United State*
Environmental Protection
Agency
Research EPA-600/2-79-O19c
Park NC 27711
Research and Dgvetopment
&ERA Source Assessment:
Phosphate Fertilizer
Industry
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental 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.
-------
EPA-600/2-79-019c
May 1979
Source Assessment:
Phosphate Fertilizer Industry
by
J. M. Nyers, G. D. Rawlings, E. A. Mullen,
C. M. Moscowitz, and R. B. Reznik
Monsanto Research Corporation
Box 8, Station B
Dayton, Ohio 45407
Contract No. 68-02-1874
Program Element No. 1AB015; ROAP 21AXM-071
EPA Project Officer: Ronald A. Venezia
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
The Industrial Environmental Research Laboratory (IERL) of the
U.S. Environmental Protection Agency (EPA) has the responsibility
for insuring that pollution control technology is available for
stationary sources to meet the requirements of the Clean Air Act,
the Federal Water Pollution Control Act, and solid waste legisla-
tion. If control technology is unavailable, inadequate, or
uneconomical, then financial support is provided for development
of needed control techniques for industrial and extractive proc-
ess industries. Approaches considered include: process modifi-
cations, feedstock modifications, add-on control devices, and
complete process substitution. The scale of the control technol-
ogy programs ranges from bench- to full-scale demonstration
plants.
The Chemical Processes Branch of the Industrial Processes
Division of IERL has the responsibility to develop control tech-
nology for a large number of operations (more than 500) in the
chemical industries. As in any technical program, the first
question to answer is, "Where are the unsolved problems?" This
is a determination which should not be made on superficial infor-
mation; consequently, each of the industries is being evaluated
in detail to determine if there is, in EPA's judgement, suffi-
cient environmental risk associated with the process to invest in
the development of control technology. This report on the phos-
phate fertilizer industry contains data necessary to make that
decision for the air, water, and solid waste discharges resulting
from the production of phosphoric acid and superphosphoric acid,
normal and triple superphosphate fertilizer, and granular ammo-
nium phosphate fertilizer.
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 reponsibility
as outlined above. Dr. Robert C. Binning serves as Program
Manager in this overall program entitled "Source Assessment,"
which includes investigation of sources in each of four cate-
gories: combustion, organic materials, inorganic materials, and
open sources. Dr. Dale A. Denny of the Industrial Processes
Division at Research Triangle Park serves as EPA Project Officer.
In this study of the phosphate fertilizer industry. Dr. R. A.
Venezia served as EPA Task Officer.
111
-------
ABSTRACT
This report describes a study of air emissions, water effluents,
and solid residues resulting from the manufacture of phosphate
fertilizers. It includes the production of wet process phosphor-
ic acid, superphosphoric acid, normal superphosphate, triple
superphosphate, and ammonium phosphate. The potential environ-
mental impact of the industry is evaluated on a multimedia basis.
Air emissions from production of phosphate fertilizers include
particulates, fluorides, ammonia, and sulfur oxides. The poten-
tial environmental effect of these emissions is evaluated by cal-
culating the source severity, defined as the ratio of the time-
averaged maximum ground level concentration of a pollutant to a
hazard factor. For particulate and sulfur oxide emissions, the
hazard factor is the primary ambient air quality standard; for
fluoride and ammonia emissions, it is a reduced threshold limit
value. Source severity values for emissions from the wet scrub-
ber system at an average phosphoric acid process are 0.18 for
fluorides and below 0.05 for particulates and sulfur oxides. For
superphosphoric acid, severity is 0.09 for fluoride and below
0.05 for particulates. For ammonium phosphate, severities are
0.43 for particulate, 0.45 for fluoride, and 0.09 for ammonia.
For normal superphosphate, source severity ranges from 0.004 to
0.35 for particulate and from 0.18 to 7.2 for fluoride. For
run-of-the-pile triple superphosphate, particulate source sever-
ity ranges from 0.009 to 0.04, and fluoride source severity is
0.77. For granular triple superphosphate, particulate source
severity ranges from 0.004 to 0.06, fluoride source severity
ranges from 0.12 to 0.36, and SOX source severity is 0.11.
Phosphate fertilizer plants control air emissions by a combina-
ation of cyclones, baghouses, and wet scrubbers. Material han-
dling operations are generally enclosed to reduce fugitive
particulate emissions. Only fluoride emissions from curing and
storage at normal superphosphate plants are typically uncontrolled,
Water effluents from the production operation arise from wet
scrubbers, barometric condensers, steam jet ejectors, gypsum
slurry, and acid sludge. Noncontact cooling water is normally
segregated from other wastewater streams. Wastewaters are con-
taminated with phosphates, fluorides, sulfates, and gypsum.
Process water is discharged to large gypsum ponds for storage
and recycle; it is normally not discharged to surface streams.
IV
-------
Solid residues generated at phosphoric acid plants are gypsum
from the filtration of wet process phosphoric acid, wet process
phosphoric acid sludge, and solids suspended in the wet scrubber
liquor. These solid waste residues are, for the most part,
stored in ponds, stacked in piles, or stored in mining pits on
site.
This report was submitted in partial fulfillment of Contract
68-02-1874 by Monsanto Research Corporation under the sponsorship
of the U.S. Environmental Protection Agency. The study covers
the period May 1976 to March 1979.
-------
CONTENTS
Preface iii
Abstract iv
Figures ix
Tables xii
Abbreviations and Symbols xvi
1. Introduction 1
2 . Summary 2
3. Source Description 10
Overview of phosphate fertilizer industry 10
Wet process phosphoric acid production 22
Superphosphoric acid production 37
Normal superphosphate production 41
Triple superphosphate production 44
Ammonium phosphate production 50
4. Air Emissions 67
Wet process phosphoric acid 67
Superphosphoric acid 73
Normal superphosphate 74
Triple superphosphate 77
Ammonium phosphates 80
Potential environmental effects 83
Air pollution control technology 104
Byproduct recovery 114
5. Water Effluents 117
Sources of wastewater 118
Potential environmental effects 130
6. Solid Residue 137
Sources of solid residue 137
Potential environmental effects 139
Control technology 139
7. Growth and Nature of the Industry 142
Present technology 142
Emerging technology 145
Industry production trends 147
References 149
Appendices
A. Phosphate fertilizer plants in the United States in
1975 or 1976 158
B. Emissions data 166
VII
-------
CONTENTS (continued)
C. Mass balances 174
D. NEDS data base 178
Glossary 182
Conversion Factors and Metric Prefixes 185
Vlll
-------
FIGURES
Number
1 Location of major phosphate rock deposits in the
United States 11
2 U.S. phosphate rock consumption pattern for various
phosphorus products 12
3 Schematic diagram of phosphate fertilizer industry . .12
4 Location of wet process and superphosphoric acid
plants 14
5 Location of NSP plants in the United States 16
6 Location of TSP plants in the United States 16
7 Location of ammonium phosphate plants in the
United States 17
8 Preparation of phosphate rock for acidulation 20
9 Raw material unloading and storage 21
10 Precipitation and stability of calcium sulfates in
phosphoric acid 23
11 Wet process for production of phosphoric acid 26
12 Digestion system designs 28
13 Phosphate rock digester and cooling system 30
14 Tilting pan filtration system 31
15 Operating cycle of rotary horizontal tilting pan
filter „ 31
16 Concentration and clarification 33
17 Distribution of WPPA plants by capacity 36
18 Distribution of SPA plants by capacity 40
19 ROP-NSP production facility 42
20 ROP-TSP production facility 46
21 Production of GTSP from cured ROP-TSP 47
22 Dorr-Oliver slurry granulation process for TSP ... .48
23 Solubility boundaries for the ammonia-phosphoric
acid-water system 52
IX
-------
FIGURES (continued)
Number Pa
24 Ammonium phosphate solubility and viscosity as a
function of NHtHPOi, mole ratio .......... 54
25 TVA ammonium phosphate process flow diagram ...... 57
26 TVA rotary ammoniator-granulator ........... 58
27 Cumulative screen analysis of DAP ........... 60
28 Product dust control system .............. 60
29 Dorr-Oliver ammonium phosphate process flow diagram. .61
30 Diagram of pugmill (blunger) ; top and end views. . . .62
31 Recent history of ammonium phosphate capacity and
production ................... . .63
32 Cumulative distribution of ammonium phosphate plants
and capacity in 1975 ................ 66
33 Schematic of emission points in WPPA manufacture . . .67
34 Source severity distribution of particulate emissions
from rock handling operations at WPPA plants . . . .86
35 Source severity distribution of particulate and
fluoride emissions from the wet scrubber at WPPA
plants ....................... 87
36 Source severity distribution of fluoride emissions
from superphosphoric acid plants .......... 88
37 x/F as a function of radial distance downwind from
gypsum pond ..................... 89
38 Distribution of distance to stated severity for fluor-
ide emissions from the gypsum pond at WPPA plants. . .89
39 Cumulative source severity distributions ....... 92
40 Severity distribution for total plant ammonia
emissions ...................... 94
41 Severity distribution for total plant particulate
emissions ...................... 95
42 Severity distribution for total plant fluoride
emissions ...................... 95
43 x/F as a function of distance from an elevated
source ...................... 102
44 General distribution of x/F as a function of
distance for a ground level source ........ 102
45 Spray-crossflow packed scrubber ........... 106
-------
FIGURES (continued)
Number Page
46 Inlet concentration versus outlet concentration at
scrubber discharge temperatures for a cyclonic
spray tower 107
47 Water-induced venturi scrubber 107
48 Cyclonic spray tower scrubbers Ill
49 Venturi scrubber Ill
50 Doyle impingement scrubber 112
51 Cyclone gas velocity control 115
52 WPPA production 119
53 NSP production 120
54 ROP-TSP production 121
55 GTSP production 121
56 DAP production 122
57 Major gypsum pond equilibrium 124
58 Recommended minimum cross section of dam 124
59 Gypsum pond water seepage control 125
60 Two-stage lime treatment plant 125
61 Species predominance diagram for 0.4 M hydrogen
fluoride solution 126
62 WPPA production trend 142
63 Superphosphate fertilizer consumption from 1966 to
1982 143
64 Ammonium phosphate capacity, production, and plant
utilization projections to 1980 144
65 Superphosphoric acid production trend 147
XI
-------
TABLES
Number Pag<
1 Production Statistics for Phosphate Fertilizer
Plants 2
2 Emission Characteristics for Phosphate Fertilizer
Processes at Average Plants 5
3 Representative Analyses of Commercial Phosphate
Rocks 18
4 Radium (226Ra), Uranium and Thoriun Concentrations in
Phosphate Mine Products and Wastes and Phosphate
Fertilizer Products and Byproducts 18
5 Conversion Factors for Phosphorus Content Units ... 19
6 Common Concentrations of Purified Phosphoric Acid
Grades 19
7 Typical Composition of Filtered WPPA 25
8 Typical Composition of Commercial Phosphoric Acid . .25
9 Equilibrium Concentration Ranges of Gypsum Pond
Water 34
10 Distribution of WPPA Plants by Production Capacity. . 35
11 Distribution of WPPA Plants by County Population
Density 36
12 Composition of Superphosphoric Acid 38
13 Distribution of SPA Plants by
Production Capacity 40
14 Distribution of SPA Plants by County
Population Density 41
15 Properties of Pure Ammonium Phosphates 54
16 Composition of Ammonium Phosphate Raw Materials ... 55
17 1975 Distribution of Ammonium Phosphate Capacity
by State 65
18 Companies Having Ammonium Phosphate Capacity
>200,000 Metric Tons P205 in 1975 65
19 Fluorine Material Balances for WPPA Manufacture ... 70
XII
-------
TABLES (continued)
Number Page
20 Average Stack Heights and Controlled Emission
Factors for Wet Process Phosphoric Acid and
Superphosphoric Acid Plants 73
21 Typical Chemical Composition of Florida Normal
Superphosphate and Triple Phosphate Fertilizer. . . 75
22 Emission Factors for an Average NSP Plant Based on
Controlled Emission Sources 76
23 Emission Factors for an Average ROP-TSP Plant Based
on Controlled Emission Sources 78
24 Emission Factors for an Average GRSP Plant Based
on Controlled Emission Sources 80
25 Emission Factors Developed from Source Test Data
Given in Appendix B 82
26 Values for Xmax anc* Source Severities for Emissions
from an Average Wet Process Phosphoric Acid and
Superphosphoric Acid Plant 86
27 Range of Source Severities and Percentage of Wet
Process Phosphoric Acid and Superphosphoric Acid
Plants Having Severities Greater than 0.05 or 1.0 . 90
28 Maximum Ground Level Concentrations and Source
Severities of Controlled Emission Species from
Average Superphosphate Plants 90
29 Range of Source Severities and Percentage of Plants
Having Severities Greater than 0.05 or 1.0 93
30 Maximum Ground Level Concentration and Severity for
an Average DAP Plant 93
31 Variation in Emission Source Stack Heights 94
32 Severity Distribution Summary 96
33 Total Annual Mass of Emissions from Wet Process
Phosphoric Acid and Superphosphoric Acid Plants . . 97
34 WPPA Industry Contributions to State and National
Atmospheric Emissions 97
35 Superphosphoric Acid Industry Contributions to State
and National Atmospheric Emissions 98
36 Annual Mass of Emissions from Superphosphate Plants
in the United States 99
37 Contribution to State Particulate Emissions Burdens
Due to Emissions from NSP Plants 99
Xlll
-------
TABLES (continued)
Number Page
38 Contribution to State Particulate Emissions Burdens
Due to Emissions from ROP-TSP Plants 100
39 Contribution to State Particulate and SO Emissions
Burdens Due to Emissions from GTSP Plants 100
40 Estimated Mass of Particulate Emissions from
Ammonium Phosphate Plants 101
41 Affected Population Values for Emissions from Wet
Process Phosphoric Acid and Superphosphoric Acid
Plants 103
42 Affected Population Values from Superphosphate
Plants .103
43 Affected Population Values from Ammonium Phosphate
Plants 104
44 Description of Phosphate Fertilizer Complexes in
the United States by Unit Operations 117
45 Typical Equilibrium Composition of Gypsum Pond Water.123
46 Reaction of Gypsum Pond Water with Lime 127
47 Laboratory Data for Phosphorus and Fluoride Removal
at Higher pH 127
48 Removal of 226Ra by Lime Treatment 127
49 Laboratory Process Water Treatment Study 128
50 Effect of Lime Treatment on Radioactivity Removal
from Effluents from a WPPA Plant 129
51 Water Effluent Disposal and Containment Practices
for the Phosphate Fertilizer Industry 131
52 Wastewater Discharge Data for Phosphate Fertilizer
Plants 133
53 Hazard Factors 135
54 Source Severities for Wastewater Discharges at
Individual Phosphate Fertilizer Complexes 136
55 Analysis of Solids from WPPA 138
56 Distribution of Phosphatic Fertilizer Materials . . .145
57 Plants Identified as No Longer Operating in 1977. . .145
A-l 1975 Production of Ammonium Phosphates 159
A-2 WPPA Plants in the United States in 1975 160
A-3 Superphosphoric Acid Plants in the United States in
1975 Which Derive Their Product from WPPA .... .162
A-4 NSP Plants in the United States in 1976 163
xiv
-------
TABLES (continued)
Number Page
A-5 TSP Plants in the United States in 1976 164
A-6 Ammonium Phosphate Plants in the United States in
1975 165
B-l WPPA Plant Source Test Data for Rock Unloading. . . .166
B-2 WPPA Plant Source Test Data for Rock Transfer and
Charging to Reactor 166
B-3 WPPA Plant Source Test Data for Wet Scrubber System .167
B-4 Superphosphoric Acid Plant Source Test Data for Wet
Scrubber System 167
B-5 Plant Source Test Data for ROP-TSP Manufacture. . . .168
B-6 Plant Source Test Data for NSP Manufacture 169
B-7 Plant Source Test Data for GTSP Manufacture 170
B-8 Stack Heights for NSP Plants 172
B-9 Stack Heights for GTSP Plants 172
B-10 Stack Heights for ROP-TSP Plants 172
B-ll Sampling Emissions Data for Ammonium Phosphate
Manufacture 173
D-l NEDS Emission Summary by State 179
D-2 State Listing of Emissions as of July 2, 1975 . . . .180
xv
-------
ABBREVIATIONS AND SYMBOLS
AAQS — ambient air quality standard
APP — ammonium polyphosphate
BPL — bone phosphate of lime or tricalcium phosphate
CD — concentration of particular pollutant, g/m3
D — distance downwind from source, m
DAP — diammonium phosphate
e - 2.72
ED/_ — emission factor for dryer/cooler
Ep — emission factor for product sizing and material
transfer
ER/A — emission factor for reactor/ammoniator-granulator
E o — composite emission factor
E_ — emission factor for total plant
F — hazard factor
GNSP — granular normal superphosphate
GTSP — granular triple superphosphate
h - emission height, m
MAP — monoammonium phosphate
N_/fl — number of emission factors for dryer/cooler
NEDS — National Emissions Data System
NOX — nitrogen oxides
Np — number of emission factors for product sizing and
material transfer
N-P-K — nitrogen-phosphorus-potassium fertilizer
N , — number of emission factors for reactor/ammoniator-
' granulator
NSP — normal superphosphate
N — number of emission factors for total plant
P205 — phosphorus pentoxide, used to express the phosphorus
content of fertilizer
xvi
-------
ABBREVIATIONS AND SYMBOLS (continued)
ppb — parts per billion
Q — mass emission rate, g/s
R — amount of rock required to produce 1 metric ton of NSP
ROP — run of pile
ROP-NSP — run-of-pile normal superphosphate
ROP-TSP — run-of-pile triple superphosphate
S — source severity
SOX — sulfur oxides
SPA — superphosphoric acid
t — averaging time, min
t — instantaneous averaging time, 3 min
TLV — threshold limit value
TSP — triple superphosphate
TVA — Tennessee Valley Authority
u — national average wind speed, 4.5 m/s
V — wastewater effluent flow rate, m3/s
VR — volumetric flow rate of receiving body above plant
discharge, m3/s
x — downwind dispersion distance from a source of
emissions
WPPA — wet process phosphoric acid
Xn , — downwind distance from an emission source at which
°'1 x/F=0.1
X, . — downwind distance from an emission source at which
1>0 x/P-1.0
TT — 3.14
a — 0.2089 xo.903i
a — 0.113 x°-911
z
X — downwind ground level concentration at reference
coordinate with emission height of h
)( — time-averaged ground level concentration
X — instantaneous maximum ground level concentration
"•max
X__ — time-averaged maximum ground level concentration
ITlclX
xvi i
-------
SECTION I
INTRODUCTION
The phosphate fertilizer industry converts insoluble phosphate
rock into water soluble fertilizers that are rich in phosphorus
and readily available for plant uptake. For this program, the
phosphate fertilizer industry is considered to include the produc-
tion of phosphoric acid by the wet process (reaction of phosphate
rock with sulfuric acid), the concentration of phosphoric acid to
superphosphoric acid, the production of normal and triple super-
phosphates, and the manufacture of granular ammonium phosphates.
Phosphoric and superphosphoric acids serve as intermediates in
the production of final fertilizer materials.
Historically, phosphate fertilizers have been one of the large
volume chemicals produced in the United States. Production is
concentrated in the state of Florida because of its extensive
phosphate rock deposits. Until the early 1960's, superphosphates
were the primary phosphate fertilizer material manufactured, but
now ammonium phosphates predominate because of their higher over-
all nutrient content.
During phosphate fertilizer production, air emissions, water
effluents, and solid residues are released into the environment.
This assessment document characterizes these discharges and
evaluates their potential environmental impact. The report
contains a source description that defines process operations,
process chemistry, plant production and capacity, and industry
locations. Emission points are identified, emission species are
characterized, and average emission rates are determined, all on
a multimedia basis. Present and emerging control technologies
are also considered in terms of their effectiveness, advantages/
disadvantages, and extent of application. The final section of
the report discusses the growth and nature of the phosphate
fertilizer industry.
-------
SECTION 2
SUMMARY
In 1975 the phosphate fertilizer industry in the United States
consumed 26.1 x 106 metric tons of phosphate rock to produce
approximately 4.89 x 106 metric tons of phosphate fertilizer.
Final products included 0.44 x 106 metric tons of run-of-the-pile
normal superphosphate, 0.90 x 106 metric tons of granular triple
superphosphate, 0.60 x 106 metric tons of ammonium phosphates,
all expressed in terms of their phosphorus pentoxide (P^Os) con-
tent. In addition, 6.29 x 106 metric tons of wet process phos-
phoric acid and 0.506 x 106 metric tons of superphosphoric acid
were manufactured as phosphate fertilizer intermediates.
Phosphate fertilizers are produced at 121 plants located in 28
states. The number of plants producing each compound and the
average production rates are given in Table 1. Approximately
30% of the plants are complexes producing more than one phosphate
material. These same plants account for the majority of produc-
tion volume. Florida, because of its large phosphate rock de-
posits, is the leader in number of plants (i.e., 16) and tonnage
of materials manufactured.
TABLE 1. PRODUCTION STATISTICS FOR PHOSPHATE FERTILIZER PLANTS
Number of Average plant production rate,
Product plants metric tons/yr (P20s basis)
Wet process phosphoric
acid 36 175,000
Superphosphoric acid 9 56,200
Ammonium phosphate 48 75,000
Normal superphosphate 66 6,650
Granular triple
superphosphate 13 69,100
Run-of-the-pile triple
superphosphate
Total industry
10
a
121
59,700
b
NA
a
Some plants produce more than one product.
b
Not applicable.
-------
Phosphate fertilizer production begins with phosphate rock con-
taining 30% to 35% PaOso This rock is crushed and mixed with
aqueous sulfuric acid to produce phosphoric acid (28% to 32%
PaOs). The reaction takes place in an attack vessel? in addition
to phosphoric acid, insoluble calcium sulfate dihydrate (gypsum)
and fluorine compounds are produced. Precipitated gypsum is
filtered from the acid, sluiced with recycled pond water, and
pumped to a gypsum pond. Fumes from the attack vessel are vented
to a packed-bed wet scrubber for fluoride removal before they are
vented to the atmosphere. The low quality (28% to 32% Pa05) acid
is concentrated to 54% P20S by evaporation.
Superphosphoric acid (Pa03 greater than or equal to 66%) is pro-
duced by further concentrating the 54% wet process phosphoric
acid using either vacuum evaporation with heat transfer surfaces
or submerged combustion/direct heating. All processing steps are
vented to a common scrubber system to remove fluorides and parti-
culates. Gypsum pond water is used as the scrubbing liquid and
then returned to the pond.
