AMMONIUM NITRATE
AP-42 Section 6.8
Reference Number
2
EPA-600/2-77-107J
September 1977
Environmental Protection Technology Series
;Note: This is a reference cited in AP 42, Compilation of Air Pollutant Emission Factors, Volume I Stationary;
\Point and Area Sources. AP42 is located on the EPA web site at www.epa.gov/ttn/chief/ap42/ !
;The file name refers to the reference number, the AP42 chapter and section. The file name :
:;"ref02_c01s02.pdf" would mean the reference is from AP42 chapter 1 section 2. The reference may be ?
:;from a previous version of the section and no longer cited. The primary source should always be checked.?
-------
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 Health 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.
-------
MRC-DA-582
EPA-600/2-77-1071
September 1977
SOURCE ASSESSMENT:
AMMONIUM NITRATE PRODUCTION
by
W. J. Search 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 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 legislation. If control technology is unavailable,
inadequate, or uneconomical, then financial support is provided
for the development of the needed control techniques for indus-
trial and extractive 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 for investing tax dollars in pro-
grams to develop control technology for a large number (>500) of
operations in the chemical industries. As in any technical pro-
gram, the first question to answer is, "Where are the unsolved
problems?" This is a determination 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 judg-
ment, sufficient environmental risk associated with the process to
invest in the development of control technology. This report con-
tains the data necessary to make that decision for the air emis-
sions from ammonium nitrate 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 categories:
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 ammonium nitrate production, Dr. Ronald A. Venezia served
as EPA Task Leader.
111
-------
ABSTRACT
This report describes a study of air pollutants emitted by the
ammonium nitrate industry. The potential environmental effect of
the source was evaluated using source severity (defined as the
ratio of the maximum ground level concentration of an emission to
the ambient air quality standard for criteria pollutants or to a
modified TLV for noncriteria pollutants).
Representative processes and an average plant were defined for the
purpose of establishing a base on which to determine the emissions
and severity of the source. The industry produces 39% of its orig-
inal solution capacity as ammonium nitrate solutions and 61% as
solids of which 92% are formed by prilling. The remaining 8% are
formed by granulation or graining. Primary emissions from ammo-
nium nitrate plants are particulates and ammonia. Processes
releasing the greatest amount of emissions are the neutralizer
(particulates and ammonia) and the prill tower (particulates).
Emission factors were found to be highly dependent on individual
plant operation.
This report was submitted in partial fulfillment of Contract
No. 68-02-1874 by Monsanto Research Corporation under the sponsor-
ship of the U.S. Environmental Protection Agency. The study
described in this report covers the period April 1975 to July 1977
IV
-------
CONTENTS
Preface iii
Abstract iv
Figures vii
Tables viii
Abbreviations and Symbols x
I Introduction 1
II Summary 2
III Source Description 6
A. Process Description 6
1. Introduction 6
2. Ammonium Nitrate Production by Prilling 9
3. Other Processes 17
B. Materials Flow 19
C. Average Plant Parameters 22
IV Emissions 24
A. Emissions from Each Process 24
1. Neutralizer Operations 25
2. Evaporator/Concentrator Operation 28
3. Prilling Tower Operation 29
4. Cooler and Dryer Operation 31
5. Fugitive Emissions 31
B. Low Density Prilling 33
C. Potential Environmental Effect 34
V Control Technology 41
A. Particulates 41
B. Ammonia or Nitric Acid 45
C. Potential Impact of Controls 45
D. Future Considerations 45
v
-------
CONTENTS (continued)
VI Growth and Nature of the Industry 47
A. Present Technology 47
B. Emerging Technology 47
C. Industry Production Trends 48
References 50
Appendices
A. Raw Data Used to Calculate Emission Factors and
Plant Characteristics 53
B. Derivation of Soruce Severity Equations 57
C. Estimate of Steam from Neutralizer 61
D. Calculation of Affected Population 62
Glossary 64
Conversion Factors and Metric Prefixes 66
VI
-------
FIGURES
Number
1 Flow diagram of ammonium nitrate production.
2 Generalized flow sheet of ammonium nitrate
production processes. 7
3 Physical states of ammonium nitrate products. 8
4 Geographical distribution of ammonium nitrate
plants. 9
5 Capacity distribution of ammonium nitrate plants. 10
6 Prilling process flow diagram. 11
7 Stengel reactor. 18
8 Materials flow diagram - prilled ammonium nitrate. 20
9 Source severity versus distance for the release
of emissions from confinement for coating
operation. 33
10 Source severity distribution for neutralizers
used in making solutions and high density
prills. 36
11 Source severity distribution for evaporator/
concentrators used in making solutions and high
density prills. 36
12 Source severity distribution for high density
prilling towers. 36
13 Source severity distribution for high density
cooling drums. 37
14 Source severity distribution for the evaporator/
concentrator in low density prilling processes. 37
15 Prilling tower showing placement and operation of
the CFCA collecting cone. 43
16 Full-scale Brink collection unit. 43
17 Brink high efficiency element collection
efficiency. 44
18 Capacity and production trends. 48
VI1
-------
TABLES
Number Page
1 Capacity Distribution of Product Types 3
2 Particulate Emission Parameters for Average
Production Operations 4
3 Size Analysis of Prilled Ammonium Nitrate 15
4 Polymorphic Changes of Ammonium Nitrate 16
5 Material Balance for Low Density Prilling Process 21
6 Material Balance for High Density Prilling Process 21
7 Emission Parameters for Ammonia from Neutralizer
Exhaust Based on Approximation of pH 26
8 Emission Parameters for Ammonia from Neutralizer
Exhaust Based on Source Testing 26
9 Emission Parameters for Particulates from
Neutralizer Exhaust Based on Dissociation/
Recombination Theory 28
10 Emission Parameters for Particulates from
Evaporator/Concentrator Based on Dissociation/
Recombination Theory 29
11 Operating Conditions Used to Obtain Data on
Particulate Emissions from Prilling Tower
Operations 30
12 Emission Parameters for Particulates from Prilling
Tower 30
13 Particulate Emission Parameters for Low Density
Prilling Operation 34
14 Particulate Emissions for Average Processes 35
15 Percent of Plants Exceeding Specified Values of
Source Severity, S 38
viii
-------
TABLES (continued)
Number Page
16 Particulate Emissions from Ammonium Nitrate Plants
by State 39
17 Particle Size Distribution in the Neutralizer and
High Density Prilling Tower Airstreams 41
18 Controlled Particulate Emissions 46
19 Production of Ammonium Nitrate by End Use, 1976 49
A-l Ammonium Nitrate Plant Locations and Capacities 54
A-2 Ammonium Nitrate Plant Particulate Emission
Factors 55
A-3 Prill Tower Emissions 56
A-4 Representative Stack Heights 56
IX
-------
ABBREVIATIONS AND SYMBOLS
AAQS
a, b, c
E
e
F
h
h
P
P
Q
Q
S
S
T
t
to
total
total
— the area around a plant site where the source
severity, S, is >1.0, km2
— one-half of the cross-sectional area of a building
perpendicular to the flow of wind, m2
— ambient air quality standard, yg/m3
— constants in the atmospheric dispersion equation
for a
z
— distance downwind from a plant building to the
plant boundary, km
— population density, persons/km2
— emission factor, g/kg
— 2.72
— hazard factor, g/m3
— stack height, m
— average stack height, m
— affected population
— pressure, cm of Hg
— emission rate, g/s
— overall emission rate for an entire plant, g/s
— source severity
— overall source severity for an entire plant
— absolute temperature, K
— averaging time used in calculating )(
I
— instantaneous averaging time of 3 min
-------
ABBREVIATIONS AND SYMBOLS (continued)
TLV — threshold limit value, mg/m3
u -- national average wind speed of 4.5 m/s
x, y, z — rectangular coordinates around a plant site; x is
the wind direction, y is perpendicular to the
wind direction, and z is the vertical direction, m
Xj, X2 — distances downwind where the source severity
equals 1.0, m
a — the ratio 2/ireu
TT — 3.14
a — standard deviation of horizontal dispersion, m
a — standard deviation of vertical dispersion, m
Z
~X — average ground level concentration, yg/m3
X — maximum ground level concentration, yg/m3
max
X — average maximum ground level concentration, yg/m3
max
total ~ overa11 Xmax for an entire plant
X / 0 0\ — concentration downwind with no vertical or hori-
{ ' ' ' zontal component
XI
-------
SECTION I
INTRODUCTION
Ammonium nitrate (NH4N03), made from ammonia (NH3) and nitric acid
(HN03) to give either liquid or solid product, is used in fertil-
izers as a source of nitrogen, in certain types of explosives, in
the production of nitrous oxide, and in other miscellaneous
applications.
This report evaluates the environmental impact of air emissions
from ammonium nitrate production by identifying emission points,
emission characteristics, and the process operating variables
which affect the emissions. A set of criteria is developed to
further evaluate the potential environmental effects of the emis-
sions. The potential impact of available air pollution control
equipment on emissions is also discussed.
-------
SECTION II
SUMMARY
Total ammonium nitrate (100% NH^MC^) production in the U.S. for
1974 was 7.12 x 106 metric tons.3 Major products were 2.78 x 106
metric tons of ammonium nitrate solution and 4.34 x 106 metric
tons solid product. Ammonium nitrate was produced at 64 plant
sites located in 30 states and 59 counties.
Figure 1 presents a flow diagram of the ammonium nitrate manu-
facturing process. Ammonia and 55% nitric acid react in a neutral-
izer to form a solution, and the heat of reaction concentrates the
solution to 83% NH^NC^. The water given off in this process, as
well as that given off by the following concentration step, is
either exhausted to the atmosphere or condensed. The 83% solution
is further concentrated to 95% or 99+% in an evaporator/concentrator,
Solid ammonium nitrate particles can then be formed from the con-
centrated solution by graining, granulation, or prilling. Both high
and low density prills are made, and together they account for
>92% of the solid ammonium nitrate produced. High density prills
(860 kg/m3 bulk density) are made from a 99+% solution, while low
density prills (770 kg/m3 bulk density) are made from a 95%
solution.
NH,
PARTICIPATES.
AMMONIA,
NITRIC ACID
NEUTRALIZER
PARTICULATES
EVAPORATOR /
1 "" CONCENTRATOR ~T
PARTICULATE5
t
PARTICLE
""" FORMATION
COA1
PARTICULATES MATE
1
DRYING
(OPTIONAL),
COOLING,
SCREENING
* 1
MARKET MARKET
\
riNG
RIAL
COATING
BAGGING
BULK
STORAGE
-
MARKET
Figure 1. Flow diagram of ammonium nitrate production.
1 metric ton = 1Q)6 grams; conversion factors and metric system
prefixes are presented at the end of this report.
-------
After forming, the particles are dried (low density prills only),
cooled, screened, and in some cases coated to enhance shelf life
characteristics. In over 90% of high density prill production
no coating takes place; rather, an additive is used to enhance
shelf life. In a very few instances (<5%) coating is used with
a prill containing an additive.
The finished product is stored in bulk or shipped in train hopper
cars or bulk trucks. Less than 10% of the product is bagged in
23-kg to 45-kg quantities for shipment.
Ammonium nitrate plants produce several final product types.
Available production facilities for each product type or combina-
tion thereof are distributed as presented in Table 1.
TABLE 1. CAPACITY DISTRIBUTION OF PRODUCT TYPES
Final product
Solutions
Prills
Grains or crystals
Prills and solutions
Other combinations
Number of
plants
18
13
5
24
4
Percent of
total capacity
26.7
40.8
5.1
23.7
3.6
An average ammonium nitrate plant has a solution capacity of
131,500 metric tons/yr and is located in a county with a popula-
tion density of 209 persons/km2. The average capacity of a high
density prilling operation is 95,000 metric tons/yr, and the aver-
age for low density prilling is 62,700 metric tons/yr. The actual
operating rate for a particular ammonium nitrate plant is a func-
tion of season and geographical area.
Emissions released during the manufacture of ammonium nitrate con-
sist of particulates and either ammonia or nitric acid. Emission
points are shown in Figure 1 and listed in Table 2 with the corre-
sponding particulate emission factors (grams of particulate per
kilogram of product) and emission rates. The level of uncertainty
associated with the average emission factors for the neutralizer
and prilling tower was obtained by applying the "student t" test
to input data. The "t" test involves the estimation of the aver-
age value of a sample population and the establishment of confi-
dence ranges within which the true average value is likely to exist,
Either ammonia or nitric acid is released from the neutralizer,
depending on which reactant is present in excess. The average
process uses ammonia in excess because it reduces particulate
emissions and hence plume opacity. Ammonia emissions have been
found to range from 0.026 to 3.14 g/kg, reflecting the wide vari-
ability in operations. Plants monitor the neutralization reaction
-------
TABLE 2. PARTICULATE EMISSION PARAMETERS FOR AVERAGE PRODUCTION OPERATIONS
Emission point
Solutions and high density prills
b
Neutralizer L
Evaporator/concentrator
Prill tower"
Coolerd
e f
Low density prills '
Neutralizer
Evaporator/concenteator
Prill tower
Predryer
Dryer
Cooler
Coating
Emission
factor,3
g/kg
1.64 ± 84%
0.47C
1.37 ± 10%
0.05C
0.045
0.088
0.496
0.015
0.009
0.016
3.09
Emission
rate,
g/s
6.85
1.96
4.1
0.15
0.089
0.17
0.987
0.030
0.018
0.032
6.0
Xmax'
539
190
44
16
7.0
17
10.6
2.9
1.7
3.3
36h
Source
severity
2.07
0.70
0.17
0.06
0.03
0.06
0.04
0.01
0.01
0.01
0.14
Affected
population,
persons
95
0
0
0
0
0
0
0
0
0
0
Emission factors are for uncontrolled operations except for predryer, dryer, and cooler.
Parameters are based on an average neutralizer capacity of 131,500 metric tons/yr.
Percent uncertainty not reported since value is determined by theoretical calculations.
d
Parameters are based on average high density prilling capacity of 95,000 metric tons/yr.
Q
Parameters are based on average low density prilling capacity of 62,700 metric tons/yr.
Percent uncertainty not reported since data are based on tests at one plant.
q
Value based on material balance.
Value of ground level concentration at plant boundary; coating is a source of fugitive
emissions at ground level, and x cannot be calculated.
by pH measurements, but may unintentionally release more ammonia
than expected.
As a measure of potential environmental impact, the average maxi-
mum ground level concentration, Xmax' and tne source severity, S,
were determined for emissions from each process based on average
capacities (Table 2). In the case of particulates, S is the ratio
of Xmax to tne ainbient air quality standard (260 yg/m3). For
ammonia the air quality standard is replaced by a reduced thres-
hold limit_yalue (TLV®); i.e., TLV x 8/24 x 1/100. The range of
values of Xmax an^ S for ammonia are 89 yg/m3 to 1,003 yg/m3
and 0.1 to 17.2, respectively; particulate values are given in
Table 2.
Those persons living in the area around a plant where the source
severity exceeds 1.0 are termed the affected population. Values
of the affected population for particulate emissions from the vari-
ous processes are shown in Table 2. For ammonia emissions, this
population may range from 0 to 1,058 persons.
-------
The national and state emission burdens of the ammonium nitrate
industry were determined for particulates. The national emission
burden is the ratio of industry particulate emissions to total
national particulate emissions from all stationary sources,
expressed as a percent. The state particulate emission burden is
the same ratio on a state-by-state basis. For the ammonium
nitrate industry the national emission burden is 0.11%. The state
emission burden for Nebraska is 1.2%. All others are below 1.0%;
80% are below 0.5%.
Control technologies with the capability of lowering particulate
emissions have been developed for the ammonium nitrate industry.
One system consists of a large collection cone in the prill tower
and a Brink® mist eliminator with a 98.6% removal efficiency to
treat a combined stream consisting of the prill tower, neutralizer,
and evaporator/concentrator exhausts. With this equipment, the
combined stream has a source severity of 0.19.
A second control method for the prill tower exhaust employs
scrubbers on the top of the tower. Pilot plant tests have shown
efficiencies >90%. Coupled with this system is a redesigned
neutralizer that emits particulates in quantities 10 to 20 times
lower than conventional neutralizers. The high efficiency wet
scrubbers that are usually used on dryer and cooler exhausts have
a removal efficiency of at least 99%.
Ammonium nitrate production is expected to increase at a rate of
2.0%/yr through 1978. With no additional controls, particulate
emissions will be 13% greater in 1978 than in 1972. However,
since plants are adding controls, and because of a possible shift
in emphasis from solid to liquid ammonium nitrate, 1978 emissions
may be lower than those of 1972.
-------
SECTION III
SOURCE DESCRIPTION
A. PROCESS DESCRIPTION
1. Introduction
Four major processes for producing ammonium nitrate are currently
used in the United States. Three of these (prilling, graining,
and granulation by spheroidization) have a common starting point
in that they first produce an 83% ammonium nitrate solution which
is then concentrated to either a range of 95% to 96% or >99.5%.
They then differ in the manner by which the solid product is
formed. The fourth, the Stengel process, differs from these three
in the method by which the feed reactants are mixed and concen-
trated as well as in the solid formation method.
A generalized flow sheet of these four ammonium nitrate processes
is shown in Figure 2. Other known processes that are either obso-
lete or not in use in the U.S. are: crystallization, the Fauser
process, the Bamag-Meguin A.G. process, and pan granulation. These
will not be treated in this investigation.
This assessment of ammonium nitrate production centers on the
prilling process, since at least 92% of all solid ammonium nitrate
in the U.S. is produced by this method. Other ammonium nitrate
manufacturing processes utilize essentially the same initial pro-
cess steps as prilling; i.e., neutralization and concentration.
The remaining processes (graining, granulation by spheroidization,
and Stengel granulation) are discussed in Section III.A.3.
In the prilling process, ammonium nitrate is produced by the fol-
lowing exothermic reaction of ammonia and nitric acid:
NK3 + HN03 —> NH^NOs (1)
When a 55% nitric acid feed stream is used, the product of the
above reaction is an aqueous solution of ammonium nitrate (61%).
In practice the heat of reaction (108.8 kJ) is used to drive off a
portion of the water and concentrate the solution to 83% ammonium
nitrate.
Two solid products can be made by the prilling process: low
density prills and high density prills. The basic difference in
-------
STENGa PROCESS
FINES RECYCLE
Figure 2. Generalized flow sheet of ammonium
nitrate production processes.
production is the solution concentration entering the prill tower.
For low density material, a 95% to 96% ammonium nitrate solution
is sprayed into the prill tower. As the droplets descend and cool,
the ammonium nitrate solidifies, trapping water in the solid
prills. The water evaporates when the prills are dried, leaving
void space which results in prill bulk density of 770 kg/m3.
Approximately 40% of all prilled ammonium nitrate is low density
material. The same process is used for high density prills except
that more concentrated ammonium nitrate solution is used (99.5+%).
This results in less void space and produces prills with bulk
density of 860 kg/m3. Approximately 60% of all prilled ammonium
nitrate is high density material.
After solidifying, prills are screened to remove oversize and
undersize material. Low density prills then go to a two-stage
drying process to drive off excess water, followed by cooling and
additional screening. High density prills do not require the
drying step and go directly to cooling followed by screening.
Low density prills which are not used as an on-site intermediate
(e.g., in explosive manufacturing or fertilizer mixing) are coated
-------
with diatomaceous earth or other material as a water barrier.
High density prills are generally not coated (^3% are coated);
instead, they are formulated with an additive to enhance shelf life.
The final product may be stored in warehouses (for a short period
of time), shipped in bulk railroad cars or bulk trucks, or bagged
in 23-kg or 45-kg quantities.
In 1974, 64 plants produced 7.12 x 106 metric tons of solution
in terms of 100% NH^NC^ (1) Figure 3 illustrates the final physical
states of the solution based on actual production and indicates
that 92% of all solid product is formed by prilling, 7% by granula-
tion, and 1% by graining.
TOTAL
SOLUTION
PRODUCTION -^ SOLIDS (61*1
PERCENTAGES REFLECT WEIGHT PERCENT.
SOLIDS)
Figure 3. Physical states of ammonium nitrate products.
Field observations and discussions with experts in the industry
indicate that particulate emissions (per unit product) from pril-
ling are greater than those from graining, Stengel granulation or
granulation by spherodization. Presumably this is because pril-
ling places greater stresses on the liquid than do the other
processes, resulting in smaller particles. This fact, in conjunc-
tion with the high air flow rates used in prilling compared with
other processes, makes prilling a worst case situation in terms of
emissions. Therefore, only prilling emissions have been studied
in detail in this assessment because they provide an upper limit
of emissions.
JFour presurveys were conducted at various plant locations to
inspect different ammonium nitrate manufacturing methods and to
observe the quantities of emissions from the different processes
These field observations are referenced in this report.
(1) Personal communication with E. A. Harre, National Fertilizer
Development Center, Tennessee Valley Authority, Muscle Shoals,
Alabama, June 1976.
8
-------
Table A-l (Appendix A) lists the 64 U.S. ammonium nitrate plants
by company name and location (city, state, and county or parish),
and includes county or parish population densities and annual
total solution capacities. The geographical distribution of
ammonium nitrate plants is shown in Figure 4 (1).
Figure 4. Geographical distribution of ammonium nitrate plants.
A frequency distribution of plant sizes is presented in Figure 5,
indicating the mean, median, and mode plant capacities. At least
60% of the plants have capacities less than or equal to the mean,
131.5 x 103 metric tons/yr. The diversity in the final product
types made at each site has previously been shown in Table 1;
this information is based on a confidential Monsanto Company survey,
2. Ammonium Nitrate Production by Prilling
The basic prilling process (2) consists of spraying hot, concen-
trated ammonium nitrate solution from the top of a tower. During
their descent countercurrent to a lower temperature airstream, the
droplets are formed into spherical particles between 84 ym and
2.38 mm in diameter. (One industry contact mentioned that occa-
sionally particles >3.36 mm were manufactured.) This basic proce-
dure is actually only one step of what is termed the prilling
process today, which includes the following:
(2) Williams, L., L. F. Wright, and R. Hendriks. Process for the
Production of Ammonium Nitrate. U.S. Patent 2,402,192 (to
Consolidated Mining and Smelting Co. of Canada, Ltd.),
June 18, 1946.
9
-------
100
PLANT CAPACITY, 103 metric tons / yr
Figure 5. Capacity distribution of ammonium nitrate plants.
• Feed preparation and recycle
• Neutralizer and liquid storage
• Evaporation/concentration
• Particle formation
• Product preparation
• Dust control
The above areas are distinguished as six sections on the general-
ized flow sheet in Figure 6 by dashed lines (3). Existing dust
control systems are discussed later in Section V.
a. Feed Preparation and Recycle Loops—
Section 1 in Figure 6 contains the feed pretreatment and the re-
cycle loops entering the main process stream before the neutralizer
The liquid ammonia feed is heated to vaporization by one of several
methods. In some processes, part of the liquid ammonia is passed
through a coil placed in the neutralizer steam exhaust, and the
remainder is used as refrigerant for the air supplying the cooler.
The temperature of the ammonia feed stream is 66°C to 77°C.
Nonprilling processes utilize countercurrent, U-tube, horizontal
heat exchangers, either one or two in series. The series arrange-
ment is used primarily for those processes requiring superheated
feeds; e.g., the Stengel process.
(3) Shearon, W. H., Jr., and W. B. Dunwoody. Ammonium Nitrate
Industrial and Engineering Chemistry, 45(3) :496-504, 1953.
10
-------
SECTION 1 J~ SECTION 2 | SECTION 3
LIQUID NH4N03
TAKEOFF FOR
SHIPMENT
LUMP-DISSOLVING
TANK
| j ^ J
Figure 6. Prilling process flow diagram.
-------
Nitric acid feed (55%) may be heated or unheated. In most cases
(>85%) it is heated, using a preheater, to ^82°C (it is usually
below 88°C to prevent extreme acid corrosion).
A lump-dissolving tank recycles the oversize and undersize mater-
ial resulting from screening the prill tower product. The over-
size and fine materials enter the tank, dissolve, and are sent to
the neutralizer or evaporator/concentrator as a weak (^60%)
recycle liquor. In some plants, this tank is a sealed unit with
an agitator. In other plants it is simply an open collection pool
at the bottom of the tower.
Some ammonium nitrate facilities are portions of entire nitrogen
fertilizer complexes consisting of ammonia, nitric acid, ammonium
nitrate, and urea plants. In such cases the feed to the ammonium
nitrate unit may come from the urea process. This occurs in <20%
of the facilities, however, and will not be considered standard
operating procedure.
b. Neutralizer and Adjusting Tank—
Section 2 of Figure 6 contains the neutralizer and an adjusting
tank. The neutralizer is a vertical, stainless steel vessel 1.8 m
to 3.7 m in diameter and 4.6 m high. Its size is dependent upon
other design considerations in the individual plant. The ammonia
gas is introduced into the neutralizer below the nitric acid noz-
zle. The reaction, in effect, is carried out under a hydrostatic
head of ammonium nitrate solution by placement of the solution
overflow pipe at the appropriate height (see neutralizer in Fig-
ure 6). Agitation is achieved by the feed sprays and water vapor
created by the heat of reaction. Since the reaction is exothermic
(NH3 + HN03 -> NHitNOs + 108.8 kJ) , the heat is sufficient to concen-
trate the ammonium nitrate leaving the neutralizer to 83% and
produce steam that leaves the reactor and, in some plants, heats
the ammonia feed as previously described.
The neutralizer can be operated with a slight excess of ammonia or
or nitric acid. When operating with nitric acid in excess, the
solution pH is automatically controlled at 1.5 by a recorder-
controller which actuates a valve in the nitric acid line. When
operating with an ammonia excess, the pH of a condensed sample
from the offgases is manually maintained between 9.5 and 11.0.
Industrial sources are quick to point out that pH control is not
easy. At the neutral point, the pH titration curve is nearly ver-
tical. In addition, the pH probes have an approximate 94°C tem-
perature limit, and the liquor must be condensed and cooled care-
fully so that salting does not occur.
Under normal operating conditions, the neutralizer temperature is
maintained at 131°C either by adding the weak (^60%) ammonium
nitrate recycle liquor or by varying the nitric acid feed strength.
To minimize ammonium nitrate decomposition, the maximum tempera-
ture permitted during abnormal operating periods is 149°C. When
impurities arising from low grade nitric acid feed are present
(such as copper, zinc, oil, wood, cotton, and chloride impurities)
12
-------
the neutralizer temperature is held to a maximum of 141°C in
periods of abnormal operation. (In normal practice, low grade
nitric acid is not used as a feed stream.)
The coil in the bottom of the neutralizer is used only when the
unit is shut down or when necessary to keep the ammonium nitrate
in solution. Heat can be introduced to the neutralizer by passing
steam through this coil during startup to facilitate ammonia
vaporization. Water may also be introduced directly into the neu-
tralizer for emergency shutdown.
The adjusting tank is used to store the 83% ammonium nitrate solu-
tion from the neutralizer, to receive the overflow from the head
tank on the prilling tower, and supply the evaporator/concentrator
on a demand basis. The actual concentration in such tanks may
therefore range from 81% to 83%.
c. Evaporator/Concentrator—
Section 3 in Figure 6 consists of the evaporator/concentrator.
In this portion of the process the 81% to 83% ammonium nitrate
solution from section 2 is concentrated to a 95% to 96% solution
for low density prill production or to a 99.5+% solution for high
density prill production. Liquid ammonium nitrate for use as a
nitrogen source in liquid fertilizer can be taken from the process
before or after the evaporator/concentrator, as shown in Figure 6,
giving solution concentrations of 83% or 95% to 96%, respectively.
The unit shown in Figure 6 is a vacuum, film-type evaporator used
by >75% of the ammonium nitrate production facilities. It oper-
ates at 57 kPa using a single steam-jet ejector and produces a 95%
ammonium nitrate product. However, this is not the only method
used for solution concentration. Other options include: (1) a
concentrator with in-tank agitation, (2) two-stage evaporation
that gives a product with 7% moisture after the first stage and
4% to 5% moisture after the final stage, and (3) barometric con-
densers as an adjunct to vacuum evaporators to condense the water
removed from the exhaust stream. Concentration to 99.5% for high
density prills is achieved in a similar manner using different
operating parameters.
d. Particle Formation—
As shown in section 4 of Figure 6, particle formation includes the
prilling tower, various sizing screens, and the dryer and cooler.
Each of these areas is discussed below.
(1) Prilling tower—In the early days of the prilling industry,
tower design was accomplished primarily by trial and error.
Consequently the towers in use today show a great variation in
size, shape, and operating parameters. However, the basic operat-
ing principle of all such towers is the same. Ammonium nitrate
solution pumped to a head tank at the top of the tower maintains a
constant pressure on a spray device that sprays the solution into
13
-------
the tower. Droplets are formed and fall countercurrent to a
rising airstream. The airstream acts as a heat transfer mechanism
that cools the ammonium nitrate below its melting point and
permits solidification of the droplet into a spherical particle.
The first prilling tower constructed was a 6.1-m-square, 21-m-high
wooden unit. The tallest tower in use today is 61 m high and the
shortest recent design is 21 m high (4). Towers of circular,
square, or rectangular cross section are presently in operation.
Originally the solution was forced into the prilling tower by
spraying it upward through nozzles at an angle of 0.785 rad to the
horizontal. This arrangement is not essential, however. Good
results have since been obtained with the spray angled upward or
downward, or directed straight downward. Most (>90%) prilling
towers spray at a downward angle or straight downward to eliminate
much of the impingement problem (3). The actual form of the spray
device ranges from a single nozzle to multiple nozzles or a spray
bucket.
Ducts at the tower bottom supply an airflow countercurrent to the
drops, primarily as a temperature control. The air may be washed
and filtered prior to tower introduction; however, this is not
common practice since the additional moisture may inhibit drying
of the prills. The air moving upward cools the prills from an
inlet solution temperature of up to 186°C to an outlet prill
temperature of 75°C. The temperatures used in each plant vary
according to such factors as humidity, ambient temperature, and
type of product being made. As the drop falls, the air suddenly
chills its surface, causing a crust to form. One modification to
this design permits the drops to fall through a section of tower
containing a quiescent zone of heated air. Fifty percent of the
towers have such quiescent zones, ranging from 3 m to 8 m in
length. Airflow rates in prill towers are highly variable. An
average range of flow rates, however, is 6,960 to 11,600 std.
m3/metric ton. Air velocities in prill towers also cover a range
of values, from 1.37 to 2.45 m/s or higher depending on the tower
geometry and airflow rate.
The design of prilling towers is complex and many variables which
affect the tower performance must be. considered, including:
• Prill tower geometry
• Liquid inlet temperature
• Desired prill exit temperature
• Airflow rate
• Air inlet temperature
(4) Sharp, J. C. Nitrogen. In: Chemistry and Technology of
Fertilizers, V. Sauchelli, ed. Reinhold Publishing Corpora-
tion, New York, New York, 1960. pp. 26-32.
14
-------
• Air inlet humidity
• Quantity of material to be produced
• Desired prill size distribution
(2) Sizing screens—In the majority of the prilling plants (>90%),
two sets of screens (as shown in Figure 6) are used to regulate
the prill size. A 9.51-mm screen at the bottom of the prill tower
discharges to the drying system after it removes the oversize
material for recycle to the lump-dissolving tank. The 1.68-mm
screen after the cooler rejects material smaller than 0.55 mm.
These fines are also recycled to the lump-dissolving tank. Table 3
presents size distributions of the prills before and after screen-
ing; i.e., the final product.
TABLE 3. SIZE ANALYSIS OF PRILLED AMMONIUM NITRATE
Percentage of sample retained on screen
Screen
size
(standard sieve
mesh
designation)
2 .38 mm
1.68 mm
1.41 mm
250 ym
88 ym
74 ym
44 ym
<44 ym
TOTAL
(8)
(10)
(12)
(60)
(170)
(200)
(325)
(pan)
Unscreened material
High density3
b
Sample A
9.12
27.63
53.29
9.92
0.028
0.009
<0.001
<0.001
100.00
b
Sample B
8.29
33.38
41.43
16.87
0.015
0.015
0.007
<0.001
100.01
Low b
density
7.17
27.25
34.05
31.04
0.024
<0.001
<0.001
<0.001
99.98
b
Sample A
13.03
30.58
47.66
8.69
0.002
0.001
<0.001
<0.001
99.96
Screened material
High density
b
Sample B
5.73
34.79
44.58
14.88
0.018
0.009
<0.001
<0.001
100.01
C .
d
Sample C
2.24
13.15
32.73
51.58
0.040
0.070
0.070
0.070
99.95
LOW
density
10.29
40.66
36.76
12.23
0.059
<0.001
<0.001
<0.001
100.00
Sample characteristic of the material leaving the prill tower.
b
Samples obtained from a plant operating line.
Sample characteristic of product leaving the plant.
Sample obtained from local fertilizer dealer in a 22.7-kg bag. Note increase in fine size and
shift in screened quantities to finer size, possibly as a result of prill breakage caused by
additional handling.
(3) Drying—Dryers are used only in the production of low density
prills since the purpose of this operation is to drive off the
water trapped in the prill during solidification. The temperature
must be increased progressively as the moisture content declines
to avoid deterioration of the crust through moisture transfer and
subsequent caking. In most cases (>95%) this is achieved by a
two-stage dryer arrangement. Thus, the temperature in the pre-
dryer and dryer can be closely controlled to minimize prill
destruction.
A second reason for close temperature control is the change in
ammonium nitrate crystal structure with temperature. Prill
destruction can result from specific volume changes associated
15
-------
with the crystal transition points shown in Table 4 (5). The
temperature sensitivity of the crystal structure can be modified
by placing an additive in the solution before it enters the prill
tower. The most common additives in use today are magnesium oxide
and calcium oxide (3). Certain companies, however, have their own
additives based on proprietary formulations. The additive also
eliminates the diurnal cycle of size change when prilled ammonium
nitrate is stored in warm weather.
TABLE 4. POLYMORPHIC CHANGES OF AMMONIUM NITRATE (5)
Phase
reaction
Melt •«-
I «-
II •<-
III •*•
IV <-
II +
+ I
-> II
-»- III
-+ IV
-»- V
-> IV
Type of modification
Liquid-«--*-cubic
Cubic-*-->-tetragonal
Tetragonal-«">a-orthorhombic
a-Orthorhombic-«--*3-orthorhoinbic
6-Orthorhombic-«~»-pseudohexagonal
Tetragonal-*--*3-orthorhombic
Temp . , °C
169.6
125.4
84.4
32.4
-17.9
50.6
Rotary hot-air drum dryers are used in at least 80% of the ammonium
nitrate facilities requiring predryer and dryer installations.
These drums are equipped with lifting flights to tumble the prill
bed while warm air flows countercurrent to the material flow. Two
basic configurations are used for predyers, dryers, and coolers.
The first is a stacked arrangement in which the material exiting
from the predryer falls into the dryer, and that from the dryer
falls into the cooler. An alternate configuration carries the exit
material from one dryer to the other by a network of conveyor belts.
This arrangement offers greater flexibility for processing both
low density and high density prills (which require only a cooler)
on the same process line.
(4) Cooling--Two types of coolers are used in the ammonium nitrate
industry. Rotary drum coolers are used in most installations, but
a recently available fluidized bed cooler has been installed in
over 30 processes worldwide in the last 5 years. In either unit,
the air quality is carefully monitored to prevent excess humidity
from entering the cooler.
e. Product Preparation—
Section 5 of Figure 6 consists of a coating drum and any preshipment
packaging facilities required. Coating materials consist primarily
(5) Miller, P., and W. C. Saeman. Properties of Monocrystalline
Ammonium Nitrate Fertilizer. Industrial and Engineering
Chemistry, 40(1):154-160, 1948.
16
-------
of diatomaceous earth (3) , fatty amine-clay mixtures (6), or
limestone dust (7). They serve as water barriers for the hygro-
scopic prills. In the case of uncoated prills an additive is
placed in the spray solution, as previously described. The indus-
try is currently using in varying degrees several proprietary addi-
tives other than magnesium oxide and calcium oxide. Permalene-34,
a mixture of boric acid, diammonium phosphate, and aluminum sul-
fate, is marketed by Mississippi Chemical Corporation. The "NUCLO-
ADD" process, marketed by C & I Girdler, Inc., causes heterogeneous
nucleation by adding inert organics to the melt. Magnesium nitrate
and various other proprietary additives are also used. In addition
to their effect on temperature sensitivity, additives also reduce
the hygroscopicity of the material, but only to a limited extent.
Therefore, some products containing an additive may also be coated.
Over 90% of the high density material produced is shipped or
stored in bulk form. Bagging is performed periodically, but only
when the plant is in the center of its market area where small
users buy bagged product. Bulk product is loaded by conveyor
through a tube into train hopper cars or bulk trucks.
3. Other Processes
As shown earlier in Figure 3, approximatley 8% of the total solid
ammonium nitrate produced is grained or granulated. The two
products differ basically in crystal form. The grained particles
have longer fiber characteristics while the granules are more
spherical.
a. Graining—
The graining process employs feed streams and a neutralizer simi-
lar to those used in the prilling process. Following neutraliza-
tion, the ammonium nitrate solution is sent to an open-pan or
falling film evaporator. The open-pan evaporator is simply an
open pan equipped with steam coils. The falling film evaporator
is a vertical tube and shell heat exchanger with the 83% NH^NOs
solution entering on the tube side. Steam is used for heat on the
shell side and air passes countercurrently on the tube side,
facilitating evaporation.
The evaporators produce a 98% NH^NC-3 solution in batch operation.
The solution is discharged to kettles approximately 2.1 m in
diameter, equipped with large plows to keep the material stirred
(6) Slack, A. V. Fertilizer Developments and Trends - 1968.
Noyes Development Corporation, Park Ridge, New Jersey, 1968.
405 pp.
(7) Ammonium Nitrate. In: Riegel's Handbook of Industrial Chem-
istry, J. A. Kent, ed. Van Nostrand Reinhold Company, New
York, New York, 1974. pp. 100-101.
17
-------
as it cools and solidifies. Small amounts of stearic acid or
other material may be added in the graining kettles to minimize
the formation of undesired large grains. The vessel jacket is
equipped with steam and cooling coils to control the cooling rate.
Temperatures used in graining can reach 152°C to 155°C. The final
product is a small grain between 0.55 mm and 1.68 mm, which is
screened, mixed with clay, and bagged.
b. Granulation—
(1) Stengel process—The Stengel process is used by very few
plants in the industry on only about 2% of the total ammonium
nitrate production.
In this process the ammonia vapor and nitric acid (58%) are pre-
heated in separate heat exchangers and fed simultaneously and
continuously to the Stengel reactor shown in Figure 7 (8, 9).
The heat of reaction vaporizes the water in the nitric acid feed.
A solution of 98% ammonium nitrate and trace ammonia flows from
the reactor to a centrifugal separator which removes the ammonium
nitrate and runs it through an air stripper that reduces the mois-
ture content to approximatley 0.2% (10) .
AMMONIA GAS
STEAM
NITRIC ACID
MOLTEN
AMMONIUM
NITRATE
Figure 7. Stengel reactor (9).
(8) Hester, A. S., J. J. Dorsey, Jr., and J. T. Kaufman. Stengel
Process Ammonium Nitrate. Industrial and Engineering Chemis-
try, 46(4):622-632, 1954.
(9) Stengel, L. A. Process for Producing Ammonium Nitrate.
U.S. Patent 2,568,901 (to Commercial Solvents Corp.),
September 25, 1951.
(10) Dorsey, J. J. Ammonium Nitrate by the Stengel Process.
Industrial and Engineering Chemistry, 47(1):11-17, 1955.
18
-------
The molten ammonium nitrate flows into a weir box for distribution
onto an endless, stainless steel, water-cooled (Sandvik) belt.
Solid ammonium nitrate is removed from the belt by a doctor blade.
This material is fed to the grinders, then screened, coated, and
bagged. The oversize material is recycled to the grinder and the
fines are returned to the process. Particle formation may also be
accomplished using a prill tower, as described previously.
(2) Spheroidization—In this process, a 99.5% ammonium nitrate
solution from a neutralizer and evaporator/concentrator, as previ-
ously described, is sprayed into a rotating drum containing
ammonium nitrate granules. The drum is equipped with lifting
flights that lift the granules and permit them to fall through the
open space in the drum. • The melt is applied to the granules in
midair and in the bed. Ambient temperature is chilled to 7°C and
passed countercurrently to cool the melt and solidify the granules.
The particles exit the drum at <0.5% moisture which minimizes the
amount of drying. Then the product is screened, cooled, coated,
and bagged.
Solid particle formation by graining, granulation, and spheroidi-
zation will not be considered further in this assessment for the
following reasons: 1) since emissions from these processes are
lower than those from prilling, a consideration of high density
prilling will give a maximum value of emissions; 2) these three
processes account for less than 8% of total ammonium nitrate pro-
duction; and 3) each of these processes is operated inside a
building, and their only emissions are fugitive.
B. MATERIALS FLOW
A materials flow diagram of the prilling process is shown in
Figure 8 based on the process flow diagram shown in Figure 6. As
noted previously the dryer section (including the predryer and
dryer) is not present in the high density prilling process. In
this case streams 11 and 12 would be absent and stream 13 would be
equivalent to stream 10. Table 5 lists approximate flow rates and
compositions of each stream in Figure 8 for the production of low
density ammonium nitrate using stoichiometric quantities of both
reactants. Table 6 lists approximate flow rates and compositions
for each applicable stream in Figure 8 for the production of high
density ammonium nitrate using stoichiometric quantities of both
reactants.
In both Tables 5 and 6 a coating application is assumed. If no
coating is actually done, the final product weight is found in
stream 17. The basic assumptions used to calculate these values
are as follows:
19
-------
1 ©
DRYER
©
COOLER
FLOW RATE AND COMPOSITIONS
OF NUMBERED STREAMS ARE
PROVIDED IN TABLES 5&6
BAGGING
Figure 8. Materials flow diagram - prilled ammonium nitrate,
• Nitric acid feed is 55% nitric acid (3).
• All steam produced as a result of the heat of
reaction exits from the neutralizer stack and
does not condense and return to the neutralizer.
• Any excess reactants are negligible.
• Recycle stream 4 contains only ammonium
nitrate and water.
• Any impurities present are negligible.
• Airflows to the dryer and cooler are given the
values of X and Y since no change in weight through
the equipment is expected.
• There is little or no moisture removal in the
prill tower (3).
• Approximately 0.5% of the material produced by the
prill tower is lost through the stack.
20
-------
TABLE 5.
MATERIAL BALANCE FOR LOW DENSITY PRILLING PROCESS
(kg/hr)
Stream
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Name
Ammonia feed
Nitric acid feed
Concentrator feed
Recycle
Concentrator exhaust
Prill tower feed
Prill tower exhaust
Air for prill tower
Wet product
Dryer feed
Air to dryer
Dryer exhaust
Cooler feed
Air to cooler
Cooler exhaust
Dry product
Coating feed
Coating material
Final product
Fines recycle
Lump recycle
Neutralizer exhaust
Makeup solution0
Purge stream"1
Ammonia
1,134
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
"
Nitric
acid
0
4,202
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Hater
0
3,437
1,358
239
1,060
298
0
0
298
296
0
198
98
0
98
0
0
0
0
0
2
2,318
237
Ammonium
nitrate
0
0
5,668
359
0
5,668
30
0
5,638
5,610
0
56
5,554
0
55
5,499
5,471
0
5,471
28
28
27
303
Aira
0
0
0
0
0
0
14.9 x ID*
14.9 x 1011
0
°b
Xb
x°
°b
Yb
y"
0
0
0
0
0
0
0
0
Clay
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
170
170
0
0
0
0
aAirflows to the dryer and cooler are given the values of X and Y since no change in
weight through the equipment is expected.
Quantities of air feed to dryer and cooler not available.
CMakeup solution may be water or weak (-^60%) ammonium nitrate solution or a combination
of the two.
Composition of stream determined by process requirements and contamination present in
recycle vessel.
TABLE 6.
MATERIAL BALANCE FOR HIGH DENSITY PRILLING PROCESS
(kg/hr)
Stream
1
2
3
4
5
6
7
8
9
13
14
15
16
17
18
19
20
21
22
23
24
Name
Ammonia feed
Nitric acid feed
Concentrator feed
Recycle
Concentrator exhaust
Prill tower feed
Prill tower exhaust
Air for prill tower
Wet product
Cooler feed
Air to cooler
Cooler exhaust
Dry product
Coating feed
Coating material
Final product
Fines recycle
Lump recycle
Neutralizer exhaust
Makeup solution0
Purge streamd
Ammonia
1,134
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
™
Nitric
acid
0
4,202
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
*"
Water
0
3,437
1,358
239
1,330
28
0
0
28
27
0
27
0.
0
0
0
0
1
2,318
238
"
Ammonium
nitrate
0
0
5,668
359
0
5,668
28
0
5,640
5,613
0
56
5,557
5,529
0
5,529
28
27
27
304
"
Aira
0
0
0
0
0
0
14.9 x 10"
14.9 x 10"
0
°b
Yb
YD
0
0
0
0
0
0
0
0
"
Clay
0
0
0
0
0
0
0
0
0
0
0
0
0
0
171
171
0
0
0
0
Airflow to the cooler is given the value of Y since no change in weight through the
equipment is expected.
Quantity of air feed to the cooler not available.
CMakeup solution may be water or weak (^SO*) ammonium nitrate solution or a combination
of the two.
Composition of stream determined by process requirements and contamination present in
recycle vessel.
21
-------
• Approximately 0.5% of the material produced by the
prill tower is recycled due to oversize (3).
• The ammonium nitrate losses from the dryer and
cooler are 1% each based on the exit material
from each unit.
• All water is evaporated by the time the product
leaves the dryer.
• Approximately 0.5% of the material produced by the
cooler is recycled because of its fine size (3) .
• There is no loss from the coating drum.
• Coating application is approximately 3% of the
final product weight.
Operating conditions are highly related to temperature. Most
(>90%) of the neutralization is done at atmospheric pressure or
with only a slight hydrostatic head. All of the other operations
are performed at atmospheric pressure; therefore, temperature is
the prime control.
In the neutralizer the temperature is maintained at 131°C by re-
lease of steam generated from the heat of reaction (108.8kJ) and
the addition of recycle liquor. Temperature rises slightly in the
evaporator to ^140°C; this can vary depending on the ammonium
nitrate concentration of the exiting solution. In the prilling
tower, the temperature of the prill drops from 140°C to 75°C thus
placing the solution well below the solidification temperature of
123°C to 125°C (for a 95% ammonium nitrate solution).
Temperatures in the drying section (where applicable) and cooler
are carefully monitored. Prill destruction could result from
changes in specific volume associated with a phase change between
tetragonal and a-orthorhombic forms if the temperature rises above
84.4°C, and between a-orthorhombic and B-orthorhombic forms if the
temperature falls below 32.4°C. Therefore the temperature from
the predryer through the cooler is kept between these two limits.
C. AVERAGE PLANT PARAMETERS
Although the ammonium nitrate industry manufactures a wide range
of products/ certain average plant parameters can be defined. An
average plant has a neutralizer solution capacity of 131,500 metric
tons/yr and is located in a county having a population density of
209 persons/km2. (See Appendix A for a complete list of plant
capacities and locations.) As shown in Table 1, there is no repre-
sentative product. However, the average high density prilling
operation has a capacity of 95,000 metric tons/yr, and the aver-
age low density prilling operation has a capacity of 62,700 metric
tons/yr. Low density product is coated, but <3% of the high den-
sity material is similarly treated.
22
-------
Feedstock to the process in the average plant consists of a super-
heated ammonia stream at 70°C and a preheated nitric acid stream
at 82°C. These reactants combine in a neutralizer to produce an
83% NHi^NOs stream which passes through a falling film evaporator
for concentration to 99.5+% NHifNO3 to make high density product,
or to 95% for low density product, and then passes into a prilling
tower. The prills encounter a quiet zone 3 m long, then fall
countercurrent to a cooling airstream for 34 m and exit through a
cooler for bulk storage or direct loading into railroad hopper
cars or bulk trucks. Scrubbers on the exhaust streams from the
cooler, and from the drying section in the case of low density
product, are the only controls used. Such scrubbers (to be
described in Section V) have an efficiency of 97% to 98% (weight)
and produce no visible emissions. A flow diagram has been pre-
sented in Figure 6.
Average stack heights (derived from the raw data presented in
Table A-4) are as follows:
Neutralizer 15.2 m ± 30%
Evaporator/concentrator 13.7 m ± 10%
Prill tower 41.2 m ± 18%
Dryer 13.7 m ± 15% (if present)
Cooler 13.2 m ± 15%
Loading car 3.3 m ± 5%
Other factors that interact and affect the operation of ammonium
nitrate plants and the resulting emission rates include but are
not restricted to:
• Prill tower geometry
• Temperature and moisture content of the various
process streams
• Amount of coating required
• Airflow rates through the different processes
• Amount and species of reactant in excess
• Amount of hydrostatic head in neutralizer and tower
• Ambient temperature and relative humidity at plant site
• Type of spraying device
• Location of individual plant's market
• Requirements of individual plant's market
• Desired prill size distribution
All of the above variables can change the operating and physical
characteristics of a plant without changing the basic flow diagram.
Loading car stack height is determined by approximating height of
a standard railroad car used in transporting this material.
23
-------
SECTION IV
EMISSIONS
The manufacture of ammonium nitrate can emit particulates (NH4N03),
ammonia, and nitric acid mist. Particulates are released from the
neutralizer, evaporator/concentrator, prilling tower, drying sys-
tem, and cooler. Ammonia or nitric acid mist may be released from
the neutralizer and the evaporator/concentrator, depending upon
which reactant is present in excess. The chemical species (NHs or
nitric acid mist) and quantity emitted are primarily dependent
upon the individual unit operation and conditions. The majority
(>75%) of ammonium nitrate plants operate under basic pH condi-
tions (NHs excess). The quantity of ammonia emitted varies from
plant to plant since it is directly related to the amount of
ammonia in excess.
A. EMISSIONS FROM EACH PROCESS
In Figure 8, streams 5, 7, 12, 15, and 22 are shown as emission
points from the ammonium nitrate process. These streams are the
exhausts from the evaporator/concentrator, prilling tower, dryer,
cooler, and neutralizer, respectively. In addition, fugitive emis-
sions are released from screening, coating, bagging, and bulk
loading operations. Each of these operations occurs within a
building, and any emissions would exhaust from doors and windows
rather than through a stack.
The following sections characterize the species emitted from each
process and determine average_emission factors. Average maximum
ground level concentrations (Xmax^ are calculated f°r emissions
from each point, based on average emission factors and process
parameters; i.e., production rates and stack heights. In addition,
the source severity, S, defined as the ratio of Xmax to the
ambient air quality standard, AAQS, is determined for emissions
from each operation. Where no air quality standard exists (NHa,
HN03), a reduced TLV is used in place of the AAQS; i.e., TLV x
8/24 x 1/100. The methodology for computing Xmax and S is Pre~
sented in Appendix B.
Percent uncertainty was established by applying a "Student t" test
to the input data. The "t" test involves the estimation of the
true average value of a sample population and the establishment of
confidence ranges within which the true average value likely
24
-------
exists (11). The "t" test is applied to the report data because
the sample size is less than 30 and may not be normally distributed.
1. Neutralizer Operations
The exhaust gas from the neutralizer is a jet of superheated steam
containing ammonia and particulates; i.e., droplets of recombined
NH3 and HN03 molecules liberated by dissociation. The steam acts
as a carrier gas for the ammonia and particulates. Ammonia is
released rather than HNO3 because NH3 is used as the excess
reactant in most (>75%) of the plants.
a. Ammonia—
Data on ammonia emissions from the neutralizer have been obtained
from three distinct sources. One source indicated that plants
(neutralizers) operating on the basic or high pH side have lost as
much as 4 kg of ammonia per metric ton of nitric acid consumed (3).
For an average plant, 1.04 x 105 metric tons of acid are consumed
per year, resulting in an anticipated ammonia loss of 4.14 x 105 metric
metric tons/yr. For continuous operation this results in an
emission rate of 13.1 g/s and an emission factor of 3.14 g of NH3
per kg of NH^NOs produced. The concomitant Xmax an<^ source
severity, derived using Equation 5 in Appendix B, are 1.0 mg/m3
and 17.2 respectively.
A second method of estimating the ammonia released by the neutral-
izer is to approximate the quantity of ammonia contained in the
neutralizer offgas by pH. The neutralizer offgas is normally held
between a pH of 9.5 and 11 as discussed previously. This is the
pH of a condensed sample of the offgas at 83°C. An ammonia solu-
tion with a pH of 10.6 is -vO.OlN and contains 0.17 g NH3/1, or
170 g NH3/m3. The total amount of steam leaving the neutralizer
as a result of heat of reaction is calculated from a material
balance to be 3.45 m3/s for an average plant. Another author has
reported a lower steam rate of 1.62 m3/s (12), presumably because
part of the steam condensed before leaving the neutralizer stack.
(See Appendix C for calculations.) If the total amount of steam
leaving the neutralizer were condensed, it would equal 0.85 kg to
1.82 kg of water per second. Combining this value with the ammo-
nia concentration of 170 g/m3, the emission rates (Q) of 0.15 and
0.32 g/s are determined. The emission factor (E), maximum ground
level concentration (Xmax^ • anc* source severity (S) are listed in
Table 7 for the two steam rate conditions discussed.
(11) Volk, W. Applied Statistics for Engineers, Second Edition.
McGraw-Hill Book Co., New York, New York, 1969. 354 pp.
(12) Metzger, T. R. Controlling Airborne Emissions from Ammonium
Nitrate Production. Presented to Ammonium Nitrate Study
Group, Sarnia, Ontario, August 1974. 7 pp.
25
-------
TABLE 7. EMISSION PARAMETERS FOR AMMONIA FROM NEUTRALIZER
EXHAUST BASED ON APPROXIMATION OF pH
Parameter
Emission rate,
Emission factor
Xmax' ugA>3
Source severity
g/s
, g/kg
1.62
0
0
12
0
Steam
m3/s
.15
.04
.1
.2
rate
3.45
0
0
25
0
m3/s
.32
.08
.8
.4
The third method of determining ammonia emissions is by source
testing. The tests conducted found concentrations between 100 ppm
and 3,000 ppm NH3 in the neutralizer exhaust, which are equivalent
to 0.072 g/m3 and 2.16 g/m3, respectively. These yield the emis-
sion parameters shown in Table 8.
TABLE 8. EMISSION PARAMETERS FOR AMMONIA FROM NEU-
TRALIZER EXHAUST BASED ON SOURCE TESTING
Parameter
NH3 concentration in
neutralizer exhaust, ppm
Emission factor, g/kg
Emission rate, g/s
X yg/m3
Source severity
n
100
0.026
0.1
8.9
0.1
62 n
to
to
to
to
to
3
0
3
Steam
/s
,000
.86
.5
280
4
.6
rate
3.
100
0.055
0.2
19
0.2
45 ii
to
to
to
to
to
3
1
7
/s
,000
.8
.5
580
9
.8
Two factors need to be considered in interpreting these data.
Ammonia was initially used as the excess reactant because it was
inexpensive and it also produced a significantly less visible
plume. (Excess ammonia vaporizes and inhibits the dissociation
reaction, NHi+NO3 -»• NH3 + HN03; the recombination of NH3 and HN03
vapor is the main source of particulates.) While the economic
aspect is no longer true, plants without controls still tend to
run on the basic side because of the lower plume opacity obtained,
Secondly, those plants operating on the basic side tend to run
with an appreciable excess of ammonia. This premise has been
26
-------
confirmed in cases where testing was done and measurements showed
3,000 ppm of NH3 in the exhaust gas. In each case, the company
involved did not realize that it was losing so much ammonia and
took corrective measures to lower the ammonia emissions.
It should be noted that the operation of the neutralizer is a
relatively subjective art rather than a science. Therefore, the
emissions variation resulting from and directly dependent upon the
plant operations can be significant from plant to plant and from
operator to operator. For example, one source test indicated a
range on the same neutralizer of 150 ppm to 3,000 ppm of NH3 in
the offgas depending upon when the sample was taken (results
varied both from shift to shift and within individual shifts) .
b. Particulates —
Particulate emissions from the neutralizer consist of ammonium
nitrate particles. Trace metal emissions resulting from the corro-
sion of the neutralizer were discounted since, even at parts per
billion concentrations, the quantity leaving the neutralizer would
be sufficient to cause equipment failure. The ammonium nitrate
particulate emission was determined by two methods: source test-
ing and dissociation theory. Some source testing has been com-
pleted and the results are listed in Appendix A. From these data,
an emission factor of 1.64 g/kg ± 84% can be Calculated. For the
average plant, this results in Q = 6.85 g/s, Xmax = 53^ V3/m3 •
and S = 2.07.
The second method for estimating particulate emissions utilizes
dissociation theory and assumes that all of the material disso-
ciated will end up as recombined droplets leaving the stack.
While this method cannot be used to supersede test data, it is a
valuable verification tool.
The vapor pressure (p, in cm of Hg) of ammonium nitrate can be
estimated using the following equation (13) :
+ 8.502 (2)
where T = absolute temperature, K
Assuming ideality and' that steam and vapors are in equilibrium,
emission rates of 3.14 g/s and 1.48 g/s are calculated for steam
rates of 1.62 m3/s and 3.45 m3/s, respectively. This results in
the following values in Table 9 for emission factor, maximum
ground level concentration and source severity.
(13) Feick, G. , and R. M. Hainer. On the Thermal Decomposition of
Ammonium Nitrate. Steady-State Reaction Temperatures and
Reaction Rate. Journal of the American Chemical Society,
76(23):5860-5863, 1954.
27
-------
If the two calculated theoretical extremes are averaged with the
source test data, an emission factor of 1.43 g/kg ± 76% can be
obtained, which results in a source severity of 1.76. While this
average cannot be taken as a true value, it demonstrates the
excellent agreement of theory to actual data available.
TABLE 9. EMISSION PARAMETERS FOR PARTICULATES FROM
NEUTRALIZER EXHAUST BASED ON DISSOCIATION/
RECOMBINATION THEORY
Parameter
Emission factor, g/kg
Xmax' lig/m
Source severity
Value
0.36 to 0.77
120 to 260
0.45 to 0.96
c. Nitric Acid—
Nitric acid is not normally added in excess to the neutralizer
because it results in higher particulate emissions which in turn
make the plume more visible. (Nitric acid is not vaporized as
readily as ammonia and therefore does not suppress the dissocia-
tion of ammonium nitrate.) Consequently, there are no source test
data available on the magnitude of HNOs emissions. The maximum
level can be determined from a material balance, based on the fact
that plants operating with excess acid use only 0.02% to 0.05% in
excess. The emission factor for an average plant is <0.26 g/kg
and the source severity (using TLV = 5 mg/m3 and h = 14 m) is <5.8.
This calculation must be viewed as the maximum for the whole
operation rather than just for the neutralizer since some of the
excess HN03 may be released beyond the neutralizer.
2. Evaporator/Concentrator Operation
a. Particulates—
Few data are available on particulate emissions from evaporator/
concentrators primarily due to the wide variation in the types of
evaporators employed. One can theoretically calculate the amount
of particulates liberated as a result of dissociation and recom-
bination since the operation is basically one of vaporization.
The theory as proposed for neutralizers is valid in this case
also. However, for these calculations, the amount of steam
released can only be approximated by a material balance. For an
average plant, ^1.5 m3/s of steam is released. With a vapor
pressure of 56.2 Pa (5.5 x 10~k atm), as determined by Equation 2,
and a temperature of 143°C, an emission rate of 1.96 g/s is
derived. Table 10 lists the resultant values for emission factor,
maximum ground level concentration, and source severity.
28
-------
TABLE 10. EMISSION PARAMETERS FOR PARTICULATES FROM
EVAPORATOR/CONCENTRATOR BASED ON DISSOCIATION/
RECOMBINATION THEORY
Parameter Value
Emission factor, g/kg 0.47
Xmax' ^/m3 190
Source severity 0.7
The variations in operation of the evaporator/concentrator are
similar to those present in the neutralizer. Another variation in
the evaporator/concentrator that is not so predominant in the
neutralizer is equipment design. As previously mentioned, ^75% of
the industry utilizes vacuum, film-type evaporators. The varia-
tion possible in the design of these evaporators is large. Other
variation possibilities include the use of in-tank agitation,
two-stage evaporation, or barometric condensers as an adjunct to
vacuum evaporators. In the last case, emissions do not reach the
atmosphere directly, but pass into the cooling tower. Therefore,
sampling of the evaporator/concentrator to obtain data characteris-
tic of the industry is difficult, if not impossible.
The source test results for evaporator/concentrators shown in
Appendix A are a factor of ten lower than this theory predicts.
However, a review of industry comments showed that the value pre-
dicted by theory is closer to average plant operations. Therefore,
this value is used in further calculations as a maximum level of
particulate emissions from the evaporator/concentrator.
b. Ammonia--
Ammonia should not be emitted from the evaporator/concentrator in
any appreciable quantities, because the feed to the evaporator has
a pH of ^5.2 to 5.4. This low pH is the result of ammonia vapor-
ization in the neutralizer. In addition, no ammonia odor was
detected during field investigations.
3. Prilling Tower Operation
a. Particulates—
More emission tests have been run on the prilling tower than on
any of the other potential emission points. Data obtained from
these source tests are presented in Table A-2. In adddition, data
were received on the emissions from a high density prill tower
during experimental studies where the operating conditions were
varied in an attempt to identify relationships between certain
process variables and emissions. These data are presented in
29
-------
Table A-3. While no significant correlations were identified,
many data were obtained on the particulate emissions for various
melt pH's, spray temperatures, airflow rates, and additives in the
ranges given in Table 11.
TABLE 11. OPERATING CONDITIONS USED TO OBTAIN
DATA ON PARTICULATE EMISSIONS FROM
PRILLING TOWER OPERATIONS
Process condition Range of values
Spray pH 5 . 4 to 7 . 0
Spray temperature 171 to 186°C
Tower airflow 2,209 to 3,880 standard m3/min
Additive addition 0.10 to 0.14 wt. % additive in melt
The conditions listed above were varied in a range characteristic
of those found throughout the industry. Consequently, the data
obtained are similar to those which would be expected from differ-
ent high density prill towers within the industry. By including
these data with those presented in Table A-2, the emission parame-
ters shown in Table 12 were calculated for an average high density
prilling operation; i.e., one having an annual capacity of 95,000
metric tons/yr.
TABLE 12. EMISSION PARAMETERS FOR PARTICULATES
FROM PRILLING TOWER
Parameter Value
Emission factor, g/kg 1.37 ± 10.0%
E_mission rate, g/s 4.1
Source severity 0.17
b. Ammonia—
Ammonia emissions from the dissociation in the melt at the top of
the prill tower are insignificant. During field investigations,
ammonia odors were not detectable at the top of the prill tower.
The odor threshold of ammonia is reported to be 20 ppm to 53 ppm
(14, 15) which corresponds to a source severity range of 0.09
(14) Ammonia. In: The Merck Index, P. G. Stecher, ed. Merck and
Company, Inc., Rahway, New Jersey, 1968. p. 64.
(15) Matheson Gas Data Book. The Matheson Company, Inc., East
Rutherford, New Jersey, 1966. pp. 13-22.
30
-------
to 0.4 based on an average prill tower height of 41.2 m and an
airflow rate of 2,209 to 3,880 m3/min. If residual ammonia is
estimated to be ^1% of the particulate emissions (i.e., assuming
that 99% of the NH^NOa that dissociates in the prill tower recom-
bines), the source severity is <0.01. Under certain conditions,
small quantities of ammonia are added to the melt to facilitate
prilling. This procedure may raise the ammonia emissions slightly,
but not above the odor threshold as indicated.
4. Cooler and Dryer Operation
Few test data are available on particulate emissions from coolers
and dryers because these operations are routinely controlled with
high efficiency scrubbers that give no visible plume. The only
test measurement reported gives an uncontrolled emission factor of
5.13 g/kg for a cooler (16). This value agrees with others
reported for uncontrolled dryers in the fertilizer industry (as
well as in other mineral industries), which generally range
between 1 g/kg and 10 g/kg (16).
Standard rotary dryers and coolers are usually controlled by high
efficiency wet scrubbers with anticipated efficiencies of >99% for
the removal of particles above 10 ym. There are several plants
(<10%) that utilize lower efficiency (85%) scrubbers; however, the
high efficiency scrubbers dominate. Assuming a conservative effi-
ciency estimate of 99% for the previously reported uncontrolled
emission factor (5.13 g/kg) results in a controlled emission fac-
tor of 0.05 g/kg and a source severity of 0.06 for the average
cooling process.
The above emission factor can also be applied to the dryers used
in the production of low density prills. In this case, the source
severity is 0.04.
5. Fugitive Emissions
In addition to the stack emission points discussed above, there
are two potential sources of nonstack or fugitive-type emissions:
the coating operations and the bulk loading operations. No partic-
ulate emissions could be seen during plant visits; however, no
actual test data are available. A worst case emission factor is
therefore derived from material balance considerations.
(16) Point Source Listings. National Emissions Data System.
U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, May 1974.
31
-------
a. Coating Operations—
All low density prills and ^3% of all high density prills are
coated with one of several coating materials. At an average low
density prilling operation, therefore, 1.99 kg/s of prilled mater-
ial would be coated. With an average coating rate of 3%, 0.06 kg/s
of coating would be applied to the prills.
If all the coating material is lost to the atmosphere through
doors and windows (a worst case assumption), an emission rate of
60 g/s results. By employing the relevant equations for emissions
from buildings (developed in Appendix B), the following relation-
ships are derived:
*max = 63-° D"1-81lt <3>
and
S = 2.42 x 105 D-1-814 (4)
A graph of the source severity equation is shown in Figure 9. For
the source severity to be <1.0, the distance from the building to
the plant boundary must be more than 929 m. The effect of the
building cross-sectional area on this distance is minimal. For
comparison the average plant is situated a distance of 800 m from
the nearest plant boundary. (Plants have been built in remote
areas because of explosive hazards.)
In this case, a worst case assumption is unrealistic because the
coating material must adhere to the prills to be effective. Indus-
trial contacts involved in coating operations indicate that a loss
of 10% of the coating material is a liberal estimate, and most of
this falls to the floor of the coating building. For this reason,
an emission rate of <6 g/s is an upper limit for actual coating
operations. The impact of this change is demonstrated by the
second line drawn on Figure 9 using a 6 g/s emission rate. The
distance from the source where the source severity is <1.0 is then
261 m.
b. Bulk Loading Operations—
Table 3 indicated the size distribution of the final product
exiting the ammonium nitrate process. As can be seen, <0.001% of
the material on the belt leaving the plant is <44 ym. From Stokes
law it can be calculated that the distance required for 44 ym
particulates to settle is 7.1 m, based on a wind speed of 4.5 m/s
and a loading height of 6.1 m. This is well inside the typical
plant boundary, so that emissions that remain airborne must have
an emission factor of <0.01 g/kg.
32
-------
100
10,000
B.
Figure 9. Source severity versus distance
for the release of emissions from
confinement for coating operation.
LOW DENSITY PRILLING
Data in previous sections described plants making ammonium nitrate
solutions and high density prills. This sections presents a brief
consideration of strictly low density prilling. Section 4 of Fig-
ure 6 is a flow diagram for a low density prilling operation. A
careful inspection of this schematic reveals the same emission
points as those of a high density operation, but with the inclu-
sion of a dryer or a drying system consisting of a predryer and
dryer in series.
Source test data obtained at a low density ammonium nitrate facil-
ity are given in Table 13. Each emission factor is an average
based on two to four tests per unit. Note that only particulate
emissions were measured. (Ammonia emissions will be proportional
to the amount of excess ammonia added to the neutralizer.) Aver-
age ground level concentrations and source severities are also
given in Table 13, and it can be seen that these are lower than
those for high density prill production.
33
-------
TABLE 13. PARTICULATE EMISSION PARAMETERS FOR
LOW DENSITY PRILLING OPERATION3
Unit
Neutralizer .
Evaporator/concentrator
Prilling tower b
Predryer
Dryer^
Cooler
Emission
factor,
g/kg
0.045
0.088
0.496
0.015
0.009
0.016
Emission
rate ,
g/s
0.089
0.17
0.987
0.030
0.018
0.032
yg/m3
7.0
17.0
10.6
2.9
1.7
3.3
S
0.03
0.06
0.04
0.01
0.01
0.01
Emission rate, Xmax/ and source severity values were calcu-
lated using the capacity and stack heights for an average low
density prilling operation; i.e., 62,700 metric tons/yr.
Uncontrolled.
cControlled.
It is believed that the lower emission factors for low density
prill production result from lower operation temperatures in the
manufacturing process. In the manufacture of high density prills,
higher temperatures are required to keep the concentrated ammonium
nitrate solutions in the liquid state. Particulates are generated
by the dissociation and subsequent recombination of NHifNO3 mole-
cules at elevated temperatures. Lower operating temperatures
therefore reduce particulate emission. If a low density produc-
tion line was maintained at the same temperature used for high
density production, its particulate emissions would increase to
the levels observed for high density prilling.
C. POTENTIAL ENVIRONMENTAL EFFECT
There are several ways in which the potential environmental effect
of emissions from the production of ammonium nitrate can be evalu-
ated: 1) by the magnitudes of Xmax an<^ source severity, S; 2) by
the number of persons in the vicinity of a plant who may be
affected by an emission; and 3) by the amount of state and
national industry emissions.
Source severity has been defined previously as the ratio of Xmax
to the appropriate ambient air quality standard (260 yg/m3 in the
case of particulates) or to a reduced TLV (18 mg/m3 x 8/24 x 1/100
in the case of ammonia). It indicates the relative environmental
severity of different air emissions.
Table ^4 summarizes the particulate emission factors, emission
rate, Xmax values, and source severities for the average
processes used in making solutions and high density prills.
34
-------
TABLE 14. PARTICULATE EMISSION PARAMETERS FOR SOLUTION
AND HIGH DENSITY PRILLING OPERATIONS
Emission point
Emission
characteristic
Emission factor, g/kg
Emission rate, g/s
X f pg/m3
Amax ^'
Source severity
Neutralizer
1.64a
6.85
539
2.07
Evaporator/
concentrator
0.47a
1.96
190
0.70
Prilling
tower
1.373
4.13
44
0.17
Cooler
0.05b
0.15
16
0.06
Affected population,
persons
95
a b
Uncontrolled. Controlled.
Values are for uncontrolled emissions except in the case of the
cooler that is controlled with a scrubber as discussed previously.
The corresponding values for ammonia emissions from the neutral-
izer are as follows:
• Emission factor: 0.026 to 3.14 g/kg
• Emission rate: 0.1 to 13.1 g/s
' Xmax: 8-9 to If030 yg/m3
• Source severity: 0.1 to 17.2
The higher severity for ammonia (17.2) corresponds to 1.5% excess
ammonia in the feed stream. Plants do not operate this way inten-
tionally because of the economic loss, and a well-monitored opera-
tion will exhibit a much lower source severity; i.e., below 1.0.
The data show that in ammonium nitrate manufacture the highest
severities result from ammonia and particulate emissions from the
neutralizer.
The impact of coating operations can be determined by evaluating
the source severity at the plant boundary. Since an average plant
has a boundary at least 800 m from the source, the source severity
for coating is 0.14.
In addition to severities for average process operations, it is
possible to calculate the distribution of severities for the whole
industry. Figures 10 through 13 show source severity distribu-
tions for operations used in the manufacture of solutions and high
density prills. The severity distribution curve for low density
evaporator/concentrators is illustrated in Figure 14. Other low
density operations are not shown because severities are <0.1
(although one prill tower has a severity of 0.11). Distributions
were calculated from average emission factors and stack heights
for each type of operation and the distribution of production
35
-------
0 LO 2.0 3.0 4.0 5.0 6.0 7.0
SOURCE SEVERITY
Figure 10.
Source severity distribution for neutralizers used
in making solutions and high density prills.
0 0.25 0.5 0.75 1.0 1.25 1.5 1.75 2.0 2.25 2.5
SOURCE SEVERITY
Figure 11. Source severity distribution for evaporator/concentra-
tors used in making solutions for high density prills,
100
80
60
° 40
i— «/> MU
6s
o o
£~ 20
0
Figure 12.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
SOURCE SEVERITY
Source severity distribution for
high density prilling towers.
36
-------
100
£^ 80
£3 60
3y
l^ ^5
° ° 40
i in "HI
II 20
0.05
0.1 0.15
SOURCE SEVERITY
0.2
0.25
Figure 13. Source severity distribution for
high density cooling drums.
0.1
SOURCE SEVERITY
0.15
0.20
Figure 14.
Source severity distribution for the evaporator/
concentrator in low density prilling processes.
37
-------
capacities for each type of product. (Production capacities were
obtained from a confidential Monsanto Company survey.) The per-
centage of plants having severities in excess of 1.0 and 0.1 for
each operation are given in Table 15.
TABLE 15. PERCENT OF PLANTS EXCEEDING SPECIFIED
VALUES OF SOURCE SEVERITY, S
Percent of plants
For S = 0.1 For S = 1.0
Solutions and high density prills
Neutralizer 100 63
Evaporator/concentrator 95 18
Prilling tower 56 0
Cooler 22 0
Low density prills
Neutralizer 0 0
Evaporator/concentrator 18 0
Prilling tower 5 0
Finishing section 0 0
Another measure of potential environmental effect, the affected
population for each pollutant, is also shown in Table 14. The
affected population is defined as the number of persons who reside
in the area around an average process where the source severity
is >1.0. The equations and rationale for the calculation of the
affected population are given in Appendix D. For ammonia emis-
sions from the neutralizer, the affected population ranges from
0 to 1,058.
Whereas the source severity and affected population indicate poten-
tial effects of emissions from ammonium nitrate manufacturing on a
local scale, the national and state emission burdens measure the
impact of emissions on a whole area. The particulate emission
burden is defined as the ratio of particulates emitted from all
ammonium nitrate production statewide or nationwide to the total
particulates emitted from all stationary sources in the state or
nation. Table 16 is a compilation of particulate emission burdens
for each state that produces ammonium nitrate. Calculations were
based on state ammonium nitrate capacities as given in Appendix A
and the average emission factors in Tables 13 and 14. It was
assumed that 39% of the capacity in each state was for a liquid
product and 61% for a solid product. Solids were assumed to be
60% high density prills and 40% low density prills. Fugitive emis-
sions (e.g., coating) were not considered in the totals. Total
38
-------
TABLE 16. PARTICULATE EMISSIONS FROM AMMONIUM NITRATE PLANTS BY STATE
State
Alabama
Arizona
Arkansas
California
Colorado
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Minnesota
Mississippi
Missouri
Nebraska
New Jersey
New York
North Carolina
Ohio
Oklahoma
Oregon
Pennsylvania
Tennessee
Texas
Utah
Washington
Wyoming
All other states
T3.S. TOTAL
Total state
particulate
emissions,
103 metric tons/yr
1,179
73
138
1,006
201
226
405
55
1,143
748
216
348
546
381
266
168
202
95
152
471
481
1,766
94
169
1,811
410
549
72
162
75
4,264
17,872
Total state
capacity,
103 metric tons/yr
153
112
349
271
16
368
496
18
318
141
349
816
91
594
208
454
819
520
45
64
560
151
316
23
150
116
440
91
163
73
0
8,285
Ammonium nitrate
particulate
emissions,
metric tons/yr
348
255
795
617
36
838
1,129
41
724
321
795
1,858
207
1,352
474
1,034
1,865
1,184
102
146
1,275
344
720
52
342
264
1,002
207
371
166
0
18,865
State
particulate
emission
burden
0.030
0.349
0.576
0.061
0.018
0.370
0.279
0.074
• 0.063
0.043
0.368
0.534
0.040
0.355
0.178
0.615
0.923
1.25
0.067
0.031
0.265
0.019
0.765
0.031
0.019
0.064
0.182
0.288
0.229
0.222
_a
0.106
This category is not applicable for all other states since they have no ammonium nitrate
production facilities and therefore no state particulate burden.
National emission burden of ammonium nitrate. Calculated by dividing the total ammonium
nitrate emissions by the total particulate emissions for the United States times 100.
state particulate emissions were obtained from the 1972 National
Emissions Report (17).
Nebraska's burden (1.2%) exceeds 1.0% of the state totals for par-
ticulates; in 16 other states, the particulate emission burdens
(17) 1972 National Emissions Report. EPA-450/2-74-012, U.S. Envi-
ronmental Protection Agency, Research Triangle Park, North
Carolina, June 1974. 422 pp.
39
-------
due to ammonium nitrate manufacturing exceed 0.1%; and the
national particulate emission burden for the industry is 0.14%.
40
-------
SECTION V
CONTROL TECHNOLOGY
A. PARTICULATES
Control technology in the ammonium nitrate industry is primarily
concerned with the removal of particulates from the exhausts of
the neutralizer, evaporator/concentrator, prilling tower, dryer
system, and cooler. These emissions are visible, and state regu-
lations limit both opacity and total particulate emissions.
The particulate emissions from ammonium nitrate production are of
two major types, fine (<3 ym) and coarse (M mm). Particulates
from the neutralizer, evaporator/concentrator and prilling tower
are primarily (>70%) <3 ym in size (12). Table 17 lists the
weight percentages of the particles in the neutralizer and high
density prilling tower gas exit streams. These fine particles are
caused by the dissociation of the NH^NOs solution at high tempera-
tures and recombination of the resulting NH3 and HNOa molecules.
TABLE 17. PARTICLE SIZE DISTRIBUTION IN THE NEUTRALIZER
AND HIGH DENSITY PRILLING TOWER AIRSTREAMS (12)
Particle size, ym Weight percent
Neutralizer
>3 25
1 to 3 25
0.5 to 1 40
0.1 to 0.5 8
<0.1 2
High density prilling tower
>3 30
1 to 3 20
0.5 to 1 35
<0.5 15
The coarser particles (^1 mm in diameter) that come from the dry-
ing and cooling operations are entrained in the cooling or drying
airstream. These emissions are commonly controlled by a high
efficiency (99%) wet scrubber. As stated previously, some plants
41
-------
only have medium efficiency (85%) scrubbers. However, since the
particles emitted from these operations are between 10 ym and
1,000 ym, either system will function equally well in meeting
state limitations.
Collection of microprills, formed primarily in prilling, is a
problem because normal wet scrubbers are not highly (>90%) effi-
cient for particles <3 ym. However, the scrubbers employed in the
past did remove particles above 3 ym, which met regulations for
weight emissions. A more efficient system was needed since this
removal did not place the exit stream within opacity regulations.
The Cooperative Farm Chemicals Association (CFCA) has developed a
collection system which minimizes the cost of removing microprills
from the prill tower exhaust by reducing the amount of air that
must be treated. Since microprills are formed as a result of
thermal stresses in the ammonium nitrate droplet shortly after it
leaves the spray nozzle, CFCA decided to segregate the air immedi-
ately around the nozzle and not to treat all of the air used in
the tower. This involved the design and construction of the
collection cone shown in Figure 15. While this cone is not a
removal system, it greatly reduces the amount of air needing
treatment by any proposed removal system.
Monsanto Enviro-Chem Systems, Inc. has developed a Brink® mist
eliminator unit which is compatible with the CFCA cone (12). The
unit will also accept input from the neutralizer and the evaporator/
concentrator as shown in Figure 16. In a typical high density
installation the combined streams pass through Brink spray catcher
elements, then through Brink high efficiency elements. Each of
these elements serves to remove particles by inertial impaction,
direct interception and Brownian movement. The spray catcher
element removes particles >3 ym with only 250 Pa to 500 Pa pressure
drop. As shown in Figure 17, the high efficiency element removes
99% of all particles <3 ym. The efficiency curve for a venturi
scrubber (7.47 kPa) is shown for comparison.
Figure 16 depicts a typical high density installation. Its various
elements and one not shown, the high velocity elements, can be
recombined in several ways to attain the required efficiency for
each installation. Standard removal efficiencies for such instal-
lations are 90% to 99.5%, and special designs are available for
higher removal efficiencies as the need arises. Actual in-line
test measurements have shown a Brink unit to be 98.6% efficient on
a stream containing neutralizer, evaporator/concentrator and
prilling tower (with the CFCA cone) effluents. It should be noted
that Brink elements tend to deteriorate in a basic atmosphere.
This could pose a problem when using the Brink unit on a combined
prill tower/neutralizer stream.
The Brink unit/CFCA cone control technology has also been shown
to reduce the opacity of prill tower emission. In one test case
42
-------
AMMONIUM NITRATE
SOLUTION
AIR
OUTL
COLLECTION
CONE "—
Ft '
r\n
AIR
i OUTLET
M
I.
-it
\ \
u.
\
\
\
1
(
1
(
LIQUOR IN-
1 '
EXHAUST
1
SCRUBBING
SYSTEM LIQUOR
OUT
ATOMIZING-v
SPRAYS \l
BRINK
HIGH EFFICIENCY
ELEMENTS
PRILLS
BRINK
SPRAY CATCHER
ELEMENTS
FROM
NEUTRALIZER,
EVAPORATOR,
AND
COLLECTION
DUCT
Figure 15. Prilling tower showing
placement and operation of
the CFCA collecting cone.
Figure 16. Full-scale Brink collection unit,
-------
o
LU
O
o
o
o
o
100
99
95
90
80
TYPICAL BRINK H-E
MIST ELIMINATOR
VENTURI
(7.47kPa)
70
0.025
0.1 0.3 0.5
PARTICLE SIZE, \im
Figure 17. Brink high efficiency element collection efficiency.
(18), where a plant was exceeding the 40% opacity allowed by state
code, a Brink unit was installed and opacity levels were reduced
to 9.3% and 10.2%, as read by state inspectors.
At least seven plants have Brink units operating or on order, rep-
resenting 21% of the total industry solution capacity or 57% of
the total high density capacity. At least two other plants have
controls of some sort on their prill towers (18).
Mississippi Chemical Corporation (MCC) has developed other methods
to control neutralizer and prill tower ammonium nitrate emissions
(19). MCC engineers have developed and patented a new type of neu-
tralizer that reduces the release of particulates and fumes to
only a fraction of those emitted from most neutralizers; a reduc-
tion on the order of 10-fold to 20-fold is claimed (19). Emissions
of ammonium nitrate, nitric acid and ammonia are brought under
control by the use of the patented design and by the close pH
control of the neutralizer which is possible only with this new
design. MCC offers this design under license agreement to other
companies. Emissions from the neutralizer have been found to run
<1 kg of ammonia and 0.5 kg of ammonium nitrate per metric ton of
ammonium nitrate production.
(18) Personal communication with T. R. Metzger, Monsanto Enviro-
Chera Systems, Inc., St. Louis, Missouri, May 1976.
(19) Personal communication with M. L. Brown, Mississippi Chemical
Corporation, Yazoo City, Mississippi, May 1976.
44
-------
A scrubber designed by Beco Engineering Company also controls
prill tower emissions (19). These scrubbing units can usually be
fitted directly on top of the prill tower a few feet above the
sprayheads. Each scrubber consists of module units of packing
which may be inserted across the tower. Pilot plant tests have
shown efficiencies >90%. MCC has two prill towers in operation
with a Beco scrubbing system on each tower (19).
B. AMMONIA OR NITRIC ACID
Emissions of ammonia or nitric acid result from the addition of one
of these chemicals in excess to the neutralizer. Control therefore
begins with the proper regulation of the pH in the neutralizer.
Apparently plants may or may not monitor this parameter closely, so
that ammonia emissions could be >1 g/kg of product. There has been
no impetus to develop elaborate monitoring systems because the
quantity of ammonia usually lost is relatively small (^1 g/kg or
less) and no emission regulations are exceeded. The chemical
industry as a whole has developed sophisticated monitoring systems
that could be applied in this situation, if the need arose.
Where a wet collection device is used to control particulates from
the neutralizer, it can also be employed to control ammonia or HNOs
resulting from normal plant operation. This only requires the
maintenance of acidic (or basic) conditions in the scrubbing solu-
tion. No information is available on the efficiency of such
controls.
C. POTENTIAL IMPACT OF CONTROLS
The potential impact of controls on the emissions from an ammonium
nitrate facility is illustrated by the values in Table 18. The
emission parameters presented are derived from applying a CFCA cone
to the prilling tower and then combining the exhaust stream from
the cone with the exhausts from the neutralizer and the evaporator/
concentrator. This combined stream is passed through a Brink unit
for removal of 98.6% of the particulates. A high-efficiency wet
scrubber removes particulate from the cooler. The above mentioned
controls are expected to reduce emissions to zero. Since ammonia
deteriorates the glass fiber packing used in the Brink unit, the
amount of excess NHs fed to the neutralizer will be controlled to
a low level. This small excess will then be absorbed in the con-
trol system.
D. FUTURE CONSIDERATIONS
In the future, the basic problem facing ammonium nitrate manufac-
turers is compliance with state opacity regulations. This is a
particular problem with ammonium nitrate production. Limitations
on emissions weight can be met, for example, with wet scrubbers.
However, because of the submicron size of the particulate emission,
more stringent control technology must be applied to comply with
opacity regulations.
45
-------
TABLE 18. CONTROLLED PARTICULATE EMISSIONS
Emission parameter
Emission factor, g/kg
Emission rate, g/s
Xmax, yg/m3
Source severity
Affected population, persons
Brink
unit
0.049
0.1813
51. 5b
0.20
0
Controlled
cooler
0.05
0.15
16C
0.06
0
aEmission rate determined by 98.6% removal from
emission rates of composite stream, not by multi-
plying the emission factor by total original solu-
tion capacity.
Height of Brink is
-------
SECTION VI
GROWTH AND NATURE OF THE INDUSTRY
A. PRESENT TECHNOLOGY
Ammonium nitrate production technology is well developed. There
have been no major breakthroughs or developments since high den-
sity prilling was begun in the early 1960's. For this reason not
many variations exist in the basic process.
In the other processes discussed, there is also little room for
change except in equipment design. The plants using such pro-
cesses differ from prill plants not only in materials flow, but
also in product. Some plants have more than one method of forming
solids so that their product type can be varied to meet market
demand. At least one plant is known to have a prilling tower,
graining kettles, and a Sandvik belt.
B. EMERGING TECHNOLOGY
Changes in neutralizer operation are among the most significant
emerging developments that may affect emissions. Neutralization
can be carried out under pressure or vacuum as well as at atmo-
spheric conditions. Pressure neutralization gives maximum
economy since the steam produced is used to concentrate the
ammonium nitrate solution and to preheat the nitric acid. However,
additional control equipment must be used to control excess ammonia
in the contaminated condensate.
In 1967, American Cyanamid Company started operation of a vacuum
neutralizer at their Hannibal, Missouri, ammonium nitrate plant
(20). Operating conditions were adjusted so that stainless steel
could be used as the construction material rather than the normal
titanium steel. Also, the solution is concentrated to 95%,
instead of the normal 83%. If this type of design were imple-
mented throughout the industry, the need for an evaporator in the
low density process would be eliminated.
(20) Falck-Muus, R. New Process Solves Nitrate Corrosion. Chem-
ical Engineering, 74(14):108-116, 1967.
47
-------
C. INDUSTRY PRODUCTION TRENDS
Figure 18 illustrates actual growth in the ammonium nitrate indus-
try from 1960 to 1975 and the projected growth through 1980 (1).
The ammonium nitrate industry continues to be the largest single
end-chemical user of nitrogen from ammonia, consuming about 21% of
domestic synthetic ammonia production in 1974.
9,000
2,000
I960
1965
1970
YEAR
1975
1980
Figure 18. Capacity and production trends.
Table 19 gives ammonium nitrate production data by end use for
1976 (21). Approximately 6.3% of ammonium nitrate produced was as
a solution intended for direct sale; 27.7% as a solution for use
in the manufacture of nitrogen solutions or other fertilizer mater-
ials; 50.2% as a solid; and 15.8% intended for other uses such as
the manufacture of explosives.
Ammonium nitrate production is expected to increase at an average
rate of 2% per year through 1978. If no additional controls are
installed in the industry, particulate emissions in 1978 should be
13% greater than in 1972. However, new controls are being added,
(21) Current Industrial Reports, Inorganic Fertilizer Materials
and Related Products. Series M28B(76)-1 through M28B(76)-12,
U.S. Department of Commerce, Washington, D.C., January
through December 1976. 6 pp. each.
48
-------
TABLE 19. PRODUCTION OF AMMONIUM NITRATE BY END USE, 1976
Production Metric tons/yr
Fertilizer
Liquid for direct sale 476,763
Liquid for consumption in other nitrogen fertilizers 2,100,507
Solid 3,821,829
Subtotal 6,399,099
Other uses 1,189,074
Production total 7,588,173
and the relative amount of ammonium nitrate solutions as a final
product versus solids is increasing. Therefore, 1978 emissions
may be lower than those in 1972 even with increased production.
49
-------
REFERENCES
1. Personal communication with E. A. Harre, National Fertilizer
Development Center, Tennessee Valley Authority, Muscle Shoals,
Alabama, June 1976.
2. Williams, L., L. F. Wright, and R. Hendriks. Process for
the Production of Ammonium Nitrate. U.S. Patent 2,402,192
(to Consolidated Mining and Smelting Co. of Canada, Ltd.),
June 18, 1946.
3. Shearon, W. H., Jr., and W. B. Dunwoody. Ammonium Nitrate.
Industrial and Engineering Chemistry, 45(3):496-504, 1953.
4. Sharp, J. C. Nitrogen. In: Chemistry and Technology of
Fertilizers, V. Sauchelli, ed. Reinhold Publishing Corpora-
tion, New York, New York, 1960. pp. 26-32.
5. Miller, P., and W. C. Saeman. Properties of Monocrystalline
Ammonium Nitrate Fertilizer. Industrial and Engineering
Chemistry, 40 (1):154-160, 1948.
6. Slack, A. V. Fertilizer Developments and Trends - 1968.
Noyes Development Corporation, Park Ridge, New Jersey, 1968.
405 pp.
7. Ammonium Nitrate. In: Riegel's Handbook of Industrial Chem-
istry, J. A. Kent, ed. Van Nostrand Reinhold Company, New
York, New York, 1974. pp. 100-101.
8. Hester, A. S., J. J. Dorsey, Jr., and J. T. Kaufman. Stengel
Process Ammonium Nitrate. Industrial and Engineering Chemis-
try, 46(4):622-632, 1954.
9. Stengel, L. A. Process for Producing Ammonium Nitrate.
U.S. Patent 2,568,901 (to Commercial Solvents Corp.),
September 25, 1951.
10. Dorsey, J. J. Ammonium Nitrate by the Stengel Process.
Industrial and Engineering Chemistry, 47(1):11-17, 1955.
11. Volk, W. Applied Statistics for Engineers, Second Edition.
McGraw-Hill Book Co., New York, New York, 1969. 354 pp.
50
-------
12. Metzger, T. R. Controlling Airborne Emissions from Ammonium
Nitrate Production. Presented to Ammonium Nitrate Study
Group, Sarnia, Ontario, August 1974. 7 pp.
13. Feick, G., and R. M. Hainer. On the Thermal Decomposition of
Ammonium Nitrate. Steady-State Reaction Temperatures and
Reaction Rate. Journal of the American Chemical Society,
76(23):5860-5863, 1954.
14. Ammonia. In: The Merck Index, P. G. Stecher, ed. Merck
and Company, Inc., Rahway, New Jersey, 1968. p. 64.
15. Matheson Gas Data Book. The Matheson Company, Inc., East
Rutherford, New Jersey, 1966. pp. 13-22.
16. Point Source Listings. National Emissions Data System.
U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina, May 1974.
17. 1972 National Emissions Report. EPA-450/2-74-012, U.S. Envi-
ronmental Protection Agency, Research Triangle Park, North
Carolina, June 1974. 422 pp.
18. Personal communication with T. R. Metzger, Monsanto Enviro-
Chem Systems, Inc., St. Louis, Missouri, May 1976.
19. Personal communication with M. L. Brown, Mississippi Chemical
Corporation, Yazoo City, Mississippi, May 1976.
20. Falck-Muus, R. New Process Solves Nitrate Corrosion. Chem-
ical Engineering, 74 (14):108-116, 1967.
21. Current Industrial Reports, Inorganic Fertzilier Materials
and Related Products. Series M28B(76)-1 through M28B(76)-12,
U.S. Department of Commerce, Washington, D.C., January
through December 1976. 6 pp. each.
22. TLVs® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1975. American Conference of Governmental Indus-
trial Hygienists, Cincinnati, Ohio, 1975. 97 pp.
23. Turner, D. B. Workbook of Atmospheric Dispersion Estimates.
Public Health Service Publication No. 999-AP-26, U.S. Depart-
ment of Health, Education, and Welfare, Cincinnati, Ohio, May
1970. 84 pp.
24. Hesketh, H. E. Understanding and Controlling Air Pollution.
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan,
1974. p. 63.
51
-------
25. Eimutis, E. C., and M. G. Konicek. Derivations of Continuous
Functions for the Lateral and Vertical Atmospheric Dispersion
Coefficients. Atmospheric Environment, 6(11):859-863, 1972.
26. Metric Practice Guide. ASTM Designation: E 380-74, American
Society for Testing and Materials, Philadelphia, Pennsylvania,
November 1974. 34 pp.
52
-------
APPENDIX A
RAW DATA USED TO CALCULATE EMISSION FACTORS
AND PLANT CHARACTERISTICS
Table A-l (1) is a list of 64 ammonium nitrate plants giving
location, population density of the county or parish in which each
is located, and annual capacity.
Table A-2 is a compilation of data obtained from the National Emis-
sions Data System, from private industry, and from field observa-
tions. All emission factors indicated are from source measurements,
Table A-3 is a compilation of particulate emission factors from
prilling operations.
Table A-4 is a listing of stack heights for various points in an
ammonium nitrate manufacturing plant. These values were obtained
from private industry contacts and field observations.
53
-------
TABLE A-l. AMMONIUM NITRATE PLANT LOCATIONS AND CAPACITIES (1)
County or parish
Company
Agway, Inc.
Air Products and Chemicals, Inc.
Allied Chemical Corp.
Union Texas Petroleum Division
Agricultural Department
American Cyanamid Co.
Agricultural Division
Apache Powder Co.
CF Industries, Inc.
Chattanooga Nitrogen Complex
Fremont Nitrogen Complex
North Carolina Nitrogen Complex
Terre Haute Nitrogen Complex
Coastal States Gas Corp.
Colorado Interstate Corp. (subsidiary)
Wycon Chemical Co. (subsidiary)
Columbia Nitrogen Corp.
Cominco American, Inc.
Cooperative Farm Chemicals Association
E. I. duPont de Nemours & Co., Inc.
Industrial Chemicals Department
Polymer Intermediates Department
Farmland Industries, Inc.
Gardinier Big River, Inc.
Goodpasture, Inc.
City/state
Olean, NY
Pensacola, FL
Geismar, LA
Omaha, NE
Hannibal, MO
Benson, AZ
Tyner, TN
Fremont, NE
Tunis, NC
Terre Haute, IN
Cheyenne , WY
Augusta, GA
Beatrice, NE
Lawrence, KS
Gibbstown, NJ
Du Pont, HA
Louviers , CO
Seneca, IL
Dodge City, KS
Helena, AR
Dimmitt, TX
Name
Cattaraugus
Escambia
Acension
Douglas
Marion
Cochise
Hamilton
Dodge
Hertford
Vigo
Laramie
Richmond
Gage
Douglas
Gloucester
Pierce
Douglas
LaSalle
Ford
Phillips
Castro
Population
density,
persons/km2
23.4
116.6
46.7
446.3
24.2
3.8
170.3
65.5
63.4
272.5
20. 5
498.0
33.5
100.1
518.8
241.8
3.8
95.6
7.8
57.8
11.7
Annual
capacity,
10* metric tons/yr
64
182
102
272
123
112
116
30
363
, 141
73
212
154
417
45
18
16
182
73
95
30
W. R. Grace & Co.
Agricultural Chemicals Group
Gulf Oil Corp.
Gulf Oil Chetnicala Co.
Industrial and Specialty Chemicals Division
Hercules, Inc.
Industrial Systems Department:
Synthetics Department
Illinois Nitrogen Co.
International Minerals fc Chemical* Corp.
Kaiser Aluminum & Chemicals Corp.
Kaiser Agricultural Chemicals Division
Lone Star Gas Co.
Nipak, Inc. (subsidiary)
Mississippi Chemicals Corp.
Mobil Oil Corp.
Mobil Chemical Co.
Petrochemicals Division
Monsanto Co.
Monsanto Agricultural Products Co.
Nitram, Inc.
N-Ren Corp.
Cherokee Nitrogen Division
Occidental Petroleum Corp.
Occidental Chemical Co. (subsidiary)
Wilmington, NC
Pittsburg, KS
Brunswick
Crawford
62.6
197
326
Bessemer , AL
Carthage, MO
Donora , PA
Hercules, CA
Louisiana, MO
Marseilles, IL
Sterlington, LA
Bainbridge, GA
North Bend, OH
Savannah , GA
Tampa, FL
Kerens, TX
Xazoo City, MS -
Beaumont, TX
El Dorado, AR
Luling, LA
Tampa, FL
Pryor, OK
Hanford, CA
Jefferson
Jasper
Washington
Alameda
Pike
LaSalle
Union
Decatur
Hamilton
Chatham
Hillsborough
Navraro
Yazoo
Jefferson
Union
St. Charles
Hillsborough
Mayes
Tolare
573.0
122.5
244.4
1,444.8
24.3
95.6
19.7
37.8
2,211.0
410.3
466.8
27.9
28.0
255.2
41.8
96.2
466.8
33.3
14.8
23
_a
136
73
459
136
170
55
93
229
50
65
454
177
254
322
136
77
33
This plant obtains solutions from another facility and produces captive grained product.
(continued)
54
-------
TABLE A-l (continued).
Company
Phillips Pacific Chemical Co.
Phillips Petroleum Co.
Fertilizer Division
Reichhold Chemicals, Inc.
St. Paul Ammonia Products, Inc.
J. R. Simplot
Minerals and Chemicals Division
Skelly Oil Co.
Hawkeye Chemical Co. (subsidiary)
Standard Oil Co. of California
Chevron Chemical Co. (subsidiary)
Ortho Division
The Standard Oil Co. (Ohio)
Vistron Corp. (subsidiary)
Chemicals Department
Tennessee Valley Authority
Terra Chemicals International, Inc.
Tyler Corp.
Atlas Powder Co. (subsidiary)
Union Oil Co. of California
Collier Carbon and Chemical Corp.
(subsidiary)
United States Steel Corp.
VSS Agri-Chemicals
Valley Nitrogen Producers, Inc.
The Williams Companies
Agrico Chemical Co. (subsidiary)
TOTAL
TABLE A- 2. AMMONIUM
Plant
A
B
C
D
E
F
G
H
I
J
K
L
M
Mean
Standard deviation
City/state
Kennevick, WA
Beatrice, ME
Ettar, TX
St. Helens, OK
St. Paul, MN
Pocatello , ID
Clinton, IA
Fort Madison, IA
Kennewick, WA
Lima, OH
Muscle Shoals, AL
Fort Neal (Sioux
City) , IA
Joplin, MO
Tamaque, PA
Brea, CA
Cherokee, AL
Crystal City, MO
Geneva , UT
El Centre, CA
Helm, CA
Henderson, KX
Verdigris, OK
County
Name
Ben ton
Gage
Moure
Columbia
Dakota
Bannock
Clinton
Lee
Ben ton
Allen
Colbert
ttoodbury
Jasper
Schuylkill
Orange
Colbert
Jefferson
Imperial
Fresno
Henderson
or parish
Population
density,
persons/km2 10 3
38.5
33.5
IS. 2
44.5
240.7
45.3
31.3
79.6
38.5
270.4
82.3
44.6
122.5
201.9
1,802.2
82.3
155.4
17.3
26.3
81.6
Annual
capacity,
metric tons/yr
50
64
168
23
208
18
136
77
95
58
39
136
146
14
55
91
91
91
41
69
91
239
8,285
NITRATE PLANT PARTICULATE EMISSION FACTORS
(gAg)
Neutral- Evapo-
izer rator
0.32 0.04
0.15 0.02
0.59
0.35
3.8
1.9
4.3
1.7
1.64 0.03
1.63 0.01
Prill
tower
2.11
3.69
0.74
0.45
1.82
1.32
3.80
1.99
1.33
Dryer Cooler
0.23
0.73
5.13a
0.66
0.54 5.13
0.27
No control device; assume for representative plant that a 98%
efficient scrubber would be added.
Note: Blanks in table indicate data not available.
55
-------
TABLE A-3.
PRILL TOWER EMISSIONS
(g/kg)
0.51
1.70
1.51
1.26
1.38
1.54
1.12
0.88
1.46
2.44
0.66
1.23
1.43
1.80
1.21
1.56
1.19
3.36
0.74
1.30
0.77
1.45
1.29
1.03
1.30
1.64
1.36
1.36
1.55
1.36
0.87
1.31
1.28
1.47
1.46
1.07
1.26
1.10
0.80
0.87
1.28
1.25
1.24
1.78
1.35
1.39
1.16
1.05
1.11
1.28
0.95
1.14
1.32
1.33
2.17
0.85
2.09
0.94
0.97
0.78
1.64
1.05
1.12
1.23
1.21
1.92
1.24
Data received from a private industry source.
TABLE A-4. REPRESENTATIVE STACK HEIGHTS (16)
(meters)
Plant Neutralizer
A 9.75
B 20.73
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
Average 15.2
Prilling
Evaporator tower Dryer
13.11 46.63 7.92
18.9 56.96
12.98 56.39
37.49 15.32
21.3
37.15 14.61
30.48
9.81 35.01 16.98
50.29
33.53
46.94
42.67
48.46
38.10
28.04
34.62
36.58
60.96
13.7 41.2 13.7
Cooler
7.92
14.61
16.98
13.2
Note: Blanks in table indicate data not available.
56
-------
APPENDIX B
DERIVATION OF SOURCE SEVERITY EQUATIONS
The potential effect of the emissions from ammonium nitrate pro-
duction on ambient air quality can be evaluated by_comparing the
time averaged maximum ground level concentration, Xmax' to tne
ambient air quality standard or a reduced threshold limit value,
TLV (22) for noncriteria pollutants for each emission.
An initial calculation of the instantaneous (3 minutes) maximum
ground level concentration, Xmax* must first be made. For an
elevated point source in neutral atmospheric conditions (atmos-
pheric stability class C) , the following equation is applicable (23)
*-* '-
where Q = pollutant emission rate, g/s
u = average wind speed, m/s
h = effective stack height, m
e = 2.72
7T = 3.14
Equation B-l is the solution to the Turner plume dispersion equa-
tion (23) at that distance downv/ind from the source for which the
ground level concentration is a maximum. The Xmax equation does
not establish the point where Xmax occurs with respect to the
source, but it does determine the maximum instantaneous ground
level concentration of the emission species.
(22) TLVs® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1975. American Conference of Governmental Indus-
trial Hygienists, Cincinnati, Ohio, 1975. 97 pp.
(23) Turner, D. B. Workbook of Atmospheric Dispersion Estimates.
Public Health Service Publication No. 999-AP-26, U.S. Depart-
ment of Health, Education, and Welfare, Cincinnati, Ohio, May
1970. 84 pp.
57
-------
The severity of an emission source can be calculated by the
general equation:
S = Sp (B_2)
where _ S = source severity
X = average maximum ground level concentration over
max a defined averaging period, g/m3
F = hazard factor, g/m*
The quantity F in Equation B-2 is equal to the ambient air quality
standard for criteria pollutants (i.e., CO, SOX, NOX, hydrocarbons,
and particulates). For noncriteria pollutants, F is calculated by
the following equation:
F = TLV ' 2! • TM (B-3)
where TLV = threshold limit value of the emitted material, g/m3
8/24 = correction factor for the 8-hr work day which is
the basis for the TLV
1/100 = safety factor
Values of F for particulates and NO are 260 vg/m3 and 100 yg/m3,
respectively. The value of F for ammonia, calculated from a TLV
of 18 mg/m3, is 60 yg/m3.
The value of Xniax is determined from the value of xmax using the
following equation:
0.17
*max = *l — J (B'4)
where to - time of instantaneous maximum ground level concen-
tration determination, i.e., 3 min
t = averaging time for emission species, min
0.17 = curve fitting exponent from correlation
For particulates the averaging time is 24 hr (1,440 min). Using
this value in Equation B-4, and substituting Equation B-l into
Equation B-4 and reducing, the following equation results for
Xmax for particulates:
xmax
= 0.0182 Q ( 5)
The source severity equation for particulates can be derived
similarly from Equations B-2 and B-5 to be:
70 Q
S = h2 (B-6)
58
-------
The averaging time for ammonia is also 24 hr and F = 60 jig/m3;
hence:
7 = 0.0182 Q (B.5)
max
and S = 303 Q (B-7)
h2
For nitric acid emissions the averaging time is also 24 hr. If
HNOa emissions were present the Xmax e{3uati°n would be the same
as Equation B-5. From this the following source severity equation
is derived, based on the TLV for nitric acid of 5 mg/m3:
S = ' (B-8)
h2
For situations in which fugitive emissions are released from
the confinement of a building, a different type of equation is
needed to determine the maximum ground level concentration and
related source severity. An equation of this type is (24) :
X(x,o/0) = (TT 0y az + A')TT (B~9)
where x / \ = concentration downwind with no vertical or
' horizontal component, g/m3
Q = emission rate, g/s
a ,0 = horizontal and vertical deviation, m
A1 = 1/2 the cross sectional area of the building
_ perpendicular to the flow of the wind, m2
u = mean wind speed = 4.5 m/s
This concentration at the distance to plant boundary (D) is equiva-
lent to Xmax' For stability class C, it has been shown that (25) :
a = 0.209 D0-903 (B-10)
cj = 0.113 D°'911 (B-ll)
z
where D = distance to plant boundary downwind from source in the
horizontal plane, m
(24) Hesketh, H. E. Understanding and Controlling Air Pollution.
Ann Arbor Science Publishers, Inc., Ann Arbor, Michigan,
1974. p. 63.
(25) Eimutis, E. C., and M. G. Konicek. Derivations of Continuous
Functions for the Lateral and Vertical Atmospheric Dispersion
Coefficients. Atmospheric Environment, 6(11):859-863, 1972.
59
-------
By substituting B-10 and B-ll into B-9 the following equation is •
derived:
max [(7.42 x ID'2) D1'81" + A1] 4.5
Employing the conversion of Xmax given by Equation B-4 for a
24-hr averaging period, the following is derived:
- = - 7.78 x 10-2 Q -
max (7.42 x ID'2) D1-8** + A.
The source severity is then calculated using the particulate
ambient air quality standard of 260 yg/m3 (24-hr averaging time) :
S =
(7.42 x ID'2) D1-81" + A1
Equation B-14 is then used to calculate the source severity of
emissions from confinement at any given distance, usually the
plant boundary.
For a worst case analysis, Equation B-14 can be maximized by
allowing A1 to approach 0. As a result of this operation, Equa-
tion B-15 is derived:
S = 4,030 Q D"1'814 (B-15)
and likewise the equation for x"max becomes:
*max = 1'05 Q D'1*811* (B-16)
60
-------
APPENDIX C
ESTIMATE OF STEAM FROM NEUTRALIZES
The quantity of steam liberated from the neutralizer by the heat
of reaction is equal to the difference in concentrations of the
initial solution and the exiting solution. For an average plant
(1.315 x 105 metric tons/yr), concentrating the initial solution
from 61% NHitN03 to an exiting solution of 83% releases ^3.45 m3/s
of steam at 150°C and 101 kPa. Another author has indicated that
the steam output from a 9.09 x 101* metric ton/yr plant is ^1.12 m3/s
(12). Based on a linear scaleup to average capacity, the quantity
becomes 1.62 m3/s. The lower value is presumably the result of
lower steam temperatures and partial condensation.
61
-------
APPENDIX D
CALCULATION OF AFFECTED POPULATION
Since the xmax e(3uation (Equation B-l) does not contain a down-
wind distance term, another form of the Gaussian plume equation
is needed to find the population affected by plant emissions.
The following is a rationale for the calculations.
If the wind directions are taken to 16 points and it is assumed
that the wind directions within each sector are distributed
randomly over a period of a month or a season, it can further be
assumed that the effluent is uniformly distributed in the horizontal
within the sector. The appropriate equation for average concen-
tration is then(22):
- _ 2.03 Q
x o uX
z
exp - i £- (g/*3)
(D-l)
Since the distances at which severity x/F equals 1.0 are desired,
the roots (X) of the following equation are determined:
:.o3 Q _r_ i /h \ZM_ lmQ
uX
exp
:L
2
)
(D-2)
In addition,
+ c
(D-3)
where a, b, and c are functions of atmospheric stability. In this
case the atmospheric stability is assumed to be class C and the
values of a, b, and c are 0.113, 0.911, and 0.0, respectively.
For a specified emission from a typical source, severity as a
function of distance can be represented as follows:
62
-------
with Xj and X2 being the distances from the source where S = 1.0.
Sweeping those distances through 6.282 rad an annulus is obtained:
The affected population is then in an area, A:
A = ir(X22 - X!2) (km2)
(D-4)
If the affected population density is D , then the total affected
population, P, is: "
P = D A (persons)
(D-5)
There could be cases where the downrange severity function appears
as follows:
UJ
Thus the affected population is zero. In other cases the affected
population will be extremely low regardless of population density
because the severity for a particular emission may correspond to
that shown below and thus result in a small annulus:
i.o
63
-------
GLOSSARY
additive: Any material added to the concentrated ammonium nitrate
before solids formation which changes the natural mechanical
characteristics of the ammonium nitrate solid particle.
affected population: Number of persons living in the area near a
representative plant where the source severity is greater
than 1.0.
emission factor: Mass of an emission per unit weight of final
product.
emission growth factor: Ratio of emissions for 1978 versus 1972.
emission rate: Mass of emissions per unit time.
fugitive emissions: Gaseous and particulate emissions that result
from industrial related operations, but which are not emitted
through a primary exhaust system, such as a stack, flue, or
control system.
graining: Process in which concentrated ammonium nitrate solution
is solidified by placing a concentrated solution in a steam
jacket-heated kettle and mixing until the water evaporates.
granulation: Process in which concentrated ammonium nitrate solu-
tion is solidified by spraying the concentrated solution on
a falling curtain or rolling bed of seed paritcles to build
a larger particle.
high density prills: Prills formed using a 99.5+% solution that
lowers the number of void spaces formed to create a more
dense prill.
low density prill: Prills formed using a 95% to 96% solution that
allows more voids in the final product, giving a less dense
prill.
national emission burden: Mass of particulates emitted from the
ammonium nitrate industry divided by the total national
particulate emissions expressed in percent.
original solution capacity: Capacity of the facility expressed as
weight of solution (not final solid product) produced per
unit time.
64
-------
prill: A spherical particle formed by the solidification of an
ammonium nitrate droplet.
prilling: Process in which concentrated ammonium nitrate solution
is solidified by spraying the concentrated solution in a
tower so that the drops formed fall countercurrent to a stream
of cooling air.
source severity: Ratio of the ground level concentration of each
emission species to its corresponding ambient air quality
standard (for criteria pollutants) or to a reduced TLV (for
noncriteria emission species).
state emission burden: Mass of particulates emitted from the
ammonium nitrate industry in a particular state divided by
the total state particulate emissions expressed in percent.
threshold limit value (TLV): Refers to the airborne concentration
of a substance and represents conditions under which it is
believed that nearly all workers may be repeatedly exposed
day after day without adverse effect for a 7- or 8-hour work-
day and 40-hour workweek.
65
-------
CONVERSION FACTORS AND METRIC PREFIXES (26)
CONVERSION FACTORS
To convert from
degree Celsius (°C)
joule (J)
kilogram (kg)
kilogram (kg)
kilometer2 (km2)
meter (m)
meter (m)
meter2 (m2)
meter3 (m3)
metric ton
pascal (Pa)
radian (rad)
to
Prefix
k
m
M
Symbol
kilo
milli
micro
degree Fahrenheit
British thermal unit
pound-mass (Ib mass
avoirdupois)
ton (short, 2,000 Ib mass)
mile2
foot
mile
foot2
foot3
pound
atmosphere
degree (°)
METRIC PREFIXES
Multiplication
factor
Multiply by
t° = 1.8
9.479 x I0~k
2.205
1.102 x 10~3
3.861 x 10~!
3.281
6.215 x IQ-*
1.076 x 101
3.531 x 101
2.205 x 103
9.869 x 10~6
5.730 x 101
+ 32
Example
io-3
10~6
1 kPa = 1 x 103 pascals
1 mm = 1 x 10"3 meter
1 jig = 1 x 10~6 gram
(26) Metric Practice Guide. ASTM Designation E-280-74, American
Society for Testing and Materials, Philadelphia, Pennsylvania,
November 1974. 34 pp.
66
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-1071
2.
3. RECIPIENT'S ACCESSION-NO.
». TITLE AND SUBTITLE
SOURCE ASSESSMENT: Ammonium Nitrate
Production
. REPORT DATE
September 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
W.J. Search and R.B. Reznik
8. PERFORMING ORGANIZATION REPORT NO
MRC-DA-582
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
Dayton Laboratory
Dayton, Ohio 45407
10. PROGRAM 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; 4/75-7/77
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES
IERL-RTP task officer for this report is Ronald A. Venezia, Mai}
Drop 62, 919/541-2547. Related reports are also in the EPA-600/2-76-032 series.
16. ABSTRACT
The report describes a study of air pollutants emitted by the ammonium
nitrate industry. The potential environmental effect of the source was evaluated.
Representative processes and an average plant were defined for the purpose of
establishing a base on which to determine the emissions and severity of the source.
The industry produces 39% of its original solution capacity as ammonium nitrate
solutions and 61% as solids, of which 92% are formed by prilling. The remaining 8%
are formed by granulation or graining. Primary emissions from ammonium nitrate
plants are particulates and ammonia. Processes releasing the greatest amount of
emissions are the neutralizer (particulates and ammonia) and the prill tower
(particulates). Emission factors were found to be highly dependent on individual
plant operation.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
:. COSATI Field/Group
Air Pollution
Ammonium Nitrate
Industrial Processes
Pelleting
Granulation
Dust
Ammonia
Air Pollution Control
Stationary Sources
Source Assessment
Emission Factors
Prilling
Particulates
13B
07B
13H
07A
07D
11G
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
78
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
67
------- |