Plastics Dept.
El Paso Natural Gas Co.
FMC Corp . , Inorganic Chemicals Div .
Farmers Chemical Assoc., Inc.
Farmers National Chemical Co.
Farmland Industries, Inc.
Felmont Oil Corp.
First Mississippi Corp.
Gardinier, Inc.
W. R. Grace and Co. , Agricultural Products Group
Goodpasture, Inc.
Green Valley Chemical Corp.
Gulf and western Industries, Inc.
Hawkeye Chemical Co.
Hercules, Inc.
New Oi
____ _^_^_____
Location
'leans, LA
Pace, JFL
Geisnu
Hopew
South
Omaha
New 0
Benso
Conda
Carls
Geisnu
Borge
Donal
Fremo
Terre
Cheye
Augus
Sterl
Lawre
Dumas
Freep
Beaum
Belle
Victo
Odess
South
Tunis
Tyner
r, LA
11, VA
Point, OH
HE
leans, LA
, AZ
ID
iad, NM
r, LA
, TX
sonville , LA
It, HE
Haute, IN
me, WY
a. GA
ngton , LA
ice, KS
TX
irt, TX
>nt, TX
WV
:ia, TX
1, TX
Charleston, WV
, NC
TO
Plainjriew, TX
Dodge
Ft. D
Hasti
Enid,
Plain
Clean
Ft. K
Tampa
Big S
City, KS
>dge, IA
igs, NE
OK
view, TX
, NY
adison, IA
, PL
jring, TX
Memphis, TO
Dimml
Great
Palme
Clint
Hercv
Louie
tt, TX
an, IA
rton, PA
on, IA
les, CA
iana, MO
Estimated
capacity,
10^ metric tons/yr
190
68
308
308
290
180
308
14
127
190
258
363
( 308
\ 308
43
122
166
118
308
308
145
104
308
308
90
104
22
190
149
54
181
181
127
362
24
77
308
118
113
300
77
32
36
123
63
63
County
population
density,
persons/km2
1,103.0
116.8
46.7
39.6
47.0
446.3
1,103.0
3.8
1.4
3.7
46.7
10.6
46.7
13.3
105.2
7.9
192.2
7.6
38.6
5.9
28.8
75.7
95.4
22.8
38.4
95.4
24.5
170.4
13.2
7.8
25.7
20.7
20.3
13.2
23.4
30.7
180.2
15.7
367.5
4.5
12.0
47.7
31.4
557.8
9.4
EPA
Region
6
4
6
3
5
7
6
9
10
6
6
6
6
7
5
8
4
6
7
6
6
6
3
6
6
3
4
4
6
7
7
7
6
6
2
7
4
6
4
6
7
3
3
9
7
(continued)
65
-------
TABLE A-l (continued)
Company
Hooker Chemical. Corp., Industrial Chemical Div.
Kaiser Aluminuii and Chemical Corp.
Jupiter Chemicnls, Inc.
Lone Star Producing Co., NIPAK, Inc., Subsidiary
MISCOA (Mississippi Chemical Corp. and Coastal Chemical Corp.)
Triad Chemical
Mobil Oil Corp., Petrochemicals Div.
Monsanto Co., Agricultural Div.
Occidental Pet:;oleum Corp. , Best Fertilizer Div.
01 in. Agricultural Chemicals Div.
pennwalt Chemicals Corp.
Phillips Pacific Chemical Co.
Phillips Petroleum Co.
Reichhold chemical, Inc.
Rohm and Haas Co.
St. Paul Ammon-ia Products, Inc.
J. R. Simplot Co., Minerals and Chemical Div.
Standard Oil of Ohio
Vistron
Standard Oil Co. of California
Chevron Chemical Co.
Standard Oil of Kentucky
Standard Oil Co. of Indiana
Tenneco , Inc .
Tennessee Valley Authority
Terra Chemical 3 International, Inc.
Tipperary Corp .
union Oil Co. of California, Collier Carbon and Chemical Corp.
United States 'iteel Corp., USS Agri-Chemicals, Inc.
Valley Nitroge.i Producers, Inc.
Vulcan Materials Co., chemicals Div.
The Williams Companies, Agrico Chemicals
Location
Tacoma, WA
Savannah, GA
Lake Charles, LA
Kerens , TX
Pryor, OK
Pascagoula, MS
Yazoo City, MS
Donaldsonville, LA
Beaumont , TX
Luling, LA
Lathrop, CA
Plainview, TX
Lake Charles, LA
Portland, OR
Pinley, WA
Beatrice , NE
Pasadena, TX
St. Helens, OR
Deer park, TX
East Oubuque, IL
Pocatello, ID
Joplin, MO
Lima, OH
El Segundo, CA
Richmond, CA
Fort Madison, IA
Pascagoula, MS
Texas City, TX
Houston, TX
Wilson Dam, AL
Sioux City, IA
Lovington , NH
Brea, CA
Kenai, AK
Cherokee, AL
Clairton, PA
Geneva , UT
El Centre, CA
Chandler, AZ
Helm, CA
Wichita, KS
Donaldsonville, LA
Tulsa, OK
Blytheville, AR
Estimated
capacity,
10s metric tons/yr
20
136
120
113
95
159
358
345
236
408
87
47
435
45
8
141
190
209
81
45
| 190
100
123
522
12
118
95
463
(92
(463
190
81
190
90
(118
463
160
363
63
190
32
160
36
308
386
308
County
population
density,
persons Am2
93.3
158.6
49.7
10.8
12.8
44.8
11.2
46.7
98.5
37.1
77.6
13.2
49.7
91 0
2.L . 2
SOO.l
14.9
11.6
385.9
17.2
385.9
13.5
17.5
47.3
104.4
5.6
151.7
30.7
44.8
160.3
385.9
31.8
44.6
4.3
695.8
0.5
31.7
841.4
26.4
6.7
40.2
26.3
133.4
46.7
267.7
26.3
EPA
Region
10
4
6
6
6
4
4
6
6
6
9
6
6
10
10
7
6
10
6
5
10
7
5
9
9
7
4
6
6
4
7
6
9
6
4
3
8
9
9
9
7
6
6
6
66
-------
APPENDIX B
TOTAL MASS OF EMISSIONS
The total mass of emissions from ammonia plants was calculated on
a state-by-state basis and compared to total state emissions from
all sources. Only criteria pollutant emissions can be compared
because a comprehensive data base is available for only these
materials. Table B-l shows the production rate and masses of
emissions for each state. State production rates were determined
by multiplying the state capacities, as summed from the appropri-
ate values in Appendix A, by the ratio of national production in
1976 to total national capacity as given in Appendix A; i.e.,
multiplying state capacities by 0.9053.
TABLE B-l.
MASS OF EMISSIONS FROM SYNTHETIC
AMMONIA PLANTS IN 1976
State
Alabama
Alaska
Arizona
Arkansas
California
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Louisiana
Mississippi
Missouri
Nebraska
New Mexico
New York
North Carolina
Ohio
Oklahoma
Oregon
Pennsylvania
Tennessee
Texas
Utah
Virginia
Washington
West Virginia
Wyoming
United States
Estimated ammonia
production,
Ma
10 3 metric tons/yr Particulate
218
419
42
279
784
168
230
206
172
110
841
475
3,263
887
168
489
253
70
172
735
763
81
361
406
2,605
57
279
146
339
150
15,168
Includes monoethanolamine and methanol .
29
57
6
38
106
23
31
28
23
15
114
64
441
120
23
66
34
9
23
99
103
11
49
55
352
8
38
20
46
20
2,051
bTotals
ss of emissions,
NOX
589
1,131
113
753
2,117
454
621
556
464
297
2,271
1,283
8,810
2,395
454
1,320
683
189
464
1,985
2,060
219
975
1,096
7,034
154
753
394
915
405
40,954
nay not
SOX
48
92
9
61
172
37
50
45
38
24
184
104
716
195
37
107
55
15
38
161
167
18
79
89
571
13
61
32
74
33
3,325
metric tons/yr
CO Hydrocarbons3
210
404
40
269
755
162
222
198
166
106
810
458
3,144 3,
855
162
471
244
67
166
708
735
78
348
391
2,510 2,
55
269
141
327
145
14,616 16,
238
457
46
305
856
183
251
225
188
120
918
519
562
968
183
534
276
76
188
802
833
88
394
443
844
62
305
159
370
164
557
agree due to rounding errors.
67
-------
Statewide mass emissions were computed by multiplying the appro-
priate emission factors from Table 15 by the quantity of ammonia
produced in each state. Since total state emissions are reported
for only criteria pollutants, the masses of ammonia, MEA, meth-
anol, or carbon dioxide emitted on a state basis were not calcu-
lated. Methanol and MEA emissions were included with total hydro-
carbons. In certain cases, calculations had to be modified as
follows:
• Because the desulfurization tank is regenerated once every
30 days, annual emissions of S02, CO, and hydrocarbons
from this emission point were determined by applying a
correction factor of 12 days/365 days = 0.0329.
• Total emissions from the primary reformer were computed
based on the assumption that 50% of the ammonia plants
burned No. 2 fuel oil for 4 months/yr. Natural gas was
used in the others. The ratio of fuel oil burned to
natural gas burned is then 1:5.
• As previously stated, 80% of the ammonia plants use the
MEiA process for scrubbing carbon dioxide from the syn-
thesis gas, and 20% use the hot potassium carbonate
process. Also, about 30% of the MEA plants and 50% of
the hot potassium carbonate plants do not vent the
regeneration gases, but sell them as a chemical feed-
stock. Therefore, the fraction of synthetic ammonia
plants that have an atmospheric discharge from this
process is (8/10 x 7/10 + 2/10 x 5/10) = 0.66.
With these considerations in mind, total emission factors were
established for each criteria pollutant. Since NOX is only emit-
ted from the primary reformer, its emission factor was 2.7 g/kg.
The total SOX emission factor was determined as follows:
Total emission factor = ~ (0.019 g/kg) + - (0.0024g/kg) + ~ (1.3 g/kg)
Desulfurization tank Reformer, Reformer,
natural gas fuel oil
= 0.22 g/kg of ammonia
In the case of CO emissions, a similar calculation was used:
Total emission factor - -j|| (6.9 gAg) + f (0.068 gAg) + ~ (0.12 gAg) + 0.66 (1.0 9A9>
Desulfurization Reformer, Reformer, C02 regenerator
tank natural gas fuel oil
a 0.96 gA9 °*~ ammonia
The total particulate emission factor was determined as follows
68
-------
Total emission factor = g- (0.072 g/kg) + i- (0.45 g/kg)
Reformer, Reformer,
natural gas fuel oil
= 0.14 g/kg of ammonia
MEA and methanol emissions were added to the total hydrocarbon
emission factors in the following manner:
Total emission factor - -~ (3.6 gAg) + f (0.012 gAg) + j (0.15 gA^> + (0.66) (0.47 gAg) + (0.8) (0.7) (0.05 gAg) + 0.6 gAg
Desulfurization Reformer, Reformer,; CO2 regenerator, COa regenerator. Steam
tank natural gas fuel oil j hydrocarbons MEA stripper,
methanol
•1.09 gAg of ammonia
Table B-2 gives total state emissions for the five criteria pol-
lutants as reported in the National Emissions Data System (47).
Table B-3, a recent emissions listing prepared by Monsanto
Research Corporation under U.S. EPA contract (48) , was used to
compute the ratios which are shown in Table 19.
69
-------
TABLE B-2. NEDS EMISSION SUMMARY BY STATE (47)
Mass of emissions, metric
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
U.S. TOTALS
Particulates
1,178,
13,
72,
137,
1,006,
201,
40,
36,
19,
226,
404,
61,
55,
1,143,
748,
216,
348,
546,
380,
49,
494,
96,
705,
266,
168,
202,
272,
95,
94.
14,
151,
160,
481,
78,
1,766,
93,
169,
1,810,
13,
198,
52,
409,
549,
71,
14,
477,
161,
213,
411,
75,
16,762,
643
913
685
817
452
166
074
808
451
460
574
621
499
027
405
493
351
214
551
155
921
160
921
230
355
435
688
338
040
920
768
044
017
978
056
595
449
598
073
767
336
704
399
692
587
494
934
715
558
427
000
1,
2,
2,
1,
1,
1,
2,
2,
1,
28,
Adjustments
The U.S. summary does not
should be considered part
nationwide emissions.
New York point sources
Forest wild fires
Agricultural burning
Structural fires
Coal refuse fires
TOTAL
U.S. SUBTOTAL
U.S. GRAND TOTAL
include
of the
311,
375,
272,
52,
100,
1,110,
16,762,
17,872,
SO
882
5
679
39
393
49
168
209
60
897
472
45
54
043
050
283
86
202
166
144
420
636
466
391
50
152
871
58
304
86
463
345
473
78
980
130
36
929
65
247
17
179
753
152
17
447
272
678
712
69
873
to
X
,731
,874
,768
,923
,326 1
,188
,068
,310
,630
,381
,418
,981
,387
,020
,541 1
,416
,974
,827
,664
,887
,037
,466
,935 2
,633
,591
,373
,235
,014
,851
,596
,736
,979
,020
,537
,333 1
,705
,776
,137 3
,761
,833
,354
,982
,098 1
,526
,751
,394
,991
,348
,393
,394
,000 21
NOX
397,
32,
123,
168,
,663,
147,
155,
58,
46,
644,
369,
44,
48,
974,
,371,
242,
233,
419,
442,
76,
265,
334,
,222,
311,
172,
448,
148,
101,
88,
67,
489,
572,
412,
85,
,101,
222,
135,
,017,
46,
521,
49,
42,
,303,
80,
24,
329,
187,
229,
408,
72,
,722,
068
757
871
989
139
496
832
407
824
794
817
221
552
372
233
524
987
142
817
741
204
379
438
834
519
300
405
94S
933
309
216
451
599
708
470
687
748
345
921
544
490
454
801
998
286
308
923
598
525
572
000
tons/yr
Hydrocarbon
643,410
28,389
189,981
195,538
2,160,710
193,456
219,661
63,886
41,789
619,872
458,010
89,530
84,230
1,825,913
600,477
316,617
309,633
326,265
1,919,662
122,918
295,867
440,481
717,891
410,674
195,950
413,130
271,824
127,821
53,673
88,469
819,482
1,262,206
447,238
70,289
1,153,493
341,358
234,669
891,763
65,833
907,833
90,478
362,928
2,218,891
98,2,2
41,980
36,416
344,643
116,155
523,930
55,319
23,994,000
1
8
2
2
6
2
1
1
1
5
1
1
3
1
1
2
4
1
5
1
3
4
1
6
1
1
1
91
CO
,885
167
815
843
,237
875
897
204
190
,695
,036
275
343
,412
,933
,440
,002
,189
,633
376
,261
,682
,243
,760
829
,854
611
569
215
256
,877
,881
,734
318
,205
,456
92
,729
283
,222
387
,469
,897
40
150
,548
,659
494
,582
303
,782
,657
,357
,454
,204
,667
,781
,580
,227
,834
,817
,010
,566
,720
,718
,780
,621
,375
,932
,827
,196
,804
,218
,526
,740
,094
,901
,061
,522
,751
,380
,319
.922
,398
,679
,719
,627
,247
,830
,650
,168
,356
,253
,748
,527
,510
,031
,117
,214
,869
,297
,000
grand total
certain source categories.
U.S.
000
000
000
000
000
000
000
000
grand
1,
28,
29,
total for a
993
15
128
076
873
949
,000
0
,000
0
,000
,000
,000 21
,000 22
more
382,
88,
29,
6,
31,
536,
,722,
,258,
The
following additions
accurate picture
000
000
000
000
000
000
000
000
127,000
529,000
272,000
61,000
62,000
1,051,000
23,994,000
25,045,000
of
3
1
5
91
96
44
,089
,451
200
308
,086
,782
,868
,000
,000
,000
,000
,000
,000
,000
,000
70
-------
TABLE B-3. STATE LISTING OF EMISSIONS AS OF MAY 13, 1977 (48)
Alabama
AlaBka
Arizona
Arkantaa
Colorado
Connecticut
Delaware
Florida
Georgia
HA wall
Idaho
Illinois
Indiana
Louisiana
Main*
Maryland
Michigan
Minnesota
Mississippi
Missouri
Montana
Pebraska
Nevada
New Joraey
Now York
Ohio
Oklahoma
Oregon
Pennsylvania
Khode Island
South Dakota
Tenneisea
Texas
Utah
Vermont
Virginia
Washington
Wisconsin
Wyoming
U.S. TOTALS
3,196,000
2.55000
2.47000
0.24600
233,800
2.43000
1.750PO
1,151,000
1,003,000
0.77400
7*4,700
0.59100
2.23000
1,516,000
1,209,000
2.43000
2.72000
2.08000
1.59000
2,317,000
2,717,000
2.15000
2,570,000
291,300
1,539,000
1.19000
1.64000
2,809,000
2.17000
129,500,000
Mass of1 emission*.
Percent of U.S
802 300
57,280
0.23500
0.93000
0.82400
37,620
6.54000
3.53000
209,200
151,900
0.62400
392,500
1.61000
1.59000
109,400
0.58700
1.21000
5.06000
3.51000
69,170
0.15900
320,100
17,640
433,000
2.47000
243,300
0.99900
36,4
0.394
1.030
0.781
17,2
6.090
4.670
93,2
33,0
0.336
1.470
1.520
66,5
0.607
1.230
4.290
3.170
116,5
0.130
85,2
3,6
148,0
2.010<
74,3
0.004(
total*
0
0
98,040
0.59600
147,600
0.16700
0
0
0.87800
1.25000
1.63000
0.12100
0
52,910
64,920
0
0
0
5.00000
2.53000
187,400
7.26000
14.90000
32,660
0
10
10
10
10
57,100
0.34400
1.47000
1.51000
209,500
1.25000
41,000
0.04650
3.26000
41,740
6.10000
1.21000
160,400
0.18200
0
0.14100
14,720
0.01670
10
10
10
0.69700
6.61000
2.05000
11,740
0.01330
4.10000
0.28200
'0
241,100
99,030
0
0.21600
0.01160
0.00647
0
4
0
0
0
0
69,930
21,100
270,800
1.69000
97,100
O.S8600
2.17000
862,900
6.071
179,100
0.11800
9,118
0.01040
71
-------
GLOSSARY
carbon monoxide shift: High and low temperature catalytic
reaction in which steam is added to transform carbon
moxioxide to carbon dioxide and hydrogen.
desulfutrization: Removal of hydrogen sulfide from natural gas
feedstock, prior to reforming, by use of an activated
carbon or zinc oxide bed.
methanation: Catalytic reaction in which hydrogen in process
gas converts trace amounts of carbon monoxide and carbon
dioxide to methane and water.
primary reformer: Set of catalyst-filled tubes in which natural
gas (methane) reacts with steam to form carbon monoxide
and hydrogen.
secondary reformer: Catalytic reactor in which compressed air
is mexed with process gas from primary reformer to produce
a synthesis gas with a hydrogen-to-nitrogen mole ratio
of 3:1.
steam stripper: Column in which process condensate flows down
the column countercurrent to steam which extracts ammonia
and methanol from condensate.
72
-------
CONVERSION FACTORS AND METRIC PREFIXES (57)
CONVERSION FACTORS
To convert from
degree Celsius (°C)
joule (J)
kilogram (kg)
kilogram (kg)
kilogram/meter3 (kg/m3)
kilometer2 (km2)
meter (m)
meter3 (m3)
meter3/second (m3/s)
metric ton
pascal (Pa)
pascal (Pa)
second (s)
to
degree Fahrenheit (°F)
British thermal unit (Btu)
pound-mass
(Ib-mass avoirdupois)
ton (short, 2000 Ib mass)
pound-mass/gallon
(Ib-m/gal)
mile2
foot
foot3
gallon/minute (gpm)
ton (short, 2000 Ib-m)
atmosphere
pound-force/inch2 (psi)
minute
Multiply by
top =
9.479
2.205
1
8
3
3
3
1
1
9
1
.102
.348
.861
.281
.531
.585
.102
.869
.450
1,
x
X
X
X
X
X
X
X
.8 tor + 32
10-"
T O™
_ o
10- 1
101
10"
10~6
10-"
1.667 x 10-2
METRIC PREFIXES
Prefix Symbol
giga
mega
kilo
milli
micro
G
M
k
m
y
Multiplication
factor
109
10 6
103
10~3
io-6
Example
5 GJ = 5 x IO9 joules
5 MPa = 5 x 106 pascals
5 kg = 5 x 103 grams
5 mg = 5 x 10"3 gram
5 yg = 5 x 10"^ gram
(57) Standard for Metric Practice. ANSI/ASTM Designation E 380-
76 e, IEEE Std 268-1976, American Society for Testing and
Materials, Philadelphia, Pennsylvania, February 1976.
37 pp.
73
-------
EPA-600/2-77-107fri *
TECHNICAL REPORT DATA
(Please read luaniction! or. ti'.c /.'ft. rsc tiulbrc completing)
1. REPORT NO.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUDTITI.E
Source Assessment: Synthetic Ammonia Production
5. REPORT DATE
November 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
G. D. Rawlings and R. B. Reznik
8. PERFORMING ORGANIZATION REPORT NO.
MRC-DA-736
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
10. PRC/GRAM ELEMENT NO.
1AB015; ROAP 21AXM-071
11. CONTRACT/GRANT NO.
68-02-1874
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 8/75-9/77
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES
IERL-RTP task officer for this report is Ronald A. Venezia, Mail
Drop 61, 919/541-2547. Similar reports are also in the EPA-600/2-76-032 series.
16. ABSTRACT
The report describes a study of air emissions from the production of syn-
thetic ammoma. In 1976, 90 synthetic ammonia plants in 30 states produced 15.2
million metric tons of anhydrous ammonia. Ammonia is synthesized by the reaction
of nitrogen and hydrogen. Most plants produce hydrogen by the catalytic steam refor-
ming of natural gas. An average ammonia plant has a capacity of 180,000 metric tons/
year. Plant air emissions result from regeneration of the desulfurization tank, from
combustior in the primary reformer, from regeneration of the CO2 scrubbing solu-
tion, and from steam stripping of process condensate. On the average, emissions
from the regeneration of the desulfurization tank are released for 10 hours, but only
once every 30 days; emissions from the other sources are continuous during plant
operation. These emission points are not controlled because no state or federal
standards are exceeded. Process modifications have reduced air emissions and
improved utilization of raw materials and energy. Potential environmental effects
from ammonia plant emissions were measured: highest continuous source severities
result from NOx emissions from the primary reformer (4.1), and ammonia emissions
from the regeneration of the CO2 scrubbing solution (2.2) and from the condensate
steam stripper (3. 2). Annual ammonia production should increase by 4-8 % through
1980. Industry emissions should also increase at this rate.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COSATl Field/Group
Air Pollution
Industrial Processes
Assessments
Ammonia
Synthesis
Nitrogen
Carbon Dioxida
Hydrogen
Natural Gas
Steam
Reforming
Catalysis
Desulfurization
Snrubbers
Air Pollution Control
Stationary Sources
CO2 Scrubber
13B
13H
14B
07B
2 ID
07D
07A
3. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS Tliix Report)
Unclassified
21. NO. OF PAGES
81
20. SECURITY CL/-. •-> (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
74
------- |