The term normal superphosphate is used to designate a fertilizer
material containing from 16% to 21% PaOs made by reacting ground
phosphate rock and sulfuric acid. Rock and acid are mixed in a
reaction vessel, held in an enclosed area (den) during the solidi-
fication process, and transferred to a storage pile for curing.
Cyclones and baghouses are used to control particulate emissions
from rock processing operations; scrubbers are used to reduce
fluoride and particulate emissions from the reactor and den.
However, no controls are normally employed on the curing building
because of the lower level of emissions and typically small
plant size.
Triple superphosphate designates a fertilizer material having a
PaOs content of over 40% made by reacting phosphate rock and
phosphoric acid. There are two principal types of triple super-
phosphate: run-of-the-pile and granular. Run-of-the-pile mate-
rial is essentially a nonuniform pulverized mass produced in a
manner similar to that used for normal superphosphate production.
In the production of granular triple superphosphate, a liquid
mixture of rock and acid is distributed onto a bed of recycled
fines in a granulator to produce a hard, uniform, pelletized gran-
ule. Cyclones, baghouses, and scrubbers are used to control par-
ticulate and fluoride emissions from the various processing steps.
In the manufacture of ammonium phosphates, phosphoric acid and
ammonia are initially reacted in a preneutralizer to an ammonia/
phosphoric acid mole ratio of approximately 1.4. The resulting
slurry passes to an ammoniator-granulator, where the injection
of additional ammonia causes further solidification. Ammonium
phosphate granules are then dried, cooled, screened, and sent to
product shipment. Exhaust streams from the preneutralizer and
ammoniator-granulator pass through a primary scrubber in which
phosphoric acid removes ammonia and particulate. Exhaust gases
-------
from the dryer, cooler, and screen go to cyclones for particu-
late removal. Materials collected in the primary scrubber and
cyclones are returned to the process. The exhaust is sent to
secondary scrubbers where recycled gypsum pond water is used as
a scrubbing liquid to control fluoride emissions. The scrubber
effluent is returned to the gypsum pond.
A summary of air emissions for the six production processes is
presented in Table 2. For each emission point, the emission
species and emission factors are reported. In addition to the
process emissions at phosphate fertilizer plants, fluorine in
the gypsum pond water is volatilized and emitted to the atmos-
phere as some form of fluoride.
In order to help evaluate the potential environmental impacts of
air emissions and water effluents, certain criteria were used:
source severity, affected population, and state and national
emission burdens. The intent was to compare the relative impacts
of a large number of source types studied. In evaluating poten-
tial environmental effects, average parameters have primarily
been employed (e.g., emission factors, stack heights, population
densities). A more detailed plant-by-plant evaluation was beyond
the scope of the project and conclusions are not drawn with re-
gard to actual environmental impacts at specific sites. In some
cases, hazard factors used in the evaluation may be conservative
due to a lack of more definitive health effects data.
Source severity (S) for air emissions compares the time-averaged
maximum ground level concentration of an emitted pollutant, Xmax>
to an estimated hazard factor, F, and is defined as x"max/F•
Values of x~max were calculated from average plants from accepted
plume dispersion equations and the emission factors in Table 2.
The hazard factor, F, is defined as the primary ambient air
quality standard (AAQS) for criteria pollutants (particulates and
sulfur dioxide). For fluoride and ammonia emissions, F is
defined in terms of the reduced threshold limit value (TLV®):
F = TLV(8/24)(1/100), where the factor 8/24 corrects for 24-hr
exposure and 1/100 is a safety factor. Calculated source sever-
ity values are shown in Table 2.
Values for x"max could not be determined for hydrogen fluoride
emissions from gypsum ponds. Instead, plume dispersion equations
were used to determine the distances downwind from the pond at
which the time-averaged pollutant concentration, x~, divided by
F was below 1.0 and 0.05.
The potential environmental impact was also measured by deter-
mining the population around a plant exposed to a contaminant
concentration exceeding an acceptable level. The affected
population is defined as the number of persons living in the
area around an average plant where 7 divided by F is greater
than 1.0. Plume dispersion equations are used to find this
-------
TABLE 2. EMISSION CHARACTERISTICS FOR PHOSPHATE FERTILIZER PROCESSES AT AVERAGE PLANTS
Process
Phosphoric acid
Superphosphoric acid
Ammonium phosphate
Normal superphosphate
Run-of-the-pile triple
superphosphate
Granular triple
superphosphate
Fertilizer complex
Emission point
Rock unloading
Rock transfer
and storage
Wet scrubber
system
Wet scrubber
Total process c
emissions
Rock unloading
Rock feeding
Mixer and den
j
Curing building
Rock unloading
Rock feeding
Cone mixer, den, and
curing building
Rock unloading
Rock feeding
Reactor, granulator
dryer, cooler, and
screens
Curing building
Gypsum pond
Emission species
Particulate
Particulate
Particulate
Fluoride
Sulfur oxides
Particulate
Fluoride
Particulate
Fluoride
Ammonia
Particulate
Particulate
Particulate
Fluoride
Particulate
Fluoride
Particulate
Particulate
Particulate
Fluoride
Particulate
Particulate
Particulate
Fluoride
Sulfur oxides
Particulate
Fluoride
Fluoride
Controlled
emission factor,
g/kg P2OS
0.15
0.045
0.054
0.010
0.032
0.011
0
1.5
0.038
0.068
0
0.055
0.26
0.10
3
1.9
0
0.014
0.16
0.10
0
0.017
0.05
0.12
1
0.10
0.018
0
(0.025
± 250%
± 180%
± 164%
± 47%
± 200%
to 0.055
.0073b
± 69%
± 30%
± 75%
.28b
± 180%
i 86%
± 120%
. 6^
± 120%
.07"
± 170%
± 50%
± 40%
.09b
± 180%
± 320%
± 30%
.866
± 240%
± 40%
.50 f
to 2.5)
a
Source severity
0.41
0.040
0.025
0.18
0.011
0.01
0.09
0.43
0.44
0.09
0.02
0.004
0.013
0.18
0.35
7.2
0.04
0.009
0.03
0.77
0.06
0.01
0.004
0.36
0.11
0.02
0.12
1.0 ' @D = 1300 m
0.05 @D = 6700 m
Affected popu-
lation, persons
S > 1.0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
539
0
0
0
0
0
0
0
0
0
0
0
0
S > 0.05
64
2
0
159
0
0
28
288
285
41
0
0
0
529
519
13,021
5
0
0
1,178
15
0
0
1,356
307
0
161
5,532
aSeverity for fluoride based on TLV for hydrogen fluoride of 2.0 mg/m3; severity for ammonia based on its TLV of 18 mg/m3.
Only two data points,
cAverage process controlled by integrated control system with a single emission point.
dUncontrolled emission factors because curing building emissions are not normally controlled.
Worst case estimate based on fuel oil sulfur content.
fEmissions from gypsum pond are uncontrolled and vary wideiy depending on pond conditions.
-------
area, which is then multiplied by an average population density
to determine the affected population. Due to uncertainties in-
herent in sampling and dispersion modeling methodologies, the
number of persons exposed to a x/F ratio greater than 0.05 is
also reported. Values for the affected population are reported
in Table 2.
Another measure of potential environmental impact is the total
mass of industry emissions of each criteria pollutant. These
values were compared to total state and national emissions from
all sources to find the emissions burden due to various segments
of the phosphate fertilizer industry. The percent contributions
to states' emissions burdens by wet process phosphoric acid
plants ranged from 0.004% to 0.4% for particulates and from
0.0002% to 0.02% for sulfur dioxide. On a national basis, wet
process phosphoric acid plants contributed 0.01% of the nation's
particulate burden and less than 0.001% of the sulfur dioxide
burden. Particulate emissions from superphosphoric acid plants
contributed from less than 0.001% to 0.005% of each state's
emissions burden. For normal and triple superphosphate produc-
tion, it was found that in each state and on a national basis the
particulate and sulfur oxide (SOx) contribution to the respective
emissions burden was less than 0.001%. Ammonium phosphate partic-
ulate emissions represent approximately 0.02% of the total nation-
al particulate emissions burden from all sources. On a statewide
basis, ammonium phosphate production contributed 0.1% or more of
the total statewide particulate emissions burden in only Florida
(0.8%), Idaho (0.4%) , and Louisiana (0.3%).
Environmental and economic concerns have prompted use of control
devices in most facets of the wet process phosphoric and super-
phosphoric acid industry, with the exception of volatile emis-
sions from the gypsum pond. Rock unloading, rock transfer, and
rock charging operations are located in partially enclosed struc-
tures with ventilation systems venting to baghouses for rock
recovery. Vaporous and particulate emissions issuing from the
attack vessel, filtration system, and clarifier are all vented
to a common venturi thread packed-bed wet scrubber. Recycled
pond water is used in the scrubber to remove emission species
and is then sent back to the gypsum pond. A similar wet scrub-
bing system is used as superphosphoric acid plants to remove
fluoride and particulate emission species.
The types of air pollution control equipment used at superphos-
phate plants are varied; however, all plants have a basic emis-
sions control system consisting of cyclones, baghouses, and wet
scrubbers. All plants use cyclones and/or baghouses to control
particulate emissions from the rock unloading and rock feeder
systems. Wet scrubbers are used to control particulate and
fluoride emissions from the mixer den, curing building, reactor,
granulator, dryer, and cooler. These scrubbers also control SOX
emissions from the dryer at granular triple superphosphate plants
when fuel oil is used. Only the fluoride emissions from the
-------
curing and storage building at normal superphosphate plants
are uncontrolled„
Stack emission from all ammonium phosphate plants have some type
of emission control. Cyclones are used for product recovery, and
wet scrubbers are used for ammonia (NHa), fluoride, and product
recovery.
Based on industry production trends and forecasts, production of
wet process phosphoric acid and superphosphoric acid are expected
to increase at annual rates of 4% to 7% and 7% to 10%, respec-
tively. Normal superphosphate production is expected to decline
by 1% to 5% until about 1982 when industry production is expected
to stabilize. Triple superphosphates, both granular and run-of-
the-pile, are expected to maintain a moderate annual growth rate
of 2%. Ammonium phosphate production from 1975 to 1980 is pro-
jected to grow at an annual rate of 7.5%, resulting in approxi-
mately 44% more production in 1980 than in 1975. If the current
level of emission control is maintained, emissions from these
production processes will increase or decrease in a similar
fashion.
Sources of process wastewater from wet process phosphoric acid
production include wet scrubber liquor, gypsum slurry water, and
barometric condensers. Gypsum pond water normally supplies most
of the water requirements for operation of wet scrubbers and
barometric condensers and also for transferring the waste gypsum
to a disposal area although variations do exist. Acid sludge,
generated in acid clarification, contains substantial amounts of
phosphate and is normally disposed of by blending into dry ferti-
lizer. Cooling water may be recirculated gypsum pond water. If
supplied by a segregated nonprocess system instead, it may be re-
cycled or discharged. Steam condensate which is contaminated,
such as that from barometric condensers and vacuum ejectors, is
discharged to the gypsum pond. Uncontaminated steam condensate
is discharged to receiving waters without treatment. Wastewater
streams contain varying quantities of phosphoric acid (H3PCU),
fluorides, sulfates, and gypsum.
Wastewater streams at superphosphoric acid plants come from baro-
metric condensers, steam jet ejectors, and wet scrubbers. These
streams contain quantities of H3POi» and fluorides. Wastewater
from superphosphoric acid plants is normally contained in a man-
ner similar to that used at wet process phosphoric acid plants.
The only source of wastewater at normal and triple superphosphate
fertilizer plants is the scrubber liquors. Scrubber systems use
recycled water from the gypsum ponds or other holding reservoirs.
Nearly all triple superphosphate plants are located at fertilizer
complexes producing wet process phosphoric acid and, as a result,
use gypsum pond water in their scrubber systems. More than 60%
of normal superphosphate plants now practice fluorine recovery
and thereby eliminate or greatly reduce the need for a pond.
-------
Plants recovering fluosilicic acid consume the small amount of
silica-containing liquid waste generated as a filler in ferti-
lizer production.
Ammonium phosphate production facilities occasionally use second-
ary wet scrubbers to remove fluorides and other contaminants from
process gas streams after preliminary scrubbing with a weak phos-
phoric acid solution for ammonia recovery. Secondary scrubbers
use recycled water from gypsum ponds or other holding reservoirs.
In a study of over 70% of the plants in the phosphate fertilizer
industry, nearly 75% reported no discharge of process wastewater.
Of the 15 plants that reported a discharge, 12 reported a dis-
charge only when necessitated by excessive rainfall. Several of
these reported that they have not treated or discharged water for
several years. In actual practice, discharge of contaminated
process water from the recycle pond system is held to an absolute
minimum due to treatment costs.
One plant was found to use river water on a once through basis
for scrubbing air emissions and for cooling. Effluent from this
plant is discharged without treatment.
Available wastewater discharge data from seven plants on file as
of October 1976 at the Florida Department of Environmental Regu-
lation were collected and analyzed by means of a water source
severity relationship.
Source severity for water effluents compares the concentration
of a particular pollutant after discharge and dilution in the
receiving body with an estimated allowable concentration denoted
as the hazard factor.
In determining the source severity of a plant, the discharge
quantity is compared to the receiving body flow rate times the
hazard factor according to the following equation:
v C
s - D P
-
where S = source severity for a particular pollutant
VD = wastewater effluent flow rate, m3/s
CD = concentration of particular pollutant, g/m3
VR = volumetric flow rate of receiving body above
plant discharge, m3/s
F = hazard factor for particular pollutant, g/m3
-------
Severities for fluoride, phosphorus, and to a lesser degree
ammonia-nitrogen in discharged waters were found in a number of
cases to be above 1.0. This was due to the extremely low flow
rates of the receiving bodies and should represent a worst case
analysis for the small number of plants that do discharge.
Solid residues generated at phosphoric acid plants are gypsum
from the filtration of wet process phosphoric acid, wet process
phosphoric acid sludge, and solids suspended in the wet scrubber
liquor. These solid waste residues are, for the most part
stored in ponds, stacked in piles, or stored in mining pits on
site. A small percentage (approximately equal to 1%) is used
as a raw material for various products. Under normal conditions,
the solid residues cause no adverse environmental effects. At
normal and triple superphosphate plants, solid residues are in
the form of slurries from the wet scrubber and are therefore
included with wastewater treatment practices.
-------
SECTION 3
SOURCE DESCRIPTION
A. OVERVIEW OF PHOSPHATE FERTILIZER INDUSTRY
Phosphorus is one of the major elements essential for normal
plant growth (1). Naturally occurring phosphorus in phosphate
rock in the form of tricalcium phosphate is almost completely
insoluble in water (solubility in cold water equals 20 g/m3
of water) (2). To enhance plant growth, the phosphate fertilizer
industry converts insoluble phosphate rock into water-soluble
fertilizer products.
1. Phosphate Rock Consumption in the United States
In 1975, 44,286,000 metric tons of phosphate rock were mined in
16 states in the United States, as shown in Figure 1 (3). Phos-
phate rock mined in Florida accounted for approximately 78% of
the U.S. production and about 29% of the total world's supply in
1975. Over 92% of this output came from the vast sedimentary
land pebble deposit in Polk and Hillsborough counties east of
Tampa, Florida. Approximately 5.7% of the phosphate rock was
mined in Tennessee and 3.6% in North Carolina. Deposits in
Tennessee are classified as brown, white, and blue rock; only
the brown rock has been of commercial importance. Phosphate
rock mined in the western states of Idaho, Montana, Wyoming, and
Utah accounts for about 14% of the total ore mined in the
United States (3).
1 metric ton equals 106 grams; conversion factors and metric
system prefixes are presented at the end of this report.
(1) Riegel's Handbook of Industrial Chemistry, Seventh Edition.
J. A. Kent, ed. Van Nostrand Reinhold Co., New York, New
York, 1974. pp. 551-569.
(2) Handbook of Chemistry and Physics, 49th Edition, R. C. Weast,
ed. The Chemical Rubber Co., Cleveland, Ohio, 1968.
p. B-187.
(3) Stowasser, W. F. Phosphate-1977. Publication No. MCP-2,
U.S. Department of the Interior, Bureau of Mines, Washington,
D.C., May 1977. 18 pp.
10
-------
NUMEROUS SCATTERED FIELDS
OF PHOSPHATE ORES
Figure 1. Location of major phosphate rock
deposits in the United States (3).
Approximately 31,029,000 metric tons (70%) of the phosphate rock
mined in 1975 were used in the United States to produce numerous
phosphorus-containing materials (3). Figure 2 illustrates the
1975 consumption pattern for the various products obtained from
phosphate rock. Significant quantities (15.9%) of phosphate rock
were consumed in several nonagricultural markets such as the pro-
duction of detergent builders and water treatment chemicals and
the treatment of aluminum and ferrous metal surfaces, as well as
in foods, beverages, pet foods, dentifrices, and fire control
chemicals.
Agriculture-related industries producing phosphate fertilizers
and animal feeds used 26,096,000 metric tons (84.1% of total pro-
duction) of phosphate rock in 1975. Of this total, 22,754,000
metric tons (89.7%) were consumed for fertilizers, and 2,688,000
metric tons (10.3%) were used to produce animal feeds.
2. Types of Fertilizer Products
The schematic diagram of the phosphate fertilizer industry pre-
sented in Figure 3 (4) shows the conversion of insoluble phos-
phate ore into the soluble form necessary for plant consumption.
Phosphate-bearing rock is mixed with sulfuric acid (H2SOt,) to
produce phosphoric acid, the building block for phosphate
fertilizers.
As Figure 3 illustrates, numerous additional processes are used
to produce phosphate fertilizer materials. These processes are
in operation because of farmer demand for a wide variety of
fertilizer mixtures.
(4) Fullam, H. T., and B. P. Faulkner. Inorganic Fertilizer and
Phosphate Mining Industries—Water Pollution and Control
(PB 206 154). Grant 12020 FPD, U.S. Environmental Protec-
tion Agency, Cincinnati, Ohio, September 1971. 225 pp.
11
-------
PHOSPHATE ROCK
31.029« ID3 metric tons/yr
INDUSTRIAL
t.m*vf
metric tons/yr
15.9*
HOSPHORUS
14.103
1.3*
1
ANHYDROUS
DERIVATIVES
279 > 103
metric tonslyr
ftEM
PHOSP
metric
14.
ELEW
JflAL
HORUS
xlO3
sns/yr
S*
NTAL
PHOSPHORUS
620 « ID3
metric tons/yr
1
FURNACE
PHOSPHORIC ACID
3.63U103
metric tons/yr
DEaUORINATED
ROW
metric tons/yr
1.5*
AGRICULTURE
26.096« 103
metric tons/yr
84.1*
NORMAL
SUPERPHOSPHATE
metric tons/yr
'•"•
SODIUM
IRIPOLYPHOSPHATE
2.327U03
metric tons/yr
7.5*
OICALCIUM
PHOSPHATE
1.30JI103
metric tons/yr
4.2*
WET PROCESS
PHOSPHORIC ACID
21. 161 < ID3
metric tontfyr
68.2%
i
|
OTHER
8.129X103
metric Bn's/yr
26.2*
DICALCIUM
PHOSPHATE
1. 117 x 103
metric tons/yr
3.6*
3.227
metric
10.
DIAMMONIUM
PHOSPHATE
8.6881 I03
metric tons/yr
?8.0%
x 10 2.2M < 10
ons/yr metric
* 7.
tons/yr
*
TRIPLE
SUPERPH05PHAU
•).«!« Id3
metric tons/yr
I/. 5*
Figure 2. U.S. phosphate rock consumption pattern
for various phosphorus products (3).
H2so4
PHOSPHATE FERTILIZER PRODUCTS
Figure 3. Schematic diagram of phosphate fertilizer industry (4)
12
-------
Product fertilizers differ in the amount and chemical form of the
three primary plant nutrients: nitrogen (N), phosphorus (P), and
potassium (K). Normal and triple superphosphate contain only one
plant nutrient—phosphorus. Ammoniated superphosphate and ammo-
nium phosphates contain two nutrients—phosphorus and nitrogen,
while solid and liquid-mixed fertilizers contain all three nutri-
ents in varying N-P-K ratios.
For evaluative purposes, the phosphate fertilizer industry is
divided into three segments: phosphoric acid and superphosphoric
acid, normal and triple superphosphate, and granular ammonium
phosphate. Ammoniated superphosphates and solid and liquid-mixed
fertilizer segments of the industry were covered in a separate
Source Assessment Document on fertilizer mixing plants (5).
a. Phosphoric Acid and Superphosphoric Acid—
In 1975, 6,979,400 metric tons of phosphoric acid [reported as
equivalent (100%) phosphorus pentoxide (PaOs)] were produced in
the United States (6). Of this total, the 36 plants shown in
Figure 4 (7) produced 90% or 6,291,400 metric tons from phosphate
rock using wet process technology (3, 6). This report does not
cover those plants which produce phosphoric acid from elemental
phosphorus (thermal process) because this high purity acid is no
longer used to produce phosphate fertilizers (8). The phosphate
fertilizer industry consumed 86% or 5,380,648 metric tons of
the wet process acid produced. The remainder (14%) of the wet
process acid was used for preparing phosphatic feed supplements
for livestock and poultry.
Phosphoric acid used in the fertilizer industry is made by the
reaction of aqueous (50% to 98%) sulfuric acid with crushed phos-
phate rock, hence the term "wet process." The reaction occurs
in an attack vessel where, in addition to phosphoric acid,
insoluble calcium sulfate dihydrate (gypsum) and fluorine com-
pounds are produced. Precipitated gypsum is filtered from the
acid, sluiced with recycled .pond water, and pumped to a gypsum
pond. Fumes from the attack vessel are vented to a packed-bed
wet scrubber for fluoride removal before they are exhausted to
(5) Rawlings, G. D., and R. B. Reznik. Source Assessment:
Fertilizer Mixing Plants. EPA-600/2-76-032c, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, March 1976. 187 pp.
(6) Inorganic Chemicals 1976. M 28A(76)-14, U.S. Department of
Commerce, Washington, D.C., August 1977. 30 pp.
(7) Hargett, N. World Fertilizer Capacity-Computer Printout.
Tennessee Valley Authority, Muscle Shoals, Alabama, 1976.
(8) TVA Plans Early Closure of Furnaces; Cities Switch to Wet-
Process Phosphoric. Chemical Marketing Reporter, 209(3),
1976.
13
-------
• PHOSPHORIC ACID PLANTS
* PHOSPHORIC AND SUPERPHOSPHORIC
ACID PLANTS
Figure 4. Location of wet process and
superphosphoric acid plants (6).
the atmosphere. Low quality (28% to 30% P205 equivalent) phos-
phoric acid is then concentrated to 54% P20s equivalent by evapo-
rating water from the solution.
Superphosphoric acid (P20s equivalent greater than or equal to
66%) is produced by further concentration of the 54% P2O5 phos-
phoric acid. Superphosphoric acid concentration is accomplished
by either vacuum evaporation employing heat transfer surfaces or
submerged combustion/direct heating. In 1975, approximately
505,900 metric tons of superphosphoric acid were produced by
nine plants in six states in the United States, as shown in
Figure 4 (6, 7) (Appendix A).
b. Normal and Triple Superphosphate—
Normal superphosphate (NSP), prepared by reacting ground phos-
phate rock with sulfuric acid, contains 16% to 22% available
P2O5- Approximately 0.44 x 106 metric tons (P205 equivalent of
NSP fertilizer were produced in 1975 (9).
Triple superphosphate (TSP), containing 45% to 55% available
P205, is made by reacting ground phosphate rock with phosphoric
acid. Two types of TSP are produced: run of the pile and granu-
lar. In 1975 approximately 0.60 x 106 metric tons (P205 equiva-
lent) , of run-of-the-pile triple superphosphate (ROP-TSP and
(9) Inorganic Fertilizer Materials and Related Products. .
M28B(75)-13, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
14
-------
0.90 x 106 metric tons of granular triple superphosphate (GTSP)
were produced in the United States (9).
Geographical locations of the 66 NSP plants in the United States
are shown in Figure 5 (7). NSP plants are located near consumers
because it is cheaper to ship phosphate rock (approximately equal
to 33% P20s) to consumption areas than it is to ship NSP from the
ore deposits. A description of each NSP plant is given in
Appendix A.
The production of TSP, unlike that of NSP, occurs in plants
located near phosphate rock deposits (Figure 6) (10) . Eleven of
the sixteen TSP plants are located in Florida, which accounts for
approximately 78% of the U.S. production of phosphate-bearing
rock. Among the 16 plants, 7 have facilities for producing both
run-of-the-pile and granular grades of products; of the remaining
9 plants, 6 produce only GTSP and 3 produce only ROP-TSP (see
Figure 6). Each of these plants is also described in Appendix A.
c. Ammonium Phosphate—
Ammonium phosphates are produced by reacting phosphoric acid with
anhydrous ammonia. Both solid and liquid ammonium phosphate
fertilizers are produced in the United States. In 1975, approxi-
mately 2.8 x 106 metric tons (PzOs equivalent) of ammonium phos-
phates were produced by 48 plants located in 17 states, as shown
in Figure 7 (7, 10, 11).
3. Raw Materials
Raw materials used in the phosphate fertilizer industry consist
of phosphate rock, sulfuric acid, and anhydrous ammonia. Phos-
phate rock is a term broadly used to denote the group of minerals
commercially valuable for their phosphorus content. The princi-
pal (greater than 80%) mineral constituent of phosphate rock is
fluorapatite, [Ca3(POU)2]3«CaF2 (12). Also found in phosphate
rock are iron oxides, aluminum oxides, magnesium, carbonates,
carbon dioxide, calcium oxide, silicon oxides, and sulfates. A
chemical analysis of phosphate rock samples from mines across the
(10) Harre, E. A., M. N. Goodson, and J. D. Bridges. Fertilizer
Trends 1976. Bulletin Y-lll, National Fertilizer Develop-
ment Center, Tennessee Valley Authority, Muscle Shoals,
Alabama, March 1977. 45 pp.
(11) Final Guideline Document: Control of Fluoride Emissions
from Existing Phosphate Fertilizer Plants. EPA-450/2-77-005
(PB 265 062) , U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1977. 277 pp.
(12) Atmospheric Emissions from Wet-Process Phosphoric Acid
Manufacture. AP-57 (PB 192 222), U.S. Department of Health,
Education, and Welfare, Raleigh, North Carolina, April 1970.
86 pp.
15
-------
Figure 5. Location of NSP plants in
the United States (7) .
A GTSP
O ROP-TSP
A GTSP AND ROP - TSP
Figure 6. Location of TSP plants in
the United States (10) .
16
-------
Figure 7. Location of ammonium phosphate plants
in the United States (7, 10, 11).
United States is shown in Table 3 (13). Trace amounts of arsenic,
lead, vanadium, and chromium which may be present in the rock are
not listed. Uranium is also present in phosphate rock, with con-
centrations in the range of 40 g to 165 g of uranium per metric
ton of rock. Table 4 (14) gives typical concentrations of radio-
active elements in Florida phosphate mine products and wastes and
phosphate fertilizer products and wastes.
Phosphorus content of the rock and/or products is commonly
expressed in one of four ways:
• BPL [bone phosphate of lime or tricalcium phosphate,
Ca3(P01+)2] .
• Phosphorus pentoxide (P2C>5).
• Elemental phosphorus
• Phosphoric acid d^P
(13) Lowenheim, F. A. Phosphorus Compounds, Inorganic. In:
Encyclopedia of Industrial Chemical Analysis, Volume 17,
F. D. Snell and L. S. Ettre, eds. John Wiley & Sons, Inc.,
New York, New York, 1973. pp. 142-144.
(14) Guimond, F. J., and S. T. Windham. Radioactivity Distribu-
tion in Phosphate Products, By-Products, Effluents, and
Wastes. ORP/CSD-75-3, U.S. Environmental Protection Aqency,
Washington, D.C., August 1975. 30 pp.
17
-------
TABLE 3. REPRESENTATIVE ANALYSES OF COMMERCIAL
PHOSPHATE ROCKS (13)
(percent reported as material shown)
Reprinted from Encyclopedia of Industrial Chemical Analysis, Vol. 17, by
courtesy of John Wiley & Sons, Inc.
U.S. location and type
Florida:
Land pebble, high grade
Hard rock, high grade
Hard rock, waste pond
Tennessee :
Brown rock, high grade
Western states:
Phosphoric rock, high grade
Phosphoric rock, low grade
Pj05
35.5
35.3
23.0
34.4
32.2
19.0
CaO
48.8
50.2
28.5
49.2
46.0
23.3
MgO
0.04
0.03
0.4
0.02
0.2
1.4
Al,0,
0.9
1.2
14.8
1.2
1.0
5.9
Fe,0,
0.7
0.9
2.9
2.5
0.8
4.0
SiO,
6.4
4.3
19.8
5.9
7.5
27.4
SO,
2.4
0.1
0.01
0.7
1.7
1.9
P
4.0
3.8
2.1
3.8
3.4
1.8
Cl
0.01
0.005
0.005
0.01
0.02
_b
CO,
1.7
2.8
1.4
2.0
2.1
4.0
Organic
carbon
0.3
0.3
0.3
0.2
1.8
5.0
NajO
0.07
0.4
0.1
0.2
0.5
1.5
K,0
0.09
0.3
0.4
0.3
0.4
1.0
H70*
l.B
2.0
7.0
1.4
2.5
3.5
'After drying at 100°C for several hours. Data not available.
TABLE 4.
RADIUM (226Ra), URANIUM, AND THORIUM CONCENTRATIONS
IN PHOSPHATE MINE PRODUCTS AND WASTES AND
PHOSPHATE FERTILIZER PRODUCTS AND BYPRODUCTS
(pCi/g)b
(14)
Material
Marketable rock
Slimes
Sand tailings
Phosphoric acid
Gypsum
Normal superphosphate
Diammonium phosphate
Triple superphosphate
Monoammonium phosphate
Sodium fluorosilicate
Animal feed
226Ra
42
45
7
1
33
25
5
21
5
0
5
.5
.3
.6
.0
.28
.5
Uranium
234
41
42
5.2
6.2
63
58
55
235
1
2
0
0
3
2
2
.9
.6
.38
.32
.0
.8
.9
238
41
44
5.3
6.0
63
58
55
Thorium
227 228 230 232
2.0 0.61 42.3 0.44
2.3 1.2 48 1.4
_C
0.97 1.4 13 0.27
1.6 0.8 65 0.4
1.2 0.9 48 1.3
Plants using Florida phosphate rock.
'Blanks indicate no data obtained.
Picocuries per gram; 1 picocurie
equals 0.037 becquerel.
Table 5 shows the factors required to convert from one set of
units to another. The common industry practice of reporting all
phosphorus-containing materials in terms of the equivalent phos-
phorus pentoxide (P20s) content is used throughout the remainder
of this document. Table 6 illustrates acid concentrations
reported in various units.
Offsite preparation of phosphate rock involves beneficiation to
remove impurities, drying to remove moisture, and grinding to
improve reactivity.
18
-------
TABLE 5. CONVERSION FACTORS FOR PHOSPHORUS CONTENT UNITS
To
%
%
%
%
%
%
%
%
%
convert from
BPL
BPL
P205
P20s
P205
HsPOit
HaPOit
p
p
To
% P
% P20s
% P
% BPL
% HsPOi,
% P
% P20s
% P205
% BPL
Multiply by
0.1997
0.4576
0.4364
2.1853
1.381
0.316
0.724
2.2914
5.0073
TABLE 6. COMMON CONCENTRATIONS OF PURIFIED PHOSPHORIC ACID GRADES
(percent)
Material HaPO^ P20s P Polyphosphate
Filtered production
phosphoric acid
Orthophosphoric acid
Superphosphoric acid
28
41
75
97
100
20
30
54
70
72
9
13
24
31
31
0
0
0
2.
10
2
Sulfuric acid used in the wet process is either made in a captive
plant or piped from a nearby sulfuric acid manufacturer. Virgin
acid made from brimstone (native sulfur) or pyrites (sulfur bear-
ing ores) is normally used. The use of byproduct sulfuric acid
from other processes may introduce impurities that cause poor
quality gypsum crystal formation and odor problems (15).
4. Rock Preparation
Phosphate rock that has been mined and beneficiated is in general
too coarse to be used directly in acidulation. The major frac-
tion of the phosphate rock (more than 98%) ranges in size from
pebbles 25 mm in diameter down to 100-ym material (4). The rock
is therefore processed through equipment to mechanically reduce
it to the particle size needed for improved reactivity during
the acidulation process (smaller than 150 urn).
Preliminary drying to remove moisture is necessary to prepare the
rock for grinding (Figure 8). Direct-fired rotary kilns 8 m to
30 m long and 2 m to 3 m in diameter are used to dry phosphate
(15) Phosphoric Acid, Volume I, A. V. Slack, ed. Marcel Dekker,
Inc., New York, New York, 1968. 1159 pp.
19
-------
rock (16). These dryers use natural gas or fuel oil as fuel and
are fired countercurrently. In recent years, the fluidized-bed
type of dryer has gained prominent importance because of its fuel
savings and increased throughput.
PHOSPHATE
ROCK
-OIL OR NATURAL GAS
•AIR
EMISSIONS
•-GROUND ROCK
STORAGE
Figure 8. Preparation of phosphate rock for acidulation.
Size reduction is accomplished with ball, roll, or bowl mills.
Rock is fed into the mills and mechanically ground to a fineness
located between the particle size levels of 80% through a 150-ym
and 95% through a 74-ym screen. After the rock enters the mill
system, all flow through the sizing and reclamation circuits is
by pneumatic means. Air is constantly exhausted from the mill
system to prevent precipitation of moisture which is released
from the rock during grinding.
Future rock grinding operations may utilize a wet grinding cir-
cuit rather than the current dry grinding practice. This change
would eliminate the gas effluent streams associated with both
rock drying and grinding operations and result in lower capital
costs (17).
Phosphate rock arrives at the phosphate fertilizer plant in
either a ground or unground form. For economic reasons, the
trend has been toward more processing at the point where the rock
is mined, especially at smaller plants (18).
(16) Heller, A. N., S. T. Cuffe, and D. R. Goodwin. Inorganic
Chemical Industry. In: Air Pollution, Volume III: Sources
of Air Pollution and Their Control, A. C. Stern, ed.
Academic Press, New York, New York, 1968. pp. 221-231.
(17) Martin, E. E. Development Document for Effluent Limitations
Guidelines and New Source Performance Standards for the
Basic Fertilizer Chemicals Segment of the Fertilizer Manufac-
turing Point Source Category. EPA-440/l-74-011-a (PB 238
652), U.S. Environmental Protection Agency, Washington, D.C.,
March 1974. 170 pp.
(18) Caro, J. H. Characterization of Superphosphate. In: Super-
phosphate: Its History, Chemistry, and Manufacture. U.S.
Department of Agriculture, Washington, D.C., December 1964.
pp. 272-284.
20
-------
Ground rock requires tight, fully enclosed material handling
equipment to reduce the loss of rock and prevent excessive air
emissions. General shipping practice includes the use of
enclosed, hopper-bottom railroad cars of the type developed for
hauling cement and other finely ground material. Little (less
than 5%) ground rock is carried by ship or barge because of
handling losses that would be incurred.
In a typical system, ground rock is unloaded from the hopper-
bottom cars into a receiving hopper located directly under the
track. A vibrator is used to keep the rock flowing freely. An
underground screw or belt conveyor carries the rock to storage
silos. A typical rock unloading facility is shown in Figure 9.
The unloading station, transfer conveyors, and storage silos are
enclosed and all ventilation points are equipped with dust
collectors.
EMISSIONS
eeeeeeeeeee
aeeeeeeeeee
BAGHOUSE
ROCK STORAGE
SILO
TO WEIGH
HOPPER
Figure 9. Raw material unloading and storage,
21
-------
B. WET PROCESS PHOSPHORIC ACID PRODUCTION
1. Process Chemistry
In the wet process production of phosphoric acid, sulfuric acid
and the tricalcium phosphate portion of the phosphate rock react
to form phosphoric acid and gypsum (17).
Ca3(POit)2 + 3H2S04 + 6H20 —»• 2H3POtt + 3(CaS04•2H20) (2)
This chemistry is straightforward; however, two factors influence
operating conditions at individual plants: the composition of
the phosphate rock and the physical form of the byproduct calcium
sulfate.
a. Effects of Phosphate Rock Composition—
Side reactions occur during acidulation, and the quantity of prod-
ucts found depends on the amounts and composition of other chemi-
cal constituents in the phosphate rock (see Table 3). These
generally undesirable side reactions form precipitates and
sludges which foul operating, handling, transfer, and storage
equipment (19). Excessive amounts of impurities also increase
acid viscosity, which affects handling operations. Metals such
as iron, aluminum, and magnesium form water-insoluble phosphate
salts, which tie up useful phosphate and remain as suspended
solid impurities in product acids. Trace metals (arsenic, lead,
and heavy metals) also contaminate the acid. Carbonates, fluo-
rine, and silica likewise are troublesome materials (19). Carbon-
ates react with sulfuric acid to produce carbon dioxide, which
contributes to foaming. The calcium fluoride constituent of the
fluorapatite ore,reacts with sulfuric acid to produce hydrogen
fluoride according to the following reaction:
CaF2 + H2SOl+ —> 2HF + CaS04 (3)
In addition, calcium fluoride reacts with phosphoric acid accord-
ing to the following reaction (17, 20):
CaF2 + 2H3POI| —> Ca(H2POu)2 + 2HF (4)
The hydrogen fluoride can evolve as a gas or react with silica in
the following manner (17, 20):
(19) Dahlgren, S. E. Chemistry of Wet-Process Phosphoric Acid
Manufacture. In: Phosphoric Acid, Volume I, A. V. Slack,
ed. Marcel Dekker, Inc., New York, New York, 1968.
pp. 91-154.
(20) Evaluation of Emissions and Control Techniques for Reducing
Fluoride Emissions from Gypsum Ponds in the Phosphoric Acid
Industry. Contract 68-02-1330, Task 3, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
November 1976. 218 pp.
22
-------
Si0
6HF — »• H2SiF6 + 2H20
(5)
During acid concentration steps, fluosilicic acid (H2SiF6) in the
phosphoric acid solution can dissociate according to the follow-
ing reaction (17, 20):
H2SiF6
2HF
(6)
Fluosilicic acid can also combine with sodium or potassium to
yield fluosilicate salts, which form scale and sludge in the proc-
essing equipment.
b. Physical Form of Calcium Sulfate—
The popular process for phosphoric acid production is based on
the quick formation of calcium sulfate dihydrate or gypsum
(CaS04-2H20). It is also possible to precipitate calcium sulfate
as the hemihydrate (CaSO^•1/2H20) or the anhydrite (CaS04). The
dihydrate processes offer basic advantages—less severe operating
conditions, lower rates of corrosion, better filterability, and
lower capital cost—which outweigh advantages in the hemihydrate
and anhydrite processes. An alternative dihydrate process which
does not involve direct formation of the dihydrate utilizes the
initial formation of calcium sulfate in the hemihydrate form
and its subsequent hydration to gypsum. Figure 10 (21) shows the
precipitation of calcium sulfates in phosphoric acid.
120
<
100-
80
60
HEMIHYDRATE (CaSOj'l
PRECIPITATED; ANHYDRITE
(CaS04) STABLE
DIHYDRATE (CaS04 ZHjOt
PRECIPITATED; ANHYDRITE
STABLE
DIHYDRATE PRECIPITATED
AND STABLE
0 10 20 30 40 50 60
ACID CONCENTRATION,percent P^
Figure 10. Precipitation and stability of calcium
sulfates in phosphoric acid (21).
Reprinted from Chemistry and Technology of Fertilizers by
courtesy of John Wiley & Sons, Inc.
(21) Slack, A. V. Chemistry and Technology of Fertilizers. John
Wiley & Sons, Inc., New York, New York, 1967. pp. 69-97.
23
-------
The entire reaction, then, between the major (more than 90%)
phosphate rock constituents and sulfuric acid is as follows (17):
Ca10 (PO^) 6F2CaC03 + llI^SOi, + llnH20
— * 6H3P01, + HCaSOi+-nH20 + 2HF + H20 + C02 (7)
where n may equal 0, 1/2, or 2 depending on the degree of hydra-
tion of the calcium sulfate. Table 7 shows weight percent values
of compounds found in filtered wet process phosphoric acid (WPPA)
(22) . Table 8 gives an elemental analysis of commercial (concen-
trated) acid (21) .
2 . Process Description
Phosphoric acid can be produced by one of two methods: hydration
of phosphorus oxide derived from burning elemental phosphorus in
air (thermal process) or digestion of phosphate rock with a min-
eral acid such as sulfuric acid (wet process) . The acid produced
by the thermal process is known as furnace grade acid and, by the
nature of the process, is higher purity acid. Furnace grade
acid, used for animal feeds, detergents, fire retardant chemicals,
and other industrial phosphorus products, is no longer used to
produce phosphate fertilizers (8) .
The second, or wet process, method produces merchant grade phos-
phoric acid. Merchant grade acid contains more impurities than
does furnace grade acid. Currently, all phosphate fertilizer
production in the United States uses WPPA.
WPPA production methods differ principally in the degree of
hydration of the calcium sulfate. The degree of hydration is a
function of the temperature and phosphorus pentoxide concentra-
tion of the acidulation slurry (see Figure 10) . Calcium sulfate
can be precipitated in the dihydrate form (gypsum) , hemihydrate
form, or anhydrous form. Currently, all WPPA plants in the
United States use the dihydrate process. The hemihydrate and
anhydrite processes find limited use in Europe and Japan.
A schematic diagram of the basic dihydrate process for producing
orthophosphoric acid by the wet process method is shown in Fig-
ure 11. Production of the acid involves four unit operations:
raw material feed preparation, phosphate rock digestion, filtra-
tion, and concentration. The following sections contain detailed
process descriptions of each of these four operations.
(22) Lehr, J. R. Purification of Wet Process Acid. In: Phos-
phoric Acid, Volume I, A. V. Slack, ed. Marcel Dekker,
Inc., New York, New York, 1968. pp. 637-686.
24
-------
TABLE 7. TYPICAL COMPOSITION OF FILTERED WFPA (22, 23)
(weight percent)
Rock
source
Florida*
b
"b
b
b
c
°'d a
Western
C
Tennessee
Acid composition
P205
27.3
28.4
31.2
26.3
30.2
30.0
27.0 to 31.9
23.2
30.0
30.0
CaO
0.15
0.1
1.0
0.4
0.1
1.26
0.01 to 0.8
0.22
0.21
0.37
T
1.7
l.S
1.4
2.0
2.0
2.36
0.9 to 3.1
1.2
1.36
2.54
A1203
0.6
1.1
0.8
0.5
1.9
1.08
0.2 to 1.6
0.8
1.01
2.66
Pe2O3 MgO
1.1 0.28
2.0
1.7
1.1
1.1
0.86
0.8 to 2.4
0.6 0.33
0.42 0.05
2.27
K20
0.03
0.06
0.05
0.06
0.07
Na2O
0.08
0.01
0.13
0.01
0.43
Si02
1.1
1.6
1.21
0.2 to 0.6
0.74
0.1
S03
1.2 •'-
3.9
0.2
1.0
3.1
3.72
0.4 to 4.6
1.0
2.63
1.49
Composition
of
suspended
solid
A, B, C
A, C
A. C
Composition of clear, supernatant acid after cessation of precipitation; compounds identified in solidei A •=
B = Cai,SOi,
-------
Figure 11. Wet process for production of phosphoric acid.
a. Raw Material Feed Preparation--
Phosphate rock is delivered to the plant site by railroad hopper
cars. Unloading of these cars takes place in a three-sided shed
where the ore drops out of the bottom of the railroad car and is
conveyed to rock storage silos.
An exhaust system is installed in the unloading and transfer
areas to remove phosphate rock dust from the air. The exhaust
stream is passed through a baghouse before it is discharged to
the atmosphere. From the silos, the rock is classified by screen-
ing (60% to 80% less than 74 ym) or by air separation and is
passed on to the acidulator.
In addition to phosphate rock, sulfuric acid (93% to 98% H2SOU)
is delivered to the plant site. This acid is piped to storage
tanks from adjacent sulfuric acid plants.
b. Phosphate Rock Digestion—
The key feature in a phosphoric acid plant is the acidulator, the
reaction vessel where phosphate rock is digested with sulfuric
acid to produce orthophosphoric acid (28% to 30% P2Os) and gypsum,
Before the 1960's, the digestion section consisted of a series of
separate reaction vessels. Today, all wet process acid plants
use a single tank design consisting of multiple compartments or
26
-------
stages (24). The types of acidulation systems currently in use
in the United States include the Prayon, Prayon/Davy Powergas,
Dorr-Oliver, Singmaster and Breyer, and Swenson.
Each system design varies in terms of the number and location of
agitators and recirculation mechanisms and in the locations of
rock and sulfuric acid injection points. In the United States,
approximately 75% of all wet process acid trains use the Prayon
or a combined Prayon/Dorr-Oliver system. As Figure 12 illus-
trates, each of the systems uses different equipment, but the
basic process and resulting product and byproducts remain the
same.
Phosphate rock and sulfuric acid are added to recirculating
slurry in the acidulator. Approximately 3.35 metric tons of 70%
BPL (32% P205) phosphate rock and 2.75 metric tons of 93% to 98%
sulfuric acid are required to produce 1.0 metric ton of i^PO^
(100% P2®5 basis) (24). Some processes use dilute sulfuric acid;
the range of concentrations is 50% to 98% sulfuric acid. The
higher concentrations of sulfuric acid are generally preferred
because they remove excess water that must be evaporated during
the concentration step.
Average retention time in the reactor system ranges from 5.5 hr
to 8 hr (21). In all systems, recirculation of slurry is
required in order to reduce the adverse effects on the process
caused by fluctuations in rock analysis and incomplete mixing.
The recycled slurry also gives the control of supersaturation
necessary for good growth of gypsum crystals. In multicompart-
ment systems such as the Prayon single tank reactor, the recycle:
product ratios range from 10:1 to 20:1 (24).
Acidulation of rock and dilution of sulfuric acid produce heat:
163 kJ to 469 kJ per mole of fluorapatite (19) . The reaction
slurry must be cooled to prevent formation of other hydrated
crystal forms of calcium sulfate. Three methods of cooling are
used: blowing air into the slurry, flowing air across the slurry,
and vacuum flash cooling. Another approach, used by Prayon, is
to apply sulfuric acid which is already diluted and cooled. When
the heat of reaction and heat of dilution of sulfuric acid are
removed by flash cooling (Figure 13), submerged slurry pumps lift
the slurry from the attack tank and introduce it into the bottom
of a distributor in the flash cooler. A large slurry surface in
the top of the cooler flashes off water; the cooled slurry then
overflows the inner and outer edges of the distributor and
returns to the attack tank.
Vapors from the flash cooler are condensed in a barometric con-
denser and sent to a hot well. Noncondensables are removed by
(24) Lutz, W. A., and C. J. Pratt. Principles of Design and
Operation. In: Phosphoric Acid, Volume I, A. V. Slack, ed,
Marcel Dekker, Inc., New York, New York, 1968. pp. 158-208,
27
-------
Figure 12a. Flow diagram for Prayon phosphoric acid plant.
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 254, by courtesy of Marcel Dekker, Inc.
JHOSPW
ROCK
FROM FILTRATION
Figure 12b. Dorr-Oliver reaction system (vacuum cooled)
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 216, by courtesy of Marcel Dekker, Inc.
Figure 12. Digestion system designs (24).
28
-------
COOLING WATER.
H2S04 ACID
WATER
DtLUTER COOLER
(OPTIONAL)
RECYCLE ACID
FROM FILTER
VAPOR TO «>—
FLUORINE RECOVERY
CONDENSER AND VACUUM
PHOSPHATE
ROCK HOPPER
__» VAPOR TO
FLUORINE RECOVERY
CONDENSER AND VACUUM
PHOSPHATE
ROCK FEEDER
,— FILTER
FEED
PRIMARY DIGESTER
SECONDARY DIGESTER
Figure 12c. Flow diagram for Singmaster and Bryer
dihydrate phosphoric acid process.
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 274, by courtesy of Marcel Dekker, Inc.
WATER-
H2S04'
'niii
HEMIHYDRATE
VACUUM
COOLER
GYPSUM
SLURRY
VACUUM
COOLER
PHOSPHATE Q
ROCK T-\ v
NO. I REACTOR
HEMIHYDRATE
DIGESTER
GYPSUM
NO. 2 DIGESTER
GYPSUM
Figure 12d.
Flow diagram of Singmaster and Breyer
hemihydrate-dihydrate process.
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 364, by courtesy of Marcel Dekker, Inc.
29
-------
WATER
STEAM
WATER
93*T098%
ACID
DILUTION-COOLER
(OPTIONAL I
CRUSHED
ROCK
1
EMISSIONS TO
WET SCRUBBER
RECYCLED
WEAK(-20%P205I
ACID FROM
FILTRATION
SYSTEM
Figure 13.
ATTACK TANK
( REACTOR)
VENTED TO
WET SCRUBBER
—SLURRY FEED TO
FILTRATION SYSTEM
Phosphate rock digester and cooling system.
steam ejection and also vented to the hot well. Fumes from the
hot well may be vented to the wet scrubber, while the water
slurry is discharged to the gypsum pond.
c. Filtration—
Slurry from the final stage of the reactor system is continuously
withdrawn and pumped to a horizontal, rotary, tilting pan type of
vacuum filter to separate gypsum solids from the liquid (32%
PaOs) phosphoric acid. Two diagrams of this type of filtration
system are shown in Figures 14 (12) and 15 (15). Slurry is dis-
charged onto the filter, the undiluted mother liquor is collected,
and the remaining slurry is subsequently washed by three continu-
ous, countercurrent stages to remove phosphoric acid liquids.
The cake is dried by suction, the filter pan is inverted, and the
cake is washed from the filter with recycled gypsum pond water.
Gypsum slurry then flows to the holding pond for cooling and
solid settling. The filter cloth is washed, dried by suction,
and is then ready for the next cycle.
Acid from the first four stages of filtration is delivered to the
vacuum receivers and then to a multicompartment filtrate seal
tank. Undiluted mother liquor is pumped to a surge tank and then
to the concentration process. Water and acid from the second and
third washes are recycled to the preceding wash stage. Weak acid
from the first wash is delivered to the attack vessel. Vapors
from the vacuum receivers are cooled and vented to the wet scrub-
ber system. Cooling water and condensed vapors are used to wash
the cloth filter in the final stage of the filtration process.
30
-------
SLURRY
FROM '
REACTOR
WASH WATER
TO MAIN
VACUUM PUMP AND
FUMES TO
WET SCRUBBER
WEAK ACID
TO ATTACK TANK"
LIQUOR 1ST WASH 2ND WASH 3RD WASH
FILTRATE
RECEIVERS
MULT)COMPARTMENT FILTRATE SEAL TANK
I I I
DISCHARGE CAKE
RECYCLE
WATER
• PROCESS
EFFLUENT
WATER
RECYCLE
• TO EVAPORATOR
TO GYPSUM POND
Figure 14. Tilting pan filtration system (12)
CMC VASHNQ
CAKE DODMUMO
nu SLUM*
CMC DBIODCINO
AND DISCHARGING
Figure 15. Operating cycle of rotary horizontal
tilting pan filter (15).
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
p. 446, by courtesy of Marcel Dekker, Inc.
31
-------
d. Concentration —
Phosphoric acid (32% P20s) from the first filtration stage is con
centrated to 54% ?20s by vacuum evaporation of water. The acid
is circulated, first through a shell-and-tube heat exchanger,
then through a series of three flash chambers at 10 kPa to 20 kPa
pressure (25, 26) separated by shell-and-tube exchangers, as
shown in Figure 16 (12) . The flash chambers serve to provide
comparatively large liquid surface areas where water vapor can be
released with minimum phosphoric acid entrainment.
Minor acid impurities, such as compounds containing fluorine,
volatilize with the water vapor. The evolved vapors containing
fluorine compounds and phosphoric acid pass to a barometric con-
denser, from which the condensed vapors, process cooling water,
and condensed steam flow to a hot well. From the hot well, the
water is recycled back to the barometric condenser that is used
in connection with the acid flash cooler. Vapors from the hot
well are vented to the wet scrubber system.
A variety of minor acid impurities such as iron and aluminum phos-
phates, soluble gypsum, and f luosilicates form supersaturated
solutions in 54% PzOs phosphoric acid and will precipitate dur j n»j
storage. These precipitates, in turn, cause problems in tank car
unloading and customer processing. It is therefore necessary to
remove these precipitated impurities before the acid is sold. As
previously illustrated in Tables 7 and 8, there is a large reduc-
tion in impurities between the filtered and product acids.
The process used in the United States for removal of precipitated
solids from 54% P205 phosphoric acid involves only physical treat-
ment of the acid rather than the more complicated and expensive
solvent extraction processes utilized in Europe and Mexico (27) .
Precipitated impurities are physically separated from the acid by
settling and/or centrifugation.
Sludge is either sent to the gypsum pond, processed into a low
quality fertilizer, or recycled to the evaporator feed tank.
Recirculation of the sludge adds precipitated solids to the evapo-
rator feed, providing crystal surfaces in the acid. Because
salts coming out of solution during the evaporation process tend
to deposit on these crystals rather than on evaporator surfaces,
scaling is reduced. The clarified acid is then stored at ambient
temperatures .
(25) Cleanup Pays Off for Fertilizer Plant. Environmental
Science and Technology, 6 (5) :400-401 , 1972.
(26) Banford, C. R. IMC ' s New Plant Shows Off Latest H3PO<* Know-
How. Chemical Engineering, 70 (11) :100-102, 1963.
(27) Legal, C. C., and 0. D. Myrick, Jr. History and Status of
Phosphoric Acid. In: Phosphoric Acid, Volume I., A. V.
Slack, ed. Marcel Dekker, Inc., New York, New York, 1968.
pp. 1-89.
32
-------
PROCESS WATER
HEAT
EXCHANGER
STEAM
TO TO FILTER
GYPSUM CAKE WASH
POND
RECYCLED TO EVAPORATOR
FEED TANK OR
TO GYPSUM POND
•STEAM
VENT TO SCRUBBER
TO BAROMETRIC
•-CONDENSER IN
EVAPORATORS ( THREE IN SERIES )
REACTION SECTION
VENT TO SCRUBBER
STORAGE
Figure 16. Concentration and clarification (12).
-------
3. Gypsum Ponds
Gypsum ponds are used not only as settling basins for calcium
sulfate dihydrate (CaSOi»«2H20) , but can be used as cooling,
storage, and reconditioning ponds for all contaminated process
water streams in the plant or complex. Cooled and clarified
supernatant water from the pond can be recycled to supply over
80% of the water requirements for the plant (4).
A typical range of equilibrium compositions of gypsum pond water
is given in Table 9 (4, 20, 28). Impurities approach equilibrium
concentration in individual ponds over a period of 3 yr to 5 yr
as the water is recycled. These concentrations are then main-
tained by either volatilization and/or precipitation.
TABLE 9. EQUILIBRIUM CONCENTRATION RANGES
OF GYPSUM POND WATER (4,20, 28)
(g/m3)
Contaminant Concentration
P205 equivalent 6,00 to 12,000
Fluoride 3,000 to 10,000
Sulfate 2,000 to 4,000
Calcium 350 to 1,200
Ammonia 0 to 100
Nitrate 0 to 100
Silica "ol,600
Aluminum 100 to 500
Iron 70 to 300
pH 1.0 to 1.8
4. Industry Characterization
All 36 WPPA plants in the United States (7) use the same basic
processes described in previous sections. Specific equipment and
operating conditions vary from plant to plant. General industry
practice has included use of closed water recycle systems and a
single scrubber unit for the collective emission sources, al-
though variations do exist. One plant, located on the Missis-
sippi River and lacking available land area for a gypsum pond,
was designed for use of river water on a once through basis for
scrubbing air emissions, for operation of the barometric con-
denser, and for meeting cooling requirements.
(28) Huffstutler, K. K. Pollution Problems in Phosphoric Acid
Production. In: Phosphoric Acid, Volume I, A. V. Slack, ed,
Marcel Dekker, Inc., New York, New York, 1968. pp. 727-739.
34
-------
The 36 WPPA plants have production capacities which range from
6,480 to 751,300 metric tons of P205 per year, with an average
plant capacity of 251,600 metric tons of P20s per year or 699
metric tons of P20s per day (see Appendix A). Individual plant
capacities vary throughout the range as shown in Table 10 and
illustrated in Figure 17. Average plant production was calcu-
lated by dividing the total annual wet process phosphoric acid
production for 1975 (6,290,000 metric tons of P205 per year) by
the total number of WPPA plants, i.e., 36. An average WPPA
plant was therefore defined as producing 175,000 metric tons of
P20s per year or 486 metric tons of P205 per day.
TABLE 10. DISTRIBUTION OF WPPA PLANTS BY PRODUCTION CAPACITY
Individual plant
capacity,
103 metric tons
P205/yr
>700
600 to 700
500 to 600
400 to 500
300 to 400
200 to 300
100 to 200
<100
Total
Number
of
plants
1
3
2
3
1
7
9
10
36
Combined capacity
for all plants in
category,
103 metric tons
P205/yr
751.3
2,023
1,187
1,358
326.5
1,720
1,297
394.2
9,057
Percent of
total capacity
8.3
22.3
13.1
15.0
3.6
19.0
14.3
4.4
100
Approximately 4 to 5 metric tons of gypsum are formed for every
metric ton of P20s (20). The magnitude of this waste is an
indication of the size of gypsum ponds, which also serve as
holding ponds for the process water, necessary for plant opera-
tion. One reported rule of thumb for sizing is 0.00223 km2 per
daily metric ton of P2O5 production (20). An average plant,
producing 486 metric tons P205 daily, would require a gypsum pond
of 1.08 km2 (263 acres).
The locations of the 36 phosphoric acid plants are listed in
Table A-2 of Appendix A, which also gives information on the popu-
lation densities in counties where the plants are located. A dis-
tribution of plants by county population density is shown in
Table 11. The predominant population density range is 40 to 49
persons/km2; the median value for the 36 plants is 46.1 persons/
km2. This value is used for the population density around an
average plant.
35
-------
-------
C. SUPERPHOSPHORIC ACID PRODUCTION
1. Process Chemistry
Superphosphoric acid is produced by dehydrating "wet process"
phosphoric acid. When phosphoric acid is heated to elevated
temperatures, molecular dehydration occurs and the molecules
combine to form polyphosphoric acid chains as shown in
Equation 8 (29) .
A
+ (x - 1)H20 (8)
As an example, tripolyphosphoric acid is formed as follows (29) :
A
~* H5P3010 + 2H20 (9)
The resulting product is a mixture of phosphoric acid (HaPO^) and
polyphosphoric acid chains of varying lengths; this mixture is
called superphosphoric acid. If temperature or retention time
is increased, a higher degree of dehydration is obtained. Prod-
uct composition is affected in that the amount of phosphoric
acid decreases while the average chain length of the polymeric
acids increases. Wet process superphosphoric acid is concen-
trated to 68.5% to 72% P205 (27). At this degree of hydration,
the P20s in the acid is approximately 40% remaining as phosphoric
acid (H3POi») / 40% as pyrophosphoric acid (H<,P205) , 5% as tripoly-
phosphoric acid, and 15% as longer chain acids (27) .
Wet process superphosphoric acid differs from pure superphos-
phoric acid produced from electric-furnace phosphorus primarily
in the chemistry associated with the impurities in the wet acid.
Major impurities in wet process superphosphoric acid are calcium,
iron, aluminum, magnesium, potassium, sodium, fluorine (hydrogen
fluoride [HF] , fluosilicic acid [H2SiF6] , silicon tetraf luoride
[SiF
-------
TABLE 12. COMPOSITION OF SUPERPHOSPHORIC ACID (15)
(percent)
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 1083, by courtesy of Marcel Dekker, Inc.
Constituent
Total P205
Ortho-P205
Nonortho-P20s
Fe203
A1203
Combined Fe20a and A1203
Fluorine
CaO
S03
Conversion to polyphosphate, %
Typical
content
69.60
42.50
27.10
2.50
2.05
4.55
0.51
0.15
2.44
29.0
Range
69 to 70
42 to 45
4 to 5
NOTE.—Blanks indicate data not available.
2. Process Description
a. Submerged Combustion—
Two commercial processes are used for the production of super-
phosphoric acid from wet process acid: submerged combustion and
vacuum evaporation. Currently, in the United States only two
plants (Allied Chemical Corp. and Occidental Petroleum Corp.),
accounting for approximately 26% of the superphosphoric acid pro-
duction capacity, use submerged combustion.
The submerged combustion process was pioneered by the Tennessee
Valley Authority (TVA). Wet acid is dehydrated by bubbling hot
combustion gases through a pool of the acid. Combustion gases
are supplied by burning natural gas in a separate chamber. The
combustion gases are diluted with air to maintain a gas tempera-
ture of 925°C for introduction into the acid evaporator. After
passage through the acid, the hot combustion gases are sent to a
separator to recover entrained acid droplets and then sent to a
wet scrubber emissions control system.
Clarified acid containing 54% P2O5 is continuously fed to the
evaporator from storage, and acid containing 72% P2O5 is with-
drawn from the evaporator to product holding tanks. Acid cooling
is accomplished by circulating water through stainless steel cool-
ing tubes in the product tanks. Superphosphoric acid production
can be controlled by regulation of the natural gas and air flows
to the combustion chamber, by the feed rate of acid to the evapo-
rator, or by the amount of excess air used in the combustion
process.
38
-------
b. Vacuum Evaporation—
Most plants in the United States (approximately 74%) employ
vacuum evaporation utilizing heat transfer surfaces in the pro-
duction of superphosphoric acid (15, 21). Two popular types of
evaporators used are the falling film evaporator developed by
Stauffer Chemical Co. and the forced circulation evaporator
developed by Swenson Evaporator Co. In the seven plants which
use vacuum evaporation, approximately 60% of superphosphoric acid
production is by the Stauffer process. The remaining 40% uses
the Swenson design.
In the Stauffer process, clarified 54% P205 phosphoric acid is
continuously fed to the evaporator recycle tank where it mixes
with superphosphoric acid from the evaporator. Some of the mix-
ture (approximately 1.2%) is drawn off as product acid, but most
(approximately 98.8%) is pumped to the top of the evaporator and
is distributed across the heat exchanger tube bundle. The fall-
ing acid, heated by high-pressure steam condensing on the outside
of the tubes, evaporates. The vapors and dehydrated acid then
enter the separator section where entrained acid mist is removed.
Product acid flows to the recycle tank, and the vapor is drawn
off, condensed in a barometric condenser, and delivered to a hot
well. Noncondensables are removed by a two-stage steam ejector
and are vented to the hot well. Superphosphoric acid flows to
the recycle tank where it is mixed with more 54% P205 phosphoric
acid and recycled or removed as product. The approximate recycle
to feed acid ratio is 80:1. The product stream is cooled and
stored before shipping. Both the hot well and cooling tank are
vented to wet scrubbing systems.
The Swenson process utilizes closed heat exchanger tubes filled
with heat exchanger fluid to provide the heat of reaction. Feed
acid (54% P205) pumped into the evaporating system mixes with
recycled superphosphoric acid. As the acid leaves the exchanger
tube bundle and enters the flash chamber, evaporation begins.
Vapors are removed by a barometric condenser. Condensed materi-
als and noncondensed vapors are delivered to a hot well. Product
acid flows toward the bottom of the flash chamber where part
(approximately 0.6%) is removed to a cooling tank and the rest
(99.4%) is recycled. An approximate recycle to feed ratio is
150:1 (compared with 80:1 for the Stauffer process).
Cooling in both systems is accomplished by circulating water
through stainless steel tubes in the holding tank.
3. Industry Characterization
Nine plants in the United States produce wet process superphos-
phoric acid. These plants have production capacities which range
from 12,960 to 295,000 metric tons of P205 per year, with an
average plant capacity of 115,900 metric tons of P205 per year
or 320 metric tons of P205 per day (see Appendix A). Plant
39
-------
capacity distributions for those plants producing superphosphoric
acid are given in Table 13 and Figure 18. Average plant pro-
duction was calculated by dividing the total annual wet process
superphosphoric and production for 1975 (506,000 metric tons of
P205) by the total number of SPA plants. An average SPA plant
was therefore defined as producing 56,200 metric tons of P205
per year or 156 metric tons per day.
TABLE 13. DISTRIBUTION OF SPA PLANTS BY PRODUCTION CAPACITY
Individual plant
capacity,
103 metric tons
P205/yr
>200
150 to 200
100 to 150
50 tO 100
<50
Total
Number
of
plants
1
3
1
1
3
9
Combined capacity
for all plants in
category,
103 metric tons
PaOs/yr
295
479
124
65.2
79.6
1042.8
Percent of
total capacity
28.3
45.9
11.9
6.3
100
o
ANNUAL PLANT PRODUCTION. 10 metric tons/yr
Figure 18. Distribution of SPA plants by capacity,
40
-------
The population densities of the counties where the nine superphos-
phoric acid plants are located range from 2.9 to 385.9 persons/
km2, is used for the population density around an average plant
(Table 14) .
TABLE 14. DISTRIBUTION OF SPA PLANTS BY
COUNTY POPULATION DENSITY
Population density,
persons/km2
0 to 9
10 to 19
20 to 39
40 to 49
236
386
Total
Number of
plants
2
2
0
3
1
1
9
Percent of
total plants
22.2
22.2
33.3
11.1
11.1
100
D. NORMAL SUPERPHOSPHATE PRODUCTION
1. Process Chemistry
Phosphate rock is composed of phosphate in the form of the min-
eral fluorapatite { [Ca3 (PO^) 2] 3»CaF2) . Phosphate in this form
is only slightly soluble in water, thus reducing its availability
for plant growth.
NSP, containing from 16% to 21% P20s, is prepared by reacting
ground phosphate rock with 65% to 75% sulfuric acid. The primary
objective of this acidulation process is to convert the fluorapa-
tite in phosphate rock to soluble monocalcium phosphate, a form
readily available to plants. While the overall chemistry is com-
plex due to the composition of the rock, the major reaction
involving phosphate may be stated simply as (4) :
lCa3(P01))2] 3 -CaP 2 + THjjSO,, -I- 3H2O -* 3 [CaHk (PO,, ) 2 -H20] + 7CaSOu + 2HF d^)
Fluorapatite Sulfuric Water Monocalcium Calcium Hydrogen
(phosphate rock) acid phosphate sulfate fluoride
monohydrate
2. Process Description
NSP is prepared by reacting ground phosphate rock with 65% to 75%
sulfuric acid. Rock and acid are mixed in a reaction vessel,
held in an enclosed area (den) while the reaction mixture solidi-
fies, and then transferred to a storage pile for curing. A gener-
alized flow diagram of the process for the production of NSP is
shown in Figure 19 (4).
41
-------
EMISSIONS
*
DUST
COLLECTOR
SULFURIC
ACIO
••EMISSIONS
(UNCONTROLLED )
PRODUCT
Figure 19. ROP-NSP production facility (4).
-------
Mixing of the phosphate rock and sulfuric acid (acidulation)
takes place in either a pan or cone mixer. The pan mixer, used
in conjunction with a batch den and largely replaced by the cone
mixer, is fitted with slowly rotating plows. Larger units are
capable of handling a 2-metric ton batch of material (16).
The cone mixer, developed by the TVA, has come into use in more
than 80% of the plants because of its relatively low capital
expense, low maintenance cost, simple operation, and lack of
moving parts (16) . Sulfuric acid is fed into the cone tangen-
tially in order to provide the necessary mixing action. Fresh
superphosphate discharges from the cone mixer to a pugmill for
additional mixing of acid and rock before discharge to a den.
This type of mixer is suitable for use with either a batch or
continuous den.
Plants are described as batch or continuous, depending upon the
type of den used. In a continuous den, solidification and con-
current evolution of reaction gas take place on a slow-moving
conveyor (den) enroute to the curing area. The low travel speed
allows about 1 hr for the solidification process to occur before
the material reaches the end of the belt. A cutting knife then
slices the solidified material from the belt. NSP as it comes
from the den is uncured and must be held in a curing building
for a period of between 2 wk and 6 wk to permit acidulation to
go to completion.
A batch den is a closed compartment except for a vent that
releases reaction gases. Batch dens commonly used in this coun-
try have capacities ranging from 35 to 275 metric tons (16).
After a setting period, ranging from 1.5 hr up to 10 hr, the
solidified NSP material must be removed from the den and trans-
ferred to storage. Dens operate either automatically, with a
cutting wheel that shaves the solidified mass from the den, or
manually, with a mechanical cutter, a drag line, or a crane.
Following curing, the product can be ground and bagged for sale,
or it can be granulated for sale as granulated superphosphate or
granular mixed fertilizer. Granular mixed fertilizers are
described in a separate report entitled "Source Assessment:
Fertilizer Mixing Plants" and are therefore not included in the
present discussion (5).
In producing a granular normal superphosphate (GNSP) material,
the hardened HOP product is first fed to a pulverizer where it is
crushed, ground, and screened. Screened material is then sent
to a rotary drum granulator. Steam or water is added, if needed,
to aid in granulation. The mixture then passes through a rotary
dryer where it is dried to set its form and sufficient moisture
is removed to eliminate the chance of the pellets binding
together. The material then goes through a rotary cooler and on
to storage bins for sale as bagged or bulk product.
43
-------
In some cases, the ROP-NSP material is granulated before curing
in a similar operation.
Sources of emissions at an NSP plant include the mixer, den, and
curing building. Emissions of fluoride and particulate from the
mixer and den are controlled by scrubbing with recycled water.
Fluorides evolved during curing and particulates released from
fertilizer handling operations (including screening and milling
in the product storage building) are uncontrolled at a typical
plant. The ground rock unloading, transfer, and storage facili-
ties together with the process rock weighers and feeders comprise
an additional source of particulate emissions. These emissions
are controlled by baghouse collectors.
3. Industry Characterization
Only a small portion (less than 10%) of total NSP production is
applied directly as ROP-NSP or GNSP product (30) . GNSP accounts
for less than 5% of total NSP production, and emissions from this
plant type are therefore not considered. Most of the NSP mater-
ial is sent to a fertilizer mixing plant and used in the prepara-
tion of fertilizers containing more than one of the following
nutrients: nitrogen, phosphorus, and potassium.
An average NSP plant is defined as one that produces 6,650 metric
tons of P205 per year of run-of-the-pile grade fertilizer. The
average NSP plant is located in a county having a population
density of 426 persons/km2. (See Appendix A for a complete list
of plant capacities and locations.) Because individual plant pro-
duction statistics are not available, the average plant produc-
tion rate was calculated by dividing the total annual NSP
production for 1975 (439,000 metric tons P2O5 per year) by the
total number of NSP plants; i.e., 66.
E. TRIPLE SUPERPHOSPHATE PRODUCTION
1. Process Chemistry
TSP, 45% to 49% P20S, contains between 2.5 and 3 times more P205
than normal superphosphate. This higher P205 content product is
achieved through the use of phosphoric acid in place of sulfuric
acid as shown in the following equation (31):
[Ca3(POlt)2]3-CaF2 + 14H3POi, + 10H2O -> 10 [CaH4 (PO,,) 2'H2O] + 2HF (11)
Fluorapatite Phosphoric Water Monocalcium Hydrogen
(phosphate rock) acid phosphate fluoride
monohydrate
(30) Personal communication with Ed Harre, Tennessee Valley
Authority, Muscle Shoals, Alabama, 14 April 1977.
(31) Background Information for Standards of Performance: Phos-
phate Fertilizer Industry, Vol. 1—Proposed Standards.
EPA-450/2-74-019a (PB 237 606), U.S. Environmental Protec-
tion Agency, Raleigh, North Carolina, October 1974. 140 pp
44
-------
Higher grade TSP materials (with 54% to 55% P205) have been manu-
factured by the TVA but only on an experimental basis (9, 32).
2. Process Description
Two principal types of TSP are produced: ROP-TSP and GTSP.
Physical characteristics and processing conditions differ for the
two materials. ROP material is essentially a nonuniform pulver-
ized mass. In contrast, GTSP is a hard, uniform, pelletized
granule. The ROP process is used for approximately 40% of total
TSP production, and the granular process is used for the remain-
ing 60%. Some overlap occurs as a portion of the ROP product is
consumed in producing a GTSP product.
a. Run-of-Pile Triple Superphosphate—
The ROP-TSP production process as shown in Figure 20 is essen-
tially identical to the NSP process except that phosphoric acid
rather than sulfuric acid is used for acidulation (31) . Mixing
of the ground rock and phosphoric acid (50% to 54% P205 content)
occurs in a cone mixer. The majority of plants (more than 90%)
in the United States use the TVA cone mixer. This mixer has no
moving parts, and mixing is accomplished by the swirling action
of rock and acid streams introduced simultaneously into the cone.
The resulting viscous slurry, on discharge from the mixer,
quickly (in 15 s to 30 s) becomes plastic and begins to solidify.
Solidification, together with the concurrent evolution of reac-
tion gases, takes place on a slow-moving conveyor (den) enroute
to the curing area.
On its way to the curing building, the mix may pass through
several mixers or plungers that increase contact between the rock
and acid and help to release trapped gases. Solidified material
takes on a honeycomb appearance because of the copious evolution
of gas throughout the mass. At the point of discharge from the
den, the material passes through a rotary mechanical cutter that
breaks up the solid mass. Coarse ROP product is sent to a stor-
age pile where it is cured for a period of 3 wk to 5 wk. Final
ROP product is then mined from the "pile" in the curing shed,
and subsequently crushed, screened, and shipped in bulk (4, 16,
31) .
This method of production gives a material that is nonuniform in
particle size with consequent inferior handling characteristics.
As a result, over 90% of all ROP-TSP is later granulated, either
by the process described in the next section, or at fertilizer
mixing plants that produce nitrogen-phosphorus-potassium (N-P-K)
fertilizers (5). The remaining ROP-TSP is used as direct appli-
cation fertilizer. Sources of air emissions and emission species
(32) Gartrell, F. E., and J. C. Barber. Pollution Control Inter-
relationships. Chemical Engineering Progress, 62(10); 44-47,
1966.
45
-------
EMISSIONS
*
T
ROCK
UNLOADING
GROUND ROCK
( (
ROCK FEEDER
SYSTEM
WET PROCESS
PHOSPHORIC
ACID
ROCK BIN
DUST
\ 7 COLLECTOR
EMISSIONS
*
RECYCLED
POND WATER
WET
SCRUBBER
•PRODUCT
Figure 20. ROP-TSP production facility (31).
-------
at a typical ROP-TSP production facility are similar to those
described for an NSP plant. Emissions of fluoride vapors and
particulates from the cone mixer, den, and curing building are
controlled by wet scrubbers using recirculated pond water.
Particulate emissions from ground rock storage and transfer
facilities are controlled by baghouse collectors.
b. Granular Triple Superphosphate—
Granulation is employed as a meansT of improving the storage and
handling properties of fertilizer materials. This process yields
larger, more uniform particles (mean particle diameters between
1 mm and 4 mm) either by agglomeration of ROP material or by
direct granulation of raw product slurry.
(1) GTSP from ROP-TSP—A generalized flow diagram of the process
for the production of GTSP from cured ROP-TSP is shown in Fig-
ure 21 (4, 16, 31). Less than 10% of the GTSP consumed in the
United States is currently produced by this method.
CURING
BUILDING
ROP - TSP
MILL - — |
i 1
SCREENS
AIR
EMISSKN
t
AIR
5 psrl" ™'S5IONS
1 p^ | I
r«— FUEL
I OIL, NATURAL GAS 1
•-- AIR
1 g
AIR EMISSIONS
|_J
MILL •— i
1
SCREENS
AIR
EMISSIONS
J
GRANULAR
PRODUCT
RECYCLE FINES
Figure 21. Production of GTSP from cured ROP-TSP (4, 16, 31).
In this process, cured ROP-TSP product is removed from storage
and sent to a pulverizer where it is ground and screened. The
screened material is then sent to a rotary drum granulator. The
addition of steam and water aids the granulation process. The
resultant wet granules are discharged to an air dryer where water
is evaporated to give a hard, dense, granular product. The dis-
charge from the dryer is screened, and acceptable product is sent
to storage. Oversized material is recycled to the pulverizer and
undersized to the granulator.
(2) Basic GTSP Process — Two methods for the direct production of
GTSP are currently available: 1) Dorr-Oliver slurry granulation
process and 2) TVA one-step granulation process. Direct granula-
tion using the Dorr-Oliver process accounts for over 90% of total
GTSP production, whereas the one-step process developed by the
TVA during the past 10 yr to 15 yr remains experimental (4, 17,
31). The Dorr-Oliver slurry granulation process is illustrated
in Figure 22 (16, 31). In this process, phosphate rock, ground
to a fineness located between specific particle size levels (80%
through a 150-um screen and 95% through a 75-|im screen) , is mixed
with phosphoric acid in a reactor or mixing tank. The phosphoric
acid used in this process is appreciably lower in concentration
(40% P20s) than that used in ROP-TSP manufacture because the
47
-------
EMISSIONS
CO
GROUND
PHOSPHATE ROCK
EMISSIONS
WET PROCESS
PHOSPHORIC 1
ACID
- r . - -
i
i
r
i
BAGHOUSE
\«EACTO/
RECYCLED
PONO WATER
ELEVATOR
CURING BUILDING
I STORAGE 4 SHIPPING I
Figure 22. Dorr-Oliver slurry granulation process for TSP (16, 31).
-------
lower strength acid maintains the slurry in a fluid state during
a mixing period of 1 hr to 2 hr (17, 20, 33). A thin slurry is
continuously removed and distributed onto dried, recycled fines
where it coats out on the granule surfaces and builds up the
granule size.
Pugmills and rotating drum granulators are used in the granula-
tion process. A pugmill is composed of a U-shaped trough carry-
ing twin contrarotating shafts upon which are mounted strong
blades or paddles. Their action agitates, shears, and kneads the
solid-liquid mix, and transports the material along the trough.
The basic rotary drum granulator consists of an open-end, slight-
ly inclined rotary cylinder, with retaining rings at each end and
a scraper or cutter mounted inside the drum shell. Drums vary in
diameter from 2 m to 3 m and in length from 3 m to 6m. A roll-
ing bed of dry GTSP material is maintained in the unit while the
liquid slurry is introduced through horizontal, multioutlet
distributor pipes set lengthwise in the drum under the bed.
Slurry-wetted granules then discharge onto a rotary dryer where
excess water is evaporated and the chemical reaction is acceler-
ated to completion by the dryer heat. Dried granules are then
sized on vibrating screens. Oversized particles are crushed and
recirculated to the screen, while undersized (smaller than 1 mm)
particles are recycled to the granulator. Product-sized (1 mm to
4 mm) granules are cooled in a countercurrent rotary drum cooler.
The product is then sent to a storage pile for curing. After a
curing period of 3 days to 5 days, granules are removed from
storage, screened, bagged and shipped (31).
In the TVA one-step granulation process, ground phosphate rock
and recycled fines are fed directly into the acidulation drum
along with concentrated phosphoric acid and steam. Granulation
occurs in this revolving cylindrical reactor. The use of steam
accelerates the reaction and ensures an even distribution of
moisture in the mix. A more concentrated phosphoric acid (con-
taining 73.5% PaOs) can be used, resulting in a higher grade
granular product containing about 54% available ^2°$ (32). After
granulation occurs in the reaction cylinder, granules are screen-
ed, cooled, and sent to storage in a manner similar to that
described for the Dorr-Oliver process.
Emissions of fluorine compounds, SOX/ and dust particles occur
during the production of GTSP by the Dorr-Oliver process (16,
31). Silicon tetrafluoride and hydrogen fluoride are released by
the acidulation reaction and evolve from the reactor, granulator,
(33) Final Guideline Document: Control of Fluoride Emissions
From Existing Phosphate Fertilizer Plants. EPA-450/2-77-005
(PB 265 062), U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1977. 277 pp.
49
-------
dryer, and cooler. Evolution of fluorides continues at a lower
rate in the curing building as the reaction proceeds. SOX enter
the dryer exhaust stream as a result of the sulfur composition of
the fuel oil. Sources of particulate emissions include the
reactor, granulator, dryer, cooler, screens, mills, and transfer
conveyors. Additional emissions of particulate result from the
unloading, storage, and transfer of ground phosphate rock.
At a typical plant, emissions from the reactor and granulator are
controlled by scrubbing the effluent gas with recycled pond
water. Emissions from the dryer, cooler, screens, mills, product
transfer systems, and storage building are sent to a cyclone
separator for removal of a portion of the dust loading before
being sent to wet scrubbers (31). Baghouses are used to control
the fine rock particulate caused by the preliminary ground rock -
handling activities.
3. Industry Characterization
For TSP production, two distinct plant types are considered:
ROP-TSP and GTSP.
a. Run-of-Pile Triple Superphosphate
An average ROP-TSP plant produces 59,700 metric tons of P205 per
year and is located in a county having a population density of
86.1 persons/km2. Average plant production was obtained by divid-
ing the total amount of ROP-TSP produced in 1975 (597,110 metric
tons P20s per year) by the total number of ROP-TSP plants; i.e.,
10.
b. Granular Triple Superphosphate
An average GTSP plant is defined as one that produces 69,100
metric tons of P20s per year by the Dorr-Oliver slurry granula-
tion process and is located in a county having a population
density of 73.8 persons/km2. The average plant production rate
was calculated by dividing the total amount of f^TSP produced in
1975 (898,900 metric tons P205 per year) by the total number of
GTSP plants; i.e., 13.
F. AMMONIUM PHOSPHATE PRODUCTION
1. Source Definition
Ammonium phosphates are produced by reacting phosphoric acid with
anhydrous ammonia. Both solid and liquid ammonium phosphate
fertilizers are produced in the United States. Ammoniated super-
phosphates are also produced by adding NSP or TSP to the mixture.
In this study, only granulation of phosphoric acid with anhydrous
ammonia by ammoniation-granulation to produce granular fertil-
izers will be discussed. An environmental source assessment of
the production of liquid ammonium phosphates and ammoniated
superphosphates is separately reported in Reference 5.
50
-------
Approximately 99% of ammonium phosphates are used as fertilizers,
with the remaining quantity consumed in fire retardants; as addi-
tives to livestock feed; in manufacture of yeast, vinegar, and
bread improvers; in flux for soldering; and for sugar purifi-
cation (34, 35). As fertilizers, product nutrient analyses for
typical ammonium phosphates range from 11% to 21% nitrogen and
20% to 55% PzOs (1) • Important ammonium phosphate fertilizer
grades in the United States are
Primarily monoammonium phosphates (MAP)
11-48-0 11-55-0
13-52-0 16-20-0
Primarily diammonium phosphates (DAP)
16-48-0 18-46-0
where N-P-K analysis represents
N = percentage of available nitrogen
P = percentage of available ?20s
K = percentage of soluble potassium oxide (K20)
In 1975, 84% (on a ?205 basis) of the ammonium phosphates pro-
duced consisted of DAP grade (9). When used as fertilizers,
ammonium phosphates are either used directly or blended with
other fertilizers, either in liquid or solid form, to produce
mixed fertilizers. However, due to the nature of various report
ing systems and the complexity of the fertilizer industry, it is
impossible to extract amounts of ammonium phosphates used for
each application (5) .
Emissions from production of mixed fertilizers using granular
'ammonium phosphates are addressed in "Source Assessment: Fertil
izer Mixing Plants" (5) . Consequently, this document will dis-
cuss emissions from production of granular ammonium phosphates
and will encompass process operations from feeding of raw materi
als to loading of product for shipment.
(34) David, M. L., J. M. Malk, and C. C. Jones. Economic Analy-
sis of Effluent Guidelines Fertilizer Industry. EPA-230/2-
74-010 (PB 241 315), U.S. Environmental Protection Agency,
Washington, D.C., January 1974.
(35) The Condensed Chemical Dictionary, Eighth Edition,
G. G. Hawley, ed. Van Nostrand Reinhold Company, New York,
New York, 1971. p. 54.
51
-------
2. Process Chemistry
The ternary solubility diagram (ammonia-phosphoric acid-water)
presented in Figure 23 (36) identifies four potential anhydrous
salts of ammonia and phosphoric acid having NH3:H3POi4 mole ratios
of 7:3, 2:1, 1:1, and 1:2. NHi+^PO^ (MAP, mole ratio 1:1) and
(NHtt)2HPOit (DAP, mole ratio 2:1) are salts of commercial fertil-
izer importance. These desired products are obtained by operat-
ing along the solubility boundary at required conditions; i.e.,
operation along the segment marked DAP yields DAP, while operat-
ion along the segment marked MAP yields MAP. Lines from the
solubility curve to the right-hand border on Figure 23 represent
paths along which solution composition would change during
crystallization or solution (36).
20
40
60
wt%
100
Figure 23. Solubility boundaries for the ammonia-
phosphoric acid-water system (36).
Reprinted from The Chemistry and Technology of Fertilizers
by courtesy of the American Chemical Society.
Production of commercial ammonium phosphates is based on four
exothermic reactions. MAP is produced from 1 mole of phosphoric
acid and 1 mole of ammonia, yielding a product having 12.2%
(36) Chemistry and Technology of Fertilizers. V. Sauchelli, ed.
Reinhold Publishing Corp., New York, New York, 1960.
pp. 251-268.
52
-------
nitrogen (N) and 61.7% available phosphorus (P20S) ;
while releasing 105 kJ/mole (37, 38).
NH3
.e.
12-62-0,
(12)
DAP production combines 1 mole of phosphoric acid with 2 moles of
ammonia yielding a product having 21.2% nitrogen and 53.8% avail-
able phosphorus; i.e., 21-54-0, while releasing 159 kJ/mole
(37, 38).
2NH
MAP also reacts with ammonia to produce DAP and 54 kJ/mole
(37, 38) .
(13)
NH3
(NHi»)aHPO*
(14)
To attain various desired product analyses, sulfuric acid is
added in appropriate quantities and reacts with ammonia to form
ammonium sulfate and to release 138 kJ/mole (17, 37, 38) .
2NH3
(15)
Properties of pure crystalline MAP and DAP are listed in Table 15
(36, 37, 39) and presented in Figure 24.
Analyses of raw materials for ammonium phosphate manufacture are
presented in Table 16. Ammonium phosphates can be made from
either furnace process phosphoric acid or WPPA. Impurities in
WPPA prevent production of fertilizers having analyses equivalent
to pure MAP or DAP composition. For some products, e.g., 16-20-0,
diluents such as sulfuric acid are added to phosphoric acid by
design to reduce available phosphorus content of product to
desired levels. Commercial grades of ammonium phosphate range
from MAP grade 11-48-0 to DAP grade 18-46-0. Intermediate grades
identified earlier are either mixtures of MAP and DAP or diluted
MAP or DAP.
(37) Waggaman, W. H. Phosphoric Acid, Phosphates, and Phosphatic
Fertilizers, Second Edition. Reinhold Publishing Corp.,
New York, New York, 1952. pp. 308-344.
(38) Himmelblau, D. M. Basic Principles and Calculations in
Chemical Engineering, Second Edition. Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey, 1967. pp. 449-454.
(39) Kirk-Othmer Encyclopedia of Chemical Technology, Second
Edition, Vol. 9. John Wiley & Sons, Inc., New York, New
York, 1966. pp. 46-132.
53
-------
TABLE 15. PROPERTIES OF PURE AMMONIUM PHOSPHATES (36, 37, 39)
Property
MAP
DAP
N, %
P205, % available
Heat of formation, kJ/mole
Specific gravity at 19°C
Solubility, g/100 g H2O:
At 20°C
At 40°C
At 75°C
Dissociation pressure, Pa:
At 100°C
At 125°C
12.2
61.7
-1,450.8
1.803
37.4
56.7
108.8
Negligible
6.7
21.2
53.8
-1,573.7
1.619
69.0
81.0
108.7
670
4,000
Figure 24
Ll L2 LJ 1.4 l.S 1.6 1.7 L8 1.9 2.0
NHjMjf>04 MOLE RATIO
Ammonium phosphate solubility and viscosity
as a function of NH3:H3POu mole ratio (36).
Reprinted from The Chemistry and Technology of Fertilizers
by courtesy of the American Chemical Society.
54
-------
TABLE 16. COMPOSITION OF AMMONIUM PHOSPHATE
RAW MATERIALS (22, 37, 40, 41)
Composition, wt %
Anhydrous
Component ammonia
NH3 99.9
P205
Ca
Fe
Al
Mg
Cr
V
Na
K
F
S03
Si02
C
Solids
Cl
Pb
Cu
As
WPPA
Filtered
_b
28.7
0.30
0.45
0.29
0.13
0.05
0.02
1.82
2.11
0.79
(average)
Concentrated
53.3
0.06
0.78
0.52
0.26
0.02
0.02
0.45
0.06
0.56
2.3
0.16
0.24
3.7
Furnace process
phosphoric acid,
ppm
54.32 wt %
0.0
2
0.0
0.2
0.01 wt %
0.0
0.4
0.0
0.0
2
0.2
0.1
Commercial food-grade phosphoric acid.
Blanks indicate data not applicable.
3. Process Description
Two basic mixer designs are used by ammoniation-granulation
plants: pugmill ammoniator and rotary-drum ammoniator. Approxi-
mately 95% of ammoniation-granulation plants in the United States
use a rotary-drum mixer developed and patented by the TVA (5).
The primary product of this technology is 18-46-0, consisting
primarily of DAP. Ammonium phosphate products having a lower
NHszHsPO^ mole ratio are made using the Dorr-Oliver process or
variations of it. The degree of ammoniation utilized with this
technology ranges from an NHstHsPO^ mole ratio of 1.0 to 1.8, and
the primary product is 16-48-0, a product containing approxi-
mately one-third MAP and two-thirds DAP.
(40) Slack, A. V. Fertilizer Developments and Trends. Noyes
Development Corp., Park Ridge, New Jersey, 1968. pp. 77-274,
(41) Kirk-Othmer Encyclopedia of Chemical Technology, Second Edi-
tion, Vol. 15. John Wiley & Sons, Inc., New York, New York,
1968. p. 260.
55
-------
a. TVA Process—
A general process flow diagram of the TVA ammonium phosphate
process is presented in Figure 25. Phosphoric acid is mixed in
an acid surge tank with 93% sulfuric acid (used for product
analysis control) along with recycle and acid from wet scrubbers.
Mixed acids have a P2°5 content of 40% to 45% (42). This analy-
sis is attained by mixing unconcentrated filtered WPPA, 28.7%
P205, and concentrated WPPA, 53.3% P2O5 (see Table 16) (11, 40).
Mixed acids are then partially neutralized with liquid or gaseous
anhydrous ammonia in an brick-lined acid reactor. In this agi-
tated atmospheric pressure tank, the mole ratio of NHsrHaPO^
is maintained at 1.3:1.0 to 1.5:1.0 (16, 39, 42-44). All phos-
phoric acid and approximately 70% of ammonia are introduced in
this vessel (45). In this molar range, ammonium phosphates are
most soluble, allowing further concentration of solution while
maintaining adequate flow characteristics (Figure 24). Heat of
reaction is used in this vessel to maintain a temperature of
100°C to 120°C and to evaporate excess water (39, 43). A slurry
which is primarily MAP and contains 18% to 22% water is produced
and flows through steam-traced lines to the ammoniator-granulator
(43). To assure no leakage from the reactor, the vessel is
ventilated with outside air. In theory, the reactor could be
designed without ventilation or atmospheric discharge, but in
practice, ventilation rates of 57 to 71 m3/min (standard condi-
tions) are common. Ventilation rate is determined by reactor
mechanical design, not process requirements (45) . Ammonia-rich
offgases from the reactor at 77°C to 82°C are wet scrubbed
before exhausting to the atmosphere (45). Primary scrubbers use
raw material-mixed acids as scrubbing liquor, and secondary
scrubbers use gypsum pond water as scrubbing liquor.
The basic rotary-drum ammoniator-granulator, Figure 26, consists
of an open-end, slightly inclined rotary cylinder with retaining
rings at each end and a scraper or cutter mounted inside the drum
(42) Shreve, R. N. Chemical Process Industries, Third Edition.
McGraw-Hill Book Company, New York, New York, 1967.
pp. 274-277.
(43) Chopey, N. P. Diammonium Phosphate: New Plant Ushers in
Process Refinements. Chemical Engineering, 69(6):148-150,
1962.
(44) Vandegrift, A. E., L. J. Shannon, E. W. Lawless, P. G. Gor-
man, E. E. Sallee, and M. Reichel. Particulate Pollutant
System Study, Vol. 3—Handbook of Emission Properties. APTD-
0745 (PB 203 522), U.S. Environmental Protection Agency,
Durham, North Carolina, 1971. pp. 313-335.
(45) Hardison, L. C. Air Pollution Control Technology and Costs
in Seven Selected Areas. EPA-450/3-73-010 (PB 231 757),
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, December 1973. pp. 11-192.
56
-------
Ul
GYPSUM
POND WATER — •
•
FILTERED PHOSPHORIC ACID
p
' SECONDARY SECONDARY ™ w
. SCRUBBER SCRUBBER _
T
PRIM
1
1ARY PRIMAR
BBER SCRUBS!
4
!-,
SULFURIC ACID •• 1
ACID SURGE
— J
Y
P a
OND
AT£R 'SECONDARY
SCRUBBER
JL
PRIMARY
SCRUBBER
*
•« — POND WATER
fl
1 AMWMIATOa
I GRAKUUTOR
I i | ^v r
NE
v/
\ tm r-.
V \
T CYaoNE T
OVERSI
TANK
ANHYDROUS AMMONIA
REACTOR
FUE1,AIR
ROTARY DRYER
ROTARY COOLER
SCREENS
PRODUCT TO STORWJE.
BAGGING. OR BULK SHIPMENT
OUST SUPPRBSANT
UNDERSIZE
Figure 25. TVA ammonium phosphate process flow diagram.
-------
Figure 26. TVA rotary ammoniator-granulator (5).
shell. Drums vary in diameter from 2 m to 3 m and in length from
3 m to 6 m. A rolling bed of recycled solids is maintained in
the unit; slurry from the reactor is distributed above the bed
while the remaining ammonia (approximately 30%) is sparged under-
neath to bring the final Nf^iHsPO^ mole ratio from 1.8:1.0 to
2.0:1.0 (5, 45). Granulation by agglomeration and by coating
particles with slurry takes place in the rotating drum and is
completed in the dryer. Recycle rates of 2.5 to 4.0 kg recycle/
kg product are typical for this type of unit (39). As with the
reactor, the granulator theoretically could be designed without
ventilation, but to prevent NH3 leakage, approximately 8.5 x 10~u
m3 (standard conditions) per metric ton P20s air inleakage into
the granulator around inlet and outlet connections is allowed
(45) .
Temperature of granular DAP in the rotary drum reaches 85°C to
105°C, while temperature of offgases reaches 38°C to 77°C (5, 43,
45). Ammonia-rich offgases pass through a wet scrubber before
exhausting to the atmosphere.
Moist DAP granules are transferred to a rotary oil- or gas-fired
cocurrent dryer which reduces product moisture content to below
2%, and then product is cooled to below 35°C. Cooling minimizes
caking and product dissociation during storage (see Table 15)
58
-------
(43, 46). Temperature of offgases from the dryer ranges from
82°C to 104°C, and temperature of offgases from the cooler ranges
from 4°C to 27°C (5, 45). Before exhausting to the atmosphere,
these offgases pass through cyclones and wet scrubbers.
Cooled granules pass to a double-deck screen in which oversize
and undersize particles are separated from product-sized parti-
cles (42, 47). Some plants screen the product before cooling
(42, 44). DAP product ranges in granule size from 1 mm to 4 mm,
with a typical product size distribution presented in Figure 27
(5, 48). The oversize are crushed, mixed with the undersize, and
recycled to the ammoniator-granulator. To reduce DAP dustiness,
some manufacturers coat product granules with 0.5% by weight of
10-wt lubricating oil using a rotating dust suppressant system
similar to that shown in Figure 28 (46, 49). DAP is either
stored, bagged, or bulk loaded for shipment.
b. Dorr-Oliver Process—
A general process flow diagram of the Dorr-Oliver process is pre-
sented in Figure 29. Phosphoric acid (24% to 36% P20s) (37) or
a mixture with sulfuric acid is fed to a series of agitated reac-
tors in which acids react with liquid or gaseous anhydrous ammo-
nia feed. The bulk of the reaction takes place in the first
reactor, with additional vessels used for pH adjustment of result-
ing slurry (37) . Reactor offgases are scrubbed with raw phos-
phoric acid feed prior to exhausting to the atmosphere (17).
Thick slurry from the final reactor flows to a pugmill (blunger)
where recycled fines are added and product is granulated (39,
40). A blunger, Figure 30, is an inclined vessel with parallel
contrarotating shafts having blades to facilitate slurry mixing
and progress through the vessel. Recycle ratios range from 6 to
12 kg recycle/kg product (37, 39). These ratios are higher than
those for processes having further ammoniation during granulation
for two reacons: 1) less water is evaporated in the blunger
(46) Achorn, F. P., and H. L. Balay. Systems for Controlling
Dust in Fertilizer Plants. In: TVA Fertilizer Conference,
Tennessee Valley Authority Bulletin Y-78, Muscle Shoals,
Alabama, August 1974. pp. 55-62.
(47) Phosphate Fertilizer Plants Final Guideline Document
Availability. Federal Register, 42(40):12022-12023, 1977.
(48) Hoffmeister, G. Quality Control in a Bulk Blending Plant.
In: TVA Fertilizer Bulk Blending Conference, Tennessee
Valley Authority Bulletin Y-62, Muscle Shoals, Alabama,
August 1973. pp. 59-70.
(49) Barber, J. C. Environmental Control in Bulk Blanding Plants.
1. Control of Air Emissions. In: TVA Fertilizer Bulk
Blending Conference, Tennessee Valley Authority Bulletin
Y-62, Muscle Shoals, Alabama, August 1973. pp. 39-46.
59
-------
100
ee
°UJ
*
40
S o
ct
UJ
a: '
uj
a.
20
3.5 3.0 2.5 2.0 1.5 1.0 0.5
PARTICLE DIAMETER, mm
6 7 8 9 10 12 14 16 20
TYLER SCREEN MESH
Figure 27. Cumulative screen analysis of DAP (43)
DAP GRANULES
DUST
SUPPRESSANT
MIXER
PRODUCT GRANULES
.Figure 28. Product dust control system (49)
60
-------
O-i
SUIFURIC
ACID
- PRODUCT TO STORAGE
BAGGING.BUU SHIPMENT
UNDERSI7E
Figure 29. Dorr-Oliver ammonium phosphate process flow diagram.
-------
MIXING BLADES
TOP VIEW
END VIEW
Figure 30. Diagram of pugmill (blunger); top and end views (1).
Reprinted from Riegel's Handbook of Industrial Chemistry
by courtesy of Litton Educational Publishing, Inc.
during granulation and 2) at a lower NH3:H3P04 mole ratio, prod-
uct slurry has higher solubility (see Figure 24) (39).
Slurry-coated granules are then dryed in a cocurrent rotary dryer,
Product is then sized, e.g., 2.4 mm to 1.7 mm granules, and over-
size are crushed, mixed with undersize, and recycled to the
blunger (36). Product is sent to bulk storage for bagging or
bulk shipment. Offgases are vented to the atmosphere through a
cyclone and wet scrubber (16).
4. Industry Characterization
Recent production history of the ammonium phosphate fertilizer
industry is presented in Figure 31. Reported production data are
for MAP and DAP materials and their processed combinations with
ammonium sulfate. Ammonium phosphates produced in combination
with potash salts to make complete mixtures are excluded. Also
excluded are nitrophosphates, calcium metaphosphates, sodium
phosphates, and wet-base goods (made by treating phosphate rock
and some organic nitrogenous materials with sulfuric acid) (9).
All production and capacity data in this report are presented as
metric tons of PaOs. The relationship between metric tons of
P205 and metric tons of gross fertilizer product is a function of
fertilizer nutrient analysis and is therefore variable from plant
to plant and within each plant as a function of time. A general
conversion factor for the entire industry in 1975 was (see Appen-
dix A) (50-61).
Gross fertilizer (metric tons) = 2.49[P2C>5 (metric tons)]
(16)
(49) Inorganic Fertilizer Materials and Related Products.
M28B(75)-11, U.S. Department of Commerce, Washington, D.C.,
January 1976. 6 pp.
(50) Inorganic Fertilizer Materials and Related Products.
M28B(75)-12, U.S. Department of Commerce, Washington, D.C.,
February 1976. 6 pp.
(continued)
62
-------
65 66 67 68 69 70 71 72 73 74 75
YEAR
Figure 31.
Recent history of ammonium phosphate
capacity and production (7, 9-11, 34)
(52) Inorganic Fertilizer Materials and Related Products.
M28B(76)-1, U.S. Department of Commerce, Washington, D.C.,
March 1976. 6 pp.
(53) Inorganic Fertilizer Materials and Related Products.
M28(76)-2, U.S. Department of Commerce, Washington, D.C.,
April 1976. 6 pp.
(54) Inorganic Fertilizer Materials and Related Products.
M28(76)-3, U.S. Department of Commerce, Washington, D.C.,
May 1976. 6 pp.
(55) inorganic Fertilizer Materials and Related Products.
M28(76)-4, U.S. Department of Commerce, Washington, D.C.,
June 1976. 6 pp.
(56) Inorganic Fertilizer Materials and Related Products.
M28(76)-5, U.S. Department of Commerce, Washington, D.C.,
(continued)
63
-------
From 1965 to 1975, ammonium phosphate production grew from
0.983 x 106 metric tons P205 to 2.767 x 106 metric tons P205
(an annual growth rate of approximately 11%), while capacity grew
from 1.512 x 106 metric tons P205 to 4.926 x 106 metric tons
P20s (an annual growth rate of approximately 12%). Over that
period, plant utilization rates varied from 47% to 83%, ending in
1975 at 56%. For the period 1970 to 1975, the average annual
utilization rate was 73%.
In 1975, 35 companies in the United States operated 48 ammonium
phosphate plants in 17 states (see Appendix A). Distribution of
plants and capacity by state in Table 17 (7, 10, 11) indicates
that Florida is the largest ammonium phosphate-producing state
(25% of plants nationally having 43% of national capacity).
Florida and Louisiana, with 35% of ammonium phosphate plants,
have 67% of national capacity. As shown in Table 18, 8 of the
35 companies have an annual capacity of over 200,000 metric tons
P205; combined, they represent 64% of total national capacity.
A cumulative distribution of ammonium phosphate plants and capac-
ity in 1975 is presented in Figure 32. The distribution shows
that many small plants collectively represent a small fraction of
capacity while a few large plants represent a large fraction of
capacity. From the graph, 50% of the plants each have annual
capacity of less than approximately 65,000 metric tons P205,
but these plants represent only approximately 15% of total
national capacity. Conversely, 50% of national capacity is
represented by plants each having annual capacity of less than
approximately 180,000 metric tons P205. Approximately 83% of
plants are below this size. Mean plant capacity in 1975 was
103,000 metric tons P205.
(continued)
July 1976. 6 pp.
(57) Inorganic Fertilizer Materials and Related Products.
M28(76)-6, U.S. Department of Commerce, Washington, D.C.,
August 1976. 6 pp.
(58) Inorganic Fertilizer Materials and Related Products.
M28(76)-7, U.S. Department of Commerce, Washington, D.C.,
September 1976. 6 pp.
(59) Inorganic Fertilizer Materials and Related Products.
M28(76)-8, U.S. Department of Commerce, Washington, D.C.,
October 1976. 6 pp.
(60) Inorganic Fertilizer Materials and Related Products.
M28(76)-9, U.S. Department of Commerce, Washington, D.C.,
November 1976. 6 pp.
(61) Inorganic Fertilizer Materials and Related Products.
M28(76)-10, U.S. Department of Commerce, Washington, D.C.,
December 1976.
64
-------
TABLE 17. 1975 DISTRIBUTION OF AMMONIUM PHOSPHATE
CAPACITY BY STATE (7, 10, 11)
State
Florida
Louisiana
Texas
Idaho
Iowa
Mississippi
California
Illinois
North Carolina
Alabama
Missouri
Utah
Minnesota
Arkansas
Washington
Michigan
Arizona
Total
Capacity,
10 3 metric tons P^O*,
2,101
1,173
293
262
228
139
118
114
92
86
84
65
63
45
27
25
11
4,926
Percent of
national
capacity
43
24
6
5
5
3
2
2
2
2
2
1
1
1
<1
<1
<1
100
Number
of
plants
12
5
4
4
2
1
7
1
1
2
1
2
1
1
1
2
1
48
TABLE 18. COMPANIES HAVING AMMONIUM PHOSPHATE CAPACITY
>200,000 METRIC TONS P205 IN 1975 (7, 10, 11)
Company
Capacity,
103 metric tons P205
Percent of
national
capacity
CF Industries, Inc.
Williams Companies,
Agrico Chemical Co., Subsidiary
Beker Industries
Occidental Petroleum Corp.,
Occidental Chemical Co., Subsidiary
Gardinier, Inc.
Farmland Industries, Inc.
IMC Chemicals Corp.
Olin Corp.
827
729
328
300
272
248
227
209
17
15
7
6
6
5
5
4
65
-------
10 100
ANNUAL PLANT CAPACITY. 103 metric tons P.O.
1.000
Figure 32. Cumulative distribution of ammonium phosphate
plants and capacity in 1975 (7, 10, 11).
As previously mentioned, DAP production using TVA technology with
WPPA is representative of the ammonium phosphate industry. An
average DAP plant is similar to the one illustrated in Figure 25
and has average parameters. The average plant has a capacity of
103,000 metric tons/yr P2O5 and an average annual utilization
factor of 73%, yielding an annual production rate of 75,000
metric tons P20s (Appendix A) .
Ammonium phosphate production facilities are located in counties
with population densities ranging from 1 person/km2 to 1686
persons/km2 (Appendix A) .
The average plant is located in a county with a population
density of 82 persons/km2 based on a plant capacity weighted
average.
66
-------
SECTION 4
AIR EMISSIONS
A. WET PROCESS PHOSPHORIC ACID
Production of WPPA generates a variety of gaseous and particulate
emission species. These emissions arise from five unit oper-
ations in the production process: rock unloading, rock storage
and conveying, acidulation, filtration, and evaporation. These
unit operations, however, release emissions to the atmosphere
in only three locations, as shown in Figure 33: rock unloading,
rock storage and conveying, and wet scrubber system. In this
study, phosphoric acid production was defined to begin with the
unloading of ground rock; however, most large plants in Florida
grind their rock on site.
PHOSPHATE ROCK
*- EMISSIONS
«- EMISSIONS
H2S04
GYPSUM SLURRY
TO POND
EMISSIONS
Figure 33.
PRODUCT
Schematic of emission points in WPPA manufacture,
67
-------
Another source of air emissions at phosphate fertilizer plants
is the gypsum pond. Water-soluble fluoride compounds are sepa-
rated from phosphate rock in the reactor, and a portion is
carried to the gypsum pond along with calcium sulfate from
the filtration operation. Volatile fluorine compounds evolve
from the pond at variable rates depending on gypsum pond
characteristics.
1. Raw Materials Handling
Ground phosphate rock transported to the plant by railroad
hopper cars or hopper trucks is delivered to rock storage bins
and elevated feed bins by combination screw conveyors, bucket
elevators, belt conveyors, and pneumatic conveyors. Elevated
feed bins allow use of gravity flow to batch weigh hoppers. A
small fixed hopper and oversized screw conveyor convert the
batch weighings to a uniform feed to the reactor. To properly
control rock dust emissions, conveyors, feeders, hoppers, and
storage bins are enclosed and vented to dust abatement equipment,
typically a baghouse. The unloading shed is also enclosed and
equipped with a bag collector for rock recovery and particulate
emissions control.
Phosphate rock is ground to 60% to 80% less than 74 ym (minus
200 mesh) for WPPA manufacture. Because no reaction has taken
place, the particulate composition is that of the raw material,
phosphate rock (17, 22).
Limited data exist on emissions from baghouses associated with
rock handling at production facilities. However, some data con-
cerning these emissions, available in public files from the
Florida Department of Environmental Regulations, are tabulated in
Appendix B.
The controlled particulate emission factor for rock unloading is
0.15 g/kg P20s ± 250% based on averaging data in Appendix B.
Uncertainty associated with the emission factor is calculated
using the "Student t" test at a 95% confidence level.
For rock transfer and charging to the reactor, the controlled
emission factor ranges from 0.012 to 0.10 g/kg P20s with an aver-
age value of 0.045 g/kg £2®$ ± 180% (see Appendix B for data).
The average value and standard deviation for the height of rock
unloading emissions is 12 ± 3 m. For rock transfer, the average
value is 21 ± 6 m (Appendix B). These values do not necessarily
represent stack heights, but an elevated point in the plant where
particulates are exhausted. These values will hereafter be
referred to as stack heights.
68
-------
2. Wet Scrubber System
Three operations responsible for creating emission species are
discussed concurrently in this section: phosphate rock acidula-
tion, filtration and evaporation. To comply with strict
criteria governing emissions, particularly of fluoride compounds,
all phosphoric acid plants employ various types of wet scrubbers
as control devices. Plants for which emissions data were avail-
able have these three unit operations housed under one roof, with
one wet scrubber collecting emissions from the operations. For
this reason, one controlled emission factor for each emission
species is obtained for the multiunit process, based on an aver-
age vent height for the wet scrubber system of 29 m (Appendix B) .
The sources and species of emissions are described below.
a. Fluoride —
Gaseous fluoride emissions consist of silicon tetraf luoride gene-
rated in the reaction and evaporation processes. Hydrogen fluor-
ide formed in the reactor is converted to SiFt, according to the
reaction (45) :
4HF + Si02 - >• 2H20 + SiFt, (17)
The reaction favors the formation of SIF^ at temperatures lower
than 100°C.
Phosphate rock typically contains 3.0% to 4.0% (by weight) fluor-
ine which is variably distributed in the product acid, gypsum
slurry, and gaseous emissions (20) . Table 19 shows two material
balances depicting final distributions of the fluorine from the
rock. To reduce air emissions, the plants utilize wet scrubbers.
Silicon tetraf luoride is removed through reaction with water to
form aqueous fluosilicic acid, and hydrogen fluoride is removed
from the gaseous stream in the form of aqueous hydrofluoric acid
and silicon tetraf luoride.
In 14 plants that represent approximately 50% of total phosphoric
acid production, fluorine is recovered in the form of fluosilicic
acid, fluorides, f luosilicates , or byproducts (7). The other 22
plants regard the fluorine materials as waste and pump the
fluorine-laden scrubbing water with the gypsum slurry to the
settling pond. Consequently, emission factors for total fluorine
from the scrubber's gaseous exhaust stream were divided into two
groups based on whether or not fluorine recovery was practiced
(Appendix B) . Comparison of the two sets of data indicate that
the emission factors are not significantly different. For ex-
ample, two plants without fluorine recovery have emissions of
0.0033 and 0.0042 g/kg P20s, which compares with two plants with
recovery of fluorine which have emission factors of 0.0033 and
0.0055 g/kg P20s. One plant recovering fluorine has an emission
factor of 0.011 g/kg P205 which compares with three plants not
recovering fluorine with emission factors of 0.012 g/kg P205 and
69
-------
TABLE 19. FLUORINE MATERIAL BALANCES FOR WPPA MANUFACTURE
Material
balance
Aa
B
Phosphate
rock
48.3
127
Fluorine,
Product
acid
36.5
16.3
106 g/day
Gypsum
slurry
and
process
H20
11.8
110.7
Air
emission
0.004
0.009
Plant
daily
production,
metric
tons P2C>5
368
907
Fluorine
emission
f.-.ctor,
g/-kg P20S
0.011
0.010
Data obtained from the public files at the Florida Department of Environ-
mental Regulations in Winter Haven, October 1976.
Data from Reference 62.
one with 0.011 g/kg P20s- One plant not recovering fluorine has
a reported emission factor of 0.035, which is high. However,
this is a very small plant with a capacity of only 6 metric tons
per hour P205 and is no doubt an old plant with a less efficient
scrubber. Emission factors probably depend more on the type and
efficiency of scrubber used, scrubber operation, and the use of
fresh water tail gas scrubbers than on whether fluorine recovery
is practiced. Plants practicing fluorine recovery send less
volatile fluorine to their pond systems and might have lower
total fluorine emissions from their ponds.
An average emission factor for the wet scrubber system was calcu-
lated by averaging data from nine plants with 15 trains (Appen-
dix B) with emission factors from the two material balances shown
in Table 19. Controlled emission factors at individual plants
range from 0.0025 to 0.035 g/kg P2O5. The average fluorine emis-
sion factor for the wet scrubber system, calculated by averaging
all industry data, is 0.01 g/kg P2°5 ± 40%.
b. Particulate—
Particulate emissions generated in the reactor consist of
unreacted phosphate rock, with lesser amounts of insoluble phos-
phate salts and calcium sulfate. This dust is physically en-
trained in reactor gases vented to the scrubber. Lack of data
precludes estimating the relative amounts of species in particu-
late emissions. Some particulate matter contains silica (Si02)
which is formed when silicon tetrafluoride reacts with water to
(62) King, W. R., and J. K. Ferrell. Fluoride Emissions from
Phosphoric Acid Plant Gypsum Ponds. EPA-650/2-74-021, U.S,
Environmental Protection Agency, Research Triangle Park,
North Carolina, October 1974. 329 pp.
70
-------
form fluosilicic acid and silica. The fact that these emissions
are insoluble in water partially explains their existence in the
scrubbed vapor streams.
Source test measurements for particulate emissions range from
0.0011 to 0.17 g/kg P205 as shown in Appendix B. The average
emission factor is 0.054 g/kg P20s ± 164% based on data from five
plants representing 16% of total U.S. production.
c. Sulfur Oxides—
The origin of SOX emissions in WPPA manufacture is not clear.
The emissions can result from dissolved sulfur dioxide in the
sulfuric acid or from reactions of the phosphate rock with sul-
furic acid (12). These gases are rarely measured at acid plants.
Data from a Public Health Service document (12) and from one
plant reporting SOX emissions (Appendix B) gave a range of emis-
sion factors of 0.0077 to 0.058 g/kg ?205 (see Appendix B). An
average of these figures gives an emission factor of 0.032 g/kg
P205 ± 240%.
d. Phosphates—
Phosphate emissions consist of phosphate rock, various phosphates,
and phosphoric acid mist. During particulate analysis of stack
gases, all of these emission species are collected, with various
efficiencies, on the filter paper.
Emissions data were obtained from one WPPA plant. In this series
of three tests (Appendix B), the filter paper was removed and
the particulates and gases were passed through three water-filled
gas bubblers. The solution was then analyzed for total phospho-
rus content and reported as grams of P205 per kilogram of P20s
produced.
Comparison of these three source test measurements at one plant
with the range of total particulates emitted at the other plants
indicates that approximately 80% of the particulate matter con-
sists of water-soluble phosphorus compounds.
Because phosphate emissions are in particulate form, phosphate
emission factors were not separately calculated; they are
included with the particulate emission factor.
3. Gypsum Pond Emissions
Emissions of volatile fluorine, hydrogen fluoride, and silicon
tetrafluoride from gypsum ponds have been the subject of numerous
71
-------
studies (20, 62-65). An EPA report (20) presents a critical
review of the major studies reporting gypsum pond fluoride
emissions.
After close scrutiny of the data, emissions from gypsum ponds
were found to range from 11 to 1,100 kg F/(km2-day) [0.1 to
10 lb/(acre-day)] with an average value of 220 kg F/(km2-day).
This results in an emission factor of 0.025 to 2.5 g F/kg of
P205 for an average plant producing 486 metric tons of P20s with
a typical gypsum pond of 1.11 km2. The average emission factor
is 0.50 g F/kg of P205 (20).
At .the end of.August 1977, a field program was carried out near
Bartow, Florida, with the cooperation of EPA for measuring
fluoride emissions from a gypsum pond (66) . Average fluoride
emission rates from the pond were estimated to be in the range
of 440 to 1,100 kg F/(km2-day) [4 to 10 lb/(acre-day)]. Data
collected by remote optical sensing indicate that fluoride emis-
sions from the gypsum pond consisted entirely of hydrogen fluor-
ide. The silicon tetrafluoride concentration was below the
detectable threshold of 0.5 ppb. Results from this study, how-
ever, are still preliminary and may be subject to change in the
final report.
4. Emission Summary
Emission factors and stack heights for WPPA manufacture are
summarized in Table 20 for each emission point. The correspond-
ing errors are based on the "Student t" test at 95% confidence
(67). Data used to generate this table are presented in
Appendix B.
(63) English, M. Fluorine Recovery from Phosphatic Fertilizer
Manufacture. Chemical Process Engineering, 48 (12) :43-47,
1967.
(64) Bowers, Q. D. Disposal as Waste Material—U.S. Practice.
in: Phosphoric Acid, Volume I, A. V. Slack, ed. Marcel
Dekker, Inc., New York, New York, 1968. pp. 505-510.
(65) Huggstutler, K. K., and W. E. Starnes. Sources and
Quantities of Fluorides Evolved with the Manufacture of
Fertilizer and Related Products. Journal of the Air Pollu-
tion Control Association, 11 (12):682-684, 1966.
(66) Preliminary Report: Remote Monitoring of Fluoride Emission
from Gypsum Ponds. EPA-69/01-4145, Task 10, U.S. Environ-
mental Protection Agency, Washington, D.C., November 1977.
35 pp.
(67) Volk, W. Applied Statistics for Engineers, Second Edition.
McGraw-Hill Book Co., New York, New York, 1969. 110 pp.
72
-------
TABLE 20. AVERAGE STACK HEIGHTS AND CONTROLLED EMISSION
FACTORS FOR WET PROCESS PHOSPHORIC ACID AND
SUPERPHOSPHORIC ACID PLANTS
Emission point
Stack
height,
m
Emission factor, g/kg
Total
fluoride Particulate
PaOs
SOx
Wet process phosphoric acid:.
Rock unloading 12
Rock transfer and conveying 21
Wet scrubber system: 29
Gypsum pond
Superphosphoric acid:
Wet scrubber 21
0
0
0.010 ± 40%
0.025 to 2.5
avg 0.50
0.0073'
0.15 ± 250% 0
0.045 ± 180% 0
0.054 ± 164% 0.032 ± 200%
0.011 to 0.055
Only two data points.
B. SUPERPHOSPHORIC ACID
The most popular process (at about 75% of existing plants) for
dehydration of 54% P205 phosphoric acid to produce greater than
66% P20s Superphosphoric acid involves the use of heat transfer
surfaces. Although some (approximately 25%) manufacturers use
submerged combustion, its large volume of effluent gases makes
this process unattractive due to the cost of extensive scrubbing
facilities. Expansion of this process is unlikely (31). Conse-
quently, only vacuum evaporation processes are evaluated in this
report.
Emission species from superphosphoric acid plants include fluo-
rine compounds and particulates. Fluorine is evolved in the
form of hydrogen fluoride. Particulates are limited to liquid
phosphoric acid aerosols and mists produced by the condensation
process. The falling film evaporator (see Section 3) can gener-
ate aerosols which are submicrometer in size (45).
Two plants for which fluorine emissions data were available use
vacuum evaporation processes. The barometric condenser, hot
well, and product cooling tank are vented to a two-state wet
scrubber. Fluorine emission factors from these plants are 0.0036
and 0.011 g/kg P205, with an average value of 0.0073 g/kg P205
(Appendix B).
One plant reported particulate emissions ranging from 0.011 to
0.055 g/kg P205.
The average stack height for the plant emissions is 21 m (Appen-
dix B). Emission factors and stack height for superphosphoric
acid manufacture are included in Table 20.
73
-------
C. NORMAL SUPERPHOSPHATE
Emission points at NSP production facilities include the mixer,
den, and curing building. Emissions are also generated by mate-
rials storage and handling operations. A list of emission
points at an average plant and corresponding emission species
follows:
• Ground rock unloading and feeder system—particulate.
• Mixer and den—fluoride compounds and particulate.
• Curing building—fluoride compounds and particulate.
Particulate emissions from materials storage and handling opera-
tions result from unloading hopper-bottom railroad cars and
transporting the ground phosphate rock to the superphosphate
plant by screw conveyors, belt conveyors, and bucket elevators.
Additional emissions issue from the product storage and curing
building as a result of fertilizer handling and shipping opera-
tions within the building. Typical composition analyses of
superphosphate fertilizers are given in Table 21 (13, 18, 20).
Concentrations of radioactive elements in phosphate fertilizer
products were reported in Table 4.
Fluorides enter the NSP production process in the phosphate rock
and are released as a result of the acidulation reaction. During
acidulation, the calcium fluoride content of the rock is attacked
by the acid (sulfuric or phosphoric), resulting in formation of
hydrofluoric acid. This in turn reacts with silica found in the
rock to form silicon tetrafluoride which hydrolyzes to form
fluosilcic acid. The reaction sequence leading to the formation
of fluosilicic acid is given below:
Phosphate rock + acid —>• HF (18)
4HF + Si02 >• SiF^ + 2H20 (19)
3SiF4 + 2H20 > 2H2SiF6 + Si02 (20)
Some of the hydrogen fluoride and silicon tetrafluoride are vola-
tilized during the process leading to fluoride emissions. Fluo-
ride vapors that evolve as hydrogen fluoride and silicon tetra-
fluoride are released from the mixer, den, and curing building.
Fluorine is also present as a constituent of the rock and ferti-
lizer particulate matter. Between 1.5 kg and 9.0 kg of fluorides
per metric ton of NSP (Appendix C) are relased during the pro-
duction and curing operations. Emissions of fluoride and par-
ticulate from the mixer and den are controlled by scrubbing with
water. Scrubber liquor may be recirculated pond water or a weak
solution of fluosilicic acid. Nearly two-thirds of the NSP
plants presently practice fluorine recovery, thereby eliminating
or greatly reducing the need for a pond. No measurements are
available for fugitive fluoride emissions from those NSP plants
that make use of a pond system, but such emissions will be less
than fluoride emissions from those gypsum ponds discussed in the
section on WPPA manufacture.
74
-------
TABLE 21. TYPICAL CHEMICAL COMPOSITION OF FLORIDA
NORMAL SUPERPHOSPHATE AND TRIPLE SUPER-
PHOSPHATE FERTILIZER (13, 18, 20)
a
Component
Aluminum
Arsenic
Ash (acid-insoluble)
Boron
Calcium, total
Calcium, water soluble
Carbon, organic
Carbon dioxide
Chlorine
Chromium
Cobalt
Copper
Fluorine
Free acid
Free acid-free water ratio
Iodine
Iron
Lead
Lithuim
Magnesium, total
Magnesium, water soluble
Manganese
Molybdenum
Nitrogen
Phosphorus, total
Potassium
Selenium
Silicon
Silver
Sodium
Sulfur, total
Sulfur, water soluble
Titanium
Vanadium .
Water, reported as "moisture"
Water, free
Water of crystallization
Zinc
Expressed
as
A1203
As
Ash
B
CaO
CaO
C
C02
Cl
cr
Co
Cu
F
H3P04
H3P01)/H20
I
Fe203
Fb
Li
MgO
MgO
Mn
Mo
N
P205
K2O
Se
Si02
Ag
Na20
S03
S03
Ti
X.
H^O
H20
H20
Zn
J.NSP contentl*
Units
percent
ppm
percent
ppm'
percent
percent
percent
percent
percent
ppm
ppm
ppm
percent
percent
ppm
percent
ppm
ppm
percent
percent
ppm
ppm
percent
percent
percent
ppm
percent
ppm
percent
percent
percent
ppm
ppm
percent
percent
percent
ppm
Range
0.21 to 1.16
4.1 to 30.6
.2.00 to 13.65
<3 to 30
27.20 to 31.13
10.19 to 14.90
0.21 to 0.27
0 to 0.44
C
70 to 72
0 to 2.8
28 to 64
1.41 to 2.15
1.30 to 2.15
0.12 to 1.19
16 to 50
0.38 to 1.37
8 to 20
_C .
0.04 to 0.12
C
65 to 95
.C
C' .
16 to 21
0.16 to 0.24
0 to 1.5
4.00 to 4.54
15 to 20
0.09 to 0.13
26.58 to 30.55
6.37 to 13.49
54 to 270
20 to 71
2.3 to 8.3
1.09 to 5.71
2.44 to 5.14
50 to 200
Average
0.72
12.5
4.45
11
29.52
13.10
0.24
0.066
0.80
71
1.3
47
1.74
1.71
0.58
33
0.67
14
2
0.07
0.03
77
1.6
0.1
20
0.20
0.6
4.35
18
0.11
28.99
10.67
162
46
5.64
3.65
3.55
134
TSP content-0
Range
1.20 to 1.95
10.5 to 14.3
2.50 to 4.90
29 to 115
16.60 to 21.57
14.60 to 16.80
0 to 0.22
<0.1
0 to 890
2.4 to 4.8
3 to 22
2.00 to 3.49
0.19 to 3.85
0.06 to 1.59
0.92 to 2.00
0 to 65
0.05 to 1.00
110 to 300
3.7 to 16.8
0.06 to 0.40
45 to 49
0 to 0.57
C
0.60 to 7.37
0.13 to 1.79
2.12 to 4.95
1.65 to 5.77
0 to 599
0 to 3,875
0.87 to 6.30
0.88 to 4.42
1.29 to 6.26
0 to 320
Average
1.68
12.2
3.55
80
19.65
0.11
513
3.4
11
2.47
2.6
0.8
1.59
26
0.38
214
8.0
0.26
48
0.35
<0.8
4.42
0.97
3.01
2.98
300
2,515
3.4
2.57
3.47
102
a
Radium, uranium and thorium are reported in Table 4.
Blanks indicate component not analyzed.
Average based on one to two measurements.
75
-------
Source test data from fertilizer plants were collected from
published literature and sampling data on file as of October 1976
at the Florida Department of Environmental Regulation in Winter
Haven. Raw data used to establish emission factors are given in
Appendix B.
Emission factors for the emission species at NSP plants as a
function of emission point are shown in Table 22. Emission
factors for the mixer-den and the curing building were calculated
by averaging the appropriate values in Appendix B. Data were
available for only one set of four tests for controlled fluoride
emissions from the product curing building. Because most (more
than 85%) curing buildings remain uncontrolled, the fluoride
emission factors were normalized to uncontrolled emissions using
the fluoride control efficiency of 97% reported by Plant A. The
low volumes of fertilizer materials handled by these storage
facilities and the decline in industry production levels for NSP
make control devices economically impractical.
TABLE 22. EMISSION FACTORS FOR AN AVERAGE NSP PLANT
BASED ON CONTROLLED EMISSION SOURCES
Emission factor, g/kg P20s
Emission source
Rock unloading
Rock feeding
Mixer and den .
Curing building
Particulates
0.28b
0.055 ± 180%
0.26 ±.86%
3.6b
Fluorides3
_c
~c
0.10 ±
1.9 ±
120%
120%
Fluoride released as a vapor.
Based on two sets of data; therefore 95%
confidence limits could not be determined.
Not emitted from this source.
Uncontrolled emission factors since curing
building emissions are not controlled at an
average plant.
Particulate emissions due to the rock unloading, storage, and
transfer operations and the fertilizer handling and shipping
activities occurring in the product curing building were not
available for NSP plants. Emission factors for the rock unload-
ing and storage activities and for the ground rock weighers and
feeders are developed in Appendix B from emission factors for
similar activities occurring at GTSP production facilities. In
order to obtain an estimate of the particulate emissions arising
from fertilizer handling and shipping operations occurring in
the curing building, two measurements for controlled particulate
emissions from the combined shipping, screening, and milling of
ROP-TSP were used (Appendix B).
76
-------
Error limits shown in Table 22 and developed in Appendix B were
established by applying a "Student t" test to the input data (66),
The "t" test is applied because the sample sizes are fewer than
30 in number and thus may not be normally distributed. The
statistical data used to establish the error limits are shown
in Appendix B.
As an aid in determining the reliability of reported fluorine
emission measurements, mass balances are developed in Appendix C
for the production of NSP. Between 7.5 g F/kg P205 and 45 g
F/kg P20s (depending on the fluoride concentration of the NSP
product) are released during the production and curing operations.
Based on data from the Florida Department of Environmental Regu-
lation, a scrubber control efficiency of 99% for fluoride removal
was used. Controlled fluoride emissions would then range from
0.07 g F/kg P205 to 0.45 g F/kg P205. This compares favorably
with our values of 0.1 g F/kg P205 and 0.05 g F/kg P205 developed
for controlled emissions from the mixer-den and curing building,
respectively (Table 22).
D. TRIPLE SUPERPHOSPHATE
1. Run-of-the-Pile Triple Superphosphate
The process for production of ROP-TSP is similar to that for
NSP. Emission points and emission species therefore closely
resemble those from NSP production facilities; namely,
• Ground rock unloading and feeder system—particulate.
• Mixer and den--fluoride compounds and particulate.
• Curing building--fluoride compounds and particulate.
• Gypsum pond—fluoride compounds.
TSP manufacture differs from that of NSP in that WPPA is used
for acidulation in place of sulfuric acid. As a result, fluo-
rides enter the TSP production process not only as a constituent
of the rock but also as an impurity in the phosphoric acid.
Emissions of fluorides are controlled by wet scrubbers that dis-
charge a fluoride-containing wastewater stream to holding ponds.
Water in the ponds is recycled for use in the scrubbers. Gaseous
fluoride is also emitted from the ponds used as reservoirs to
hold contaiminated scrubber water. The development of emission
factors for the gyspum ponds is covered under WPPA manufacture,
and will therefore not be considered here.
Emission factors for the emission species from ROP-TSP plants are
given in Table 23. The raw data used to compile these factors
are presented in Appendix B.
77
-------
TABLE 23. EMISSION FACTORS FOR AN AVERAGE ROP-TSP PLANT
BASED ON CONTROLLED EMISSION SOURCES
Emission factor/ g/kg
Emission source Participates
Rock unloading 0.07
Rock feeding 0.014 ± 170%
Cone mixer, den, curing building 0.16 ± 50%
Fluorides3
_c
0.10 ± 40%
Fluoride released as a vapor.
Based on two sets of data; therefore 95% confidence limits
could not be calculated.
CNot emitted from this source.
The fluoride emission factor in Table 23 was averaged from source
test data available for Plants A and B, Appendix B. Fluoride
emissions data from Plant C did not take into account emissions
from the curing building and were not included in the averaging
procedure. Emissions from the mixer, den, and curing building at
a typical plant are vented to a common stack; therefore, individ-
ual emission factors for each source were not developed.
In order to estimate particulate emissions for mixing-denning-
curingrshipping operations, source test data for mixing-denning
and screening-milling at Plant C (Appendix B) were utilized.
Particulate emissions data from fertilizer screening and milling
operations were used in deriving the curing building emission
factor, because these activities represent the major source of
particulates from a curing building. Particulate emission
factors for the ground rock unloading and transfer operations
were developed from Appendix B using emission factors for
similar activities occurring at GTSP production facilities.
An estimated 8 g F/kg Pz°5 are released during the production and
curing of ROP-TSP. This value is based on a material balance
developed in Appendix C. A scrubber efficiency of 99% would
result in a controlled emission factor of 0.08 g F/kg PaOs-
This value can be compared to the average controlled emission
factor of 0.10 g F/kg P2Os based on actual source tests.
2 . Granular Triple Superphosphate
Five plant operations release emissions at TSP plants using the
Dorr-Oliver direct granulation process. They are described in
detail in Section 3. The emission points and the emission
species associated with each are as follows:
78
-------
• Ground rock unloading and feeder system—particulate.
• Reactor and granulator—fluoride compounds and particulate.
• Dryer and cooler—SOX, fluoride compounds, and particulates.
• Screens and oversize mills—particuiate.
• Storage and shipping—fluoride compounds and particulate.
Fluorides enter the TSP process in the phosphate rock and the
WPPA and are volatilized and evolved during the acidulation
reaction. Evolution of fluoride vapors continues throughout the
manufacturing process and during storage as the reaction proceeds
to near completion. Emissions of fluorides are in the form of
the water-soluble gases, silicon tetrafluoride, and hydrogen
fluoride. Fluorine is also released as a constituent of the rock
and fertilizer particulate matter.
An estimated 7 g of fluoride vapors per metric ton of GTSP
(Appendix B) are released during production and curing. The con-
trol of fluoride emissions is accomplished by scrubbing the
exhaust gas streams with recycled pond water. Fluoride emissions
from gypsum ponds are considered in the section on the manufac-
ture of WPPA.
In addition to fluoride compounds and dust particles, the dryer
exhaust contains SOX. These emissions result from the combustion
of fuel oil containing sulfur.
To calculate emission factors, source test data from GTSP plants
were collected from published literature and sampling data on
file at the Florida Department of Environmental Regulation in
Winter Haven. The raw data used to establish emission factors
are given in Appendix B.
Emission factors at GTSP plants as a function of emission point
are shown in Table 24. Emissions from the reactor, granulator,
dryer, cooler, screens, and mills at an average plant are vented
to a common stack. As a result, individual emission factors were
not developed for separate segments of the production process.
There are no source test data for SOX emissions from the dryer.
Estimates of uncontrolled SOX emissions were calculated by
Plants A and E (Appendix B) on the basis of fuel oil consumption
and sulfur content.
A check on the reliability of fluoride emission measurements can
be made by comparing the estimated fluoride release based on a
mass balance. On this basis (Appendix C), an estimated 15.2 g
F/kg ?205 are released during the production and curing of GTSP.
A scrubber efficiency of 99% would result in a controlled emis-
sion factor of 0.152 g F/kg P205. This can be compared with the
79
-------
TABLE 24. EMISSION FACTORS FOR AN AVERAGE GTSP PLANT
BASED ON CONTROLLED EMISSION SOURCES
Emission factor, g/kg P20s
Emission source Particulates Fluorides^ SOX
Rock unloading 0.09 -c -c
Rock feeding 0.017 ± 180% - -c
Reactor, granulator,
screens
Curing building
0.05 ± 320%
0.10 ± 240%
0.12 ± 30%
0.018 ± 40%
1.86d
_c
Fluoride released as a vapor.
Based on two sets of data; therefore, 95% confidence
limits could not be calculated.
:1
dr
Not emitted from this source.
Worst case estimate based on fuel oil sulfur content.
controlled emission of 0.156 g F/kg P2O5 developed by adding
average measured values of 0.099 g F/kg and 0.57 g F/kg from the
reactor-den and curing building, respectively.
E. AMMONIUM PHOSPHATES
Air emissions from production of ammonium phosphate fertilizers
by ammoniation-granulation of phosphoric acid and ammonia result
from six process operations. Emission sources and their related
emission species are:
Reactor--ammonia, fluorides.
Ammoniator-granulator—ammonia, fluorides, particulates.
Dryer—ammonia, fluorides, particulates, combustion gases.
Cooler—ammonia, fluorides, particulates.
Product sizing and material transfer—particulates.
Gypsum pond—fluorides.
Ammonia emissions are volatilized from the reactor and ammoniator-
granulator due to incomplete chemical reactions and excess free
ammonia. Ammonia emitted from the dryer and cooler is due to
dissocation of fertilizer product. Particulate emissions result
from entrainment of MAP and DAP dusts in ventilation air streams.
Particulate emission species may also include ammonium fluoride
and ammonium fluosilicates (45) .
Fluoride emissions originate from the fluoride content of phos-
phoric acid. Air emissions are formed based on the following
set of equilibrium reactions:
80
-------
H2SiF6 +± 2 HF + SiF4 (21)
4 HF + Si02 ^ H20 + SiF^ (22)
At operating temperatures associated with MAP and DAP production,
emissions of silicon tetraf luoride are favored over hydrogen
fluoride (45) .
Dryer offgases contain natural gas or fuel oil combustion prod-
ucts. EPA found combustion product pollutants in such minor
concentrations that they were dismissed from consideration during
EPA's development of background information for air standards for
the phosphate fertilizer industry (31) . Therefore, these emis-
sion species will not be considered further in this study.
Emissions from the first five emission points reach the atmos-
phere through a stack, while gypsum pond emissions are fugitive.
Although there are six emission sources, there may be fewer
emission points because some plants combine flue gases from
multiple sources for subsequent emission control.
Emission factors were developed for air emission species from
each emission point from data in published literature and from
sampling data on file at the Florida Department of Environmental
Regulation, Winter Haven, Florida. Raw data used to calculate
emission factors were compiled and are presented in Appendix B.
Emission factors are reported in the literature in units of grams
per kilogram of P205 input, grams per kilogram of P20s output,
and grams per kilogram of product. All P205, except losses due
to emissions, is assumed to reach the product. Therefore, input
and output emission factors are equivalent. For those emission
factors expressed as grams per kilogram of product, a 46% P205
content was assumed. All emission factors developed for this
study are expressed in units of grams per kilogram of P2C>5.
Emission factors presented in Table 25 were calculated by averag-
ing appropriate values from Appendix B. Due to the nature of
both emissions data and pollution control practices at plants,
emissions from the reactor and ammoniator-granulator were com-
bined and reported as from one emission point. Dryer and cooler
emissions were treated in the same manner. Table 25 also shows
95% confidence intervals associated with each emission factor as
calculated by the "Student t" method.
As Appendix B indicates, 53% of the raw data are from plants
which collectively report all air emissions as "total plant"
emissions. Therefore, total plant emission factors were calcu-
lated from these data and are also shown in Table 25. Because
emission factors for individual emission species from the three
81
-------
process-related emission points are similar in magnitude to those
reported as total plant emissions, a total plant emission factor
for each emissions species was calculated from all data in
Appendix B according to the following equation:
"Total
_ (ER/A + ED/C * EP)(NR/A + ND/C + Np)
TABLE 25.
ETP NTP
N
R/A + ND/C
N
N
TP
EMISSION FACTORS DEVELOPED FROM SOURCE
TEST DATA GIVEN IN APPENDIX B
(23)
a
Emission point
Reactor/ammoniator-granulator :
Fluoride (as F)
Particulate
Ammonia
Dryer/cooler :
Fluoride (as F)
Particulate
Ammonia
Product sizing and material transfer:
Fluoride (as F)
Particulate
Ammonia
Reported as total plant emissions :
Fluoride (as F)
Particulate
Ammonia
Controlled
Mean,
g/kg P20s
0.023 S^'
0.76
_b .oi\
0.015 &§^o
0.75 .\~;
_b ,®>9^=
0 . 001 i ^
0.03
_b ,00;
0.038
0.15e
0.068
emission factors
95% Confidence
interval, % of mean
±80
±90
_b
±160
'-^ ±60
_C
~C
5 >
±30
±120
±75
Fugitive emissions are included in the text.
No information available; although ammonia is emitted from these unit
operations, it is reported as a total plant emission.
"Emission factor represents only 1 sample.
A fluoride emission guideline of 0.03 g/kg
promulgated by EPA (47).
x
"Based on limited data from only 2 plants.
input has been
where E
R/A'
Ep, and Erpp are emission factors from raw data
for the"reactor/ammoniator-granulator, dryer/cooler, product
sizing and material transfer, and total plant, respectively.
NR/A' ND/C' NP' and NTP are tne corresponding number of samples
82
-------
used to generate each emission factor. This calculation results
in the following total plant stack emission factors:
Particulates: 1.5 g/kg P205 * 69%3
Fluoride (as F) : 0.038 g/kg P205 ± 30%
Ammonia: 0.068 g/kg P205 ± 75%
Information on fluoride emissions from the gypsum pond is re-
ported in the section on WPPA manufacture. One-half of the 48
ammonium phosphate plants are located at fertilizer complexes
producing WPPA. No measurements are available for fugitive
.fluoride emissions from ponds located at plants producing only
ammonium phosphates. However, pond systems at ammonium phosphate
plants not located at fertilizer complexes are proportionately
smaller and would have lower fluoride emissions than those at
complexes.
F. POTENTIAL ENVIRONMENTAL EFFECTS
The source assessment program employs certain criteria to help
evaluate the relative impacts of the source types studied. These
parameters are source severity, affected population, state and
national emission burdens, and growth factor. In evaluating
potential environmental effects, average parameters have been
employed (e.g., emission factors, stack heights, population
densities) . A more detailed plant-by-plant evaluation was be-
yond the scope of the project and conclusions are not drawn with
regards to actual environmental impacts at specific plant sites.
1 . Source Severity
Source severity compares the time-averaged maximum ground level
concentration of an emitted pollutant, x" , to an estimated
hazard factor, F (Equation 24) . max
S - (24)
The hazard factor, F, is defined as the primary ambient air
quality standards presently exist for particulates, sulfur oxides
(SOX) , nitrogen oxidants (NOX) , carbon monoxide (CO) , hydrocar-
bons, c and oxidants. For noncriteria emission species (fluoride
and ammonia) , F is derived from the threshold limit value (TLV©)
Estimated uncertainty based on process-related emissions.
Estimated uncertainty based on total plant emissions.
£
The value of 160 pg/m3 used for the primary ambient air quality
standard for hydrocarbons in this report is a recommended guide
line for meeting the primary ambient air quality standard for
photochemical oxidants.
83
-------
for the chemical substance (68) as TLV (8/24)(1/100). The factor
8/24 corrects for 24-hr exposure and 1/100 is a safety factor.
In the calculation of source severity a conservative safety fac-
tor is used due to the lack of definitive health effects data.
The time-averaged maximum downwind ground level concentration of
each emission species is given by (69):
v = V I
Amax Amaxl+-
0.17
where
X
- 2 Q
max
iieuh2
(25)
(26)
and
vmax
t
Q
IT
e
u
h
short-term (i.e., 3 min) maximum ground level
concentration, g/m3
instantaneous averaging time, 3 min
averaging time, 1,440 min
emission rate, g/s
3.14
2.72
average wind speed, m/s
stack height, m
For criteria pollutants, the averaging time, t, is the same as
that for the corresponding ambient air quality standard. For
noncriteria emission species,_t is 1,440 min (24 hr). A wind
speed of 4.5 m/s is used for u.
The equation for Xmax (Equation 26) is derived from the general
plume dispersion equation for an elevated source (69). For fugi-
tive emissions occurring at ground level (i.e., from materials
handling operations or from the gypsum pond), a special form of
the Gaussian plume dispersion equation is developed, taking the
following form (69, 70) :
(68) TLVs® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1976. American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, 1976. 94 pp.
(69) Turner, D. B. Workbook of Atmospheric Dispersion Estimates,
Public Health Service Publication No. 999-AP-26, U.S. De-
partment of Health, Education, and Welfare, Cincinnati,
Ohio, 1969. 62 pp.
(70) Reznik, R. B. Source Assessment: Flat Glass Manufacturing
Plants. EPA-600/2-76-032b, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1976.
147 pp.
84
-------
X = 2— (27)
A TTO a u
y z
where Y = ground level downwind pollutant concentration, g/m3
a = 0.2089 x0-9031
a7. = 0.113 x°-911
Q = emission rate, g/s
TT = 3.14
x = average wind speed, m/s
x = radial distance downwind from the source, m
Values of x are then calculated to determine at what distance
downwind from the source the severity falls below 0.05 and 1.0
for an average emission factor.
The 24-hr ambient air quality standards of 260 ug/m3 for particu-
lates and 365 yg/m3 for SOX were used as hazard factors to
calculate source severities. For fluoride emissions, a TLV of
2.0 mg/m3 (based on hydrogen fluoride) was used to- calculate F
for use in source severity calculations. The corresponding TLV
for ammonia is 18 mg/m3 (68).
The'source severity calculation does not consider the distance at
which maximum ground level concentrations of an emitted pollutant
occurs. In some cases, depending on individual plant layouts,
the point of maximum severity may occur within plant boundaries.
As mentioned earlier this parameter is used as a basis for com-
paring a large number of emission sources, and a detailed plant-
by-plant analysis was not conducted.
a. Phosphoric Acid and Superphosphoric Acid Plants—
Values for Ymax anc^ ^ were calculated for each emission point at
an average plant. These values are presented in Table 26.
Source severities were also calculated for each plant based on
average emission factors and stack heights. Plant production
rates used in severity calculations were derived for phosphoric
acid and Superphosphoric acid plants by multiplying plant
capacity data in Appendix A by utilization factors of 0.70 and
0.49, respectively, obtained by dividing 1975 annual productions
by available industry capacities. The resulting severity dis-
tributions are presented in Figures 34, 35, and 36 for particu-
late emissions from rock handling operations at WPPA plants, for
particulate and fluoride emissions from the wet scrubber at WPPA
plants, and for fluoride emissions from Superphosphoric acid
plants, respectively. Each severity distribution is plotted as
cumulative percent of the number of plants versus severity for
each emissions species from each emission point.
Source severity distributions were not calculated for SOX emis-
sions from the wet scrubber at WPPA plants or for particulate
emissions from Superphosphoric acid plants because of the smaller
amount of emissions data.
85
-------
TABLE 26. VALUES FOR Xmax AND SOURCE SEVERITIES FOR
EMISSIONS FROM AN AVERAGE WET PROCESS PHOS-
PHORIC ACID AND SUPERPHOSPHORIC ACID PLANT
*roax' "*"»"
Emission point
Wet process phosphoric acid:
Rock unloading
nock transfer and conveying
Wet scrubber system
Gypsum pond
Superphosphoric acid:
Wet scrubber
Total
fluoride
oa
0
1.2
_b
0.55
Particulate
106
10.4
6.5
0
2.5
SOx
0
0
3.9
0
0
Source severity
T6tal
fluoride
Oa
0
0.18
_b
0.09
Particulate
0.
0.
0.
0
0.
41
040
025
01
SO
0
0
0.
0
0
Oil
Zero indicates this species is not emitted from this source.
Hot applicable.
100
90
i 80
u. 70
o
£ 50
|.
I 30
20
10
ROCK UNLOAD ING
w
100
90
' 80
£ 70
o
| 60
o
5 50
UJ
>= 40
20
10
ROCK TRANSFER AND
CHARGING
I- i I i i i I i_
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
SOURCE SEVERITY
°0 0.02 0.04 0.06 0.080.10 0.12 0.14 0.16
SOURCE SEVERITY
Figure 34. Source severity distribution of
particulate emissions from rock
handling operations at WPPA plants.
86
-------
100
90
in
£ 80
S
t 70
o
S 60
o
£ 50
£
5 ^
1 30
PARTICULATE
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.080.09
SOURCE SEVERITY
100
90
80
^O
60
50
P 40
I 30
20
10
FLUORIDE
0.10 0.20 0.30 0.40
SOURCE SEVERITY
0.50
0.60
Figure 35.
Source severity distribution of
particulate and fluoride emissions
from the wet scrubber at WPPA plants
87
-------
100
w
! »
1 70
AUORIDC
50
«
20
10
Figure 36.
"o 0.05 a 10 a is 0.20 0.25
SOURCE SEVERITY
Source severity distribution of fluoride
emissions from superphosphoric acid plants,
Because no stack height is associated with fluoride emissions
from gypsum ponds, source severity had to be calculated differ-
ently. From Equations 25 and 27 and for 24-hr averaging times,
the value of x~ divided by F yielded the graph shown in Figure 37.
Dashed lines give the change in ~x/F with distance from the center
of a typical gypsum pond for emission rates of 11 and 1,100 kg
F/(km2-day) [0.1 and 10 lb/(acre-day)]. The solid line is for
an average emission factor of 220 kg F/(km2-day). Fluoride emis-
sions from the gypsum pond are treated as a point source located
at the center of the pond and represent a worst case analysis.
Note that the value of x~/F falls below 1.0 at approximately
1300 m from the center of the pond for an average emission rate,
and it falls below 0.05 at approximately 6700 m. A severity
distribution for fluoride emissions from the gypsum pond at
individual WPPA plants is presented in Figure 38, based on an
average emission factor. Table 27 presents severity ranges for
each species and emission point and also shows the percentage
of plants having a source severity exceeding 0.05 and 1.0.
b. Normal Superhphosphate and Triple Superphosphate Plants—
Table 28 presents the values of x" and S for each emission
point and for each emission species from three average super-
phosphate plants. Values are based on the current level of
emission control at these plants.
Average stack heights in Table 28 were developed from stack
heights for individual plants reported in Appendix B. A stack
height of 15 m was determined from plant data for emissions from
the baghouses controlling rock unloading and transfer operations.
Emissions from the NSP curing building at an average plant are
not controlled; they are exhausted from the building by ducts
along one side. The height of the curing building, 12 m, was
therefore used as the stack height for this source.
88
-------
.01
0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000
RADIAL DISTANCE DOWNWIND FROM GYPSUM POND, m
Figure 37. "x/F as a function of radial distance
downwind from gypsum pond.
100
o
75
3
ft 50
o
an
cc
o
25
1
2 °o
2.000
12,000 14,000
4,000 6,000 8,000 10,000
DISTANCE TO STATED SEVERITY, m
Figure 38. Distribution of distance to stated severity for
fluoride emissions from the gypsum pond at WPPA
plants.
89
-------
TABLE 27. RANGE OF SOURCE SEVERITIES AND PERCENTAGE OF WET
PROCESS PHOSPHORIC ACID AND SUPERPHOSPHORIC ACID
PLANTS HAVING SEVERITIES GREATER THAN 0.05 OR 1.0
Emission point
Wet process phosphoric acid:
Rock unloading
Rock transfer and conveying
Wet scrubber system
Superphosphoric acid:
Wet scrubber
Species
Particulate
Particulate '
Particulate
Total fluoride
SOx
Particulate
Total fluoride
Source severity, S
Minimum
0.011
0.001
<0.001
0.005
<0.001
<0.001
0.01
Maximum
1.26
0.12
0.078
0.56
0.01
<0.001
0.32
Percentage
S > 0.05
86
28
19
?!a
a
65
of plants
S > 1.0
14
0
0
°a
_a
0
Distribution was not calculated because of the small amount of emissions data available.
TABLE 28. MAXIMUM GROUND LEVEL CONCENTRATIONS AND
SOURCE SEVERITIES OF CONTROLLED EMISSION
SPECIES FROM AVERAGE SUPERPHOSPHATE PLANTS
Emission source category
NSP plants:
Rock unloading
Rock feeders '
Mixer and den
Curing building
GTSP plants:
Rock unloading
Rock feeders
Reactor, granulator.
screen, cooler, dryer
Curing building
ROP-TSP plants:
Cone mixer, den,
storage building
Rock feeders
Rock unloading
Average —
stack "max' pg/
height, SOX as
m Fluoride S02B>k
15 -d
15
18 1.2
12 50
15
15
44 2.5 39
30 0.81
26 5.3
15
15
'It.3
Particulate
4.9
1.0
3.0
92
16
3.0
1.1
. 4.6
8.1
2.2
11
S
c S0* aS b
Fluoride SOza'b Particulate
0.02
0.004
0.18 0.01
7.2 0.35
0.062
0.012
0.36 0.11 0.0042
0.12 0.018
0.77 0.031
0.009
0.042
For worst case analysis, based on uncontrolled emission factor.
Primary ambient air quality 24-hr standard for participates equals 0.26 mg/m3: for SOx it equals 0.365 rag/m3.
CTLV equals 2.0 mg/m3; F equals 6.7 pg/m3. Blanks indicate emission species not emitted from the source category.
e.
'Uncontrolled emissions.
90
-------
To complement the source severity values based on plants repre-
sentative of the industry, source severity distributions for the
whole industry were calculated for all species emitted from each
emission point. Plant production rates used in severity calcu-
lations were derived by multiplying plant capacity data in
Appendix A by utilization factors of 0.66 and 0.65 for normal
superphosphate and triple superphosphate plants, respectively,
obtained by dividing 1975 productions by available industry
capacities. Where actual stack heights were unknown, the
average stack heights shown in Table 28 were used. A graphic
representation of this result is shown in Figure 39, presented
as the cumulative percent of plants with a source severity less
than a specific value. Those emission points and associated
emission species not illustrated in Figure 39 had source severi-
ties for all plants less than 0.01. Table 29 presents severity
ranges for each species and each emission point and also shows
the percentage of plants having a source severity exceeding 0.05
and 1.0.
Because no source test data were available for SOX emissions
from the dryer at GTSP plants, an emission factor was developed
based on fuel analysis and consumption. Values of xmax an<^ ^
for SOX emissions are based on a worst case analysis assuming no
control, even though some control results when effluent gas
streams are scrubbed by acidic pond water before discharge.
c. Ammonium Phosphate Plants—
Table 30 presents values for x"max and source severity for stack
emissions from an average plant. Although some plants have
multiple emission points, this evaluation sums all stack emission
factors and assumes a single emission point having a stack height
of 24 m. This simplification can be justified by examining the
variation in stack heights from individual emission points in
Table 31 (71) . Variation in stack heights between emission
points is well within one standard deviation of the mean.
In order to illustrate potential environmental impact of air
emissions from the entire industry, source severity distributions
were calculated and are presented in Figures 40 through 42.
Table 32 presents severity ranges for each species and each
emission point and also shows the percentage of plants having a
source severity exceeding 0.05 and 1.0.
2. Total Emissions
Potential environmental effects of the emissions from phosphate
fertilizer plants can also be evaluated by determining the total
(71) National Emissions Data System Point Source Listing.
SCC 3-01-030-01, 3-01-030-02, 3-01-030-99, 1976. 190 pp
91
-------
0.01 0.1 01 U
U 10 U 1J U i.l 21
uuta sivnin
a.
II U 01
b.
c.
e.
d.
a. PARTICIPATE AND aUORIDE EMISSIONS FROM
NSP MIXER AND DEN
b. PARTICULATE AND FLUORIDE EMISSIONS FROM
NSP CURING BUILDING
C. SOX AND FLUORIDE EMISSIONS FROM
GTSP REACTOR
d. PARTICULATE AND FLUORIDE EMISSIONS FROM
GTSP CURING BUILDING
e. PARTICULATE AND FLUORIDE EMISSIONS FROM
ROP - TSP MIXER
Figure 39. Cumulative source severity distributions.
92
-------
TABLE 29. RANGE OF SOURCE SEVERITIES AND PERCENTAGE OF PLANTS
HAVING SEVERITIES GREATER THAN 0.05 OR 1.0
Emission ooint
Source severityPercentage of plants
Species Minimum Maximum S > 0.05 S > 1.0
NSP:
Rock unloading
Rock feeding
Mixer and den
Curing building
ROP-TSP:
Rock unloading
Rock feeding
Cone mixer, den,
curing building
GTSP:
Rock unloading
Rock feeding
Reactor, granulator, dryer,
cooler, screens
Curing building
Particulate
Particulate
Particulate
Fluoride
Particulate
Fluoride
Particulate
Particulate
Particulate
Fluoride
Particulate
Particulate
Particulate
Fluoride
SOX
Particulate
Fluoride
0.0036
0.054
0.0046
0.011
0.0065
0.16
0.063
0.018
0.0038
0.027
0.13
1.93
0.057
0.82
0.093
2.28
1.45
0.41
0.035
0.25
0
0
3
100
2
95
0
0
30
100
0
0
0
100
76
0
85
0
0
0
2
0
0
0
0
0
60
0
0
0
12
0
0
0
NOTE.—Blanks indicate that the source severity for all plants is less than 0.01.
TABLE 30. MAXIMUM GROUND LEVEL CONCENTRATION AND
SEVERITY FOR AN AVERAGE DAP PLANT
Stack emissions from total plant
Species
TLV, mg/m3 finax. ug/m3
Fluoride (as F)
Particulate
Ammonia
2.0
0.263
18
2.9
110
5.2
0.44
0.43
0.09
aPrimary ambient air quality standard.
93
-------
TABLE 31. VARIATION IN .EMISSION SOURCE STACK HEIGHTS (71)
Source
Mean stack Standard
height, m deviation, m
Ammoniation-granulation 25^
Cooler/dryer 23
Combined all stack height data 24
9.4
9.3
9.3
Average of 49 stack heights.
Average of 51 stack heights.
100
a 001
a 01
SEVERITY
Figure 40,
ai
10
Severity distribution for total
plant ammonia emissions.
94
-------
a oi
0.04 0.1
SEVERITY
Figure 41. Severity distribution for total
plant particulate emissions.
100
10.0
SEVERITY
Figure 42
Severity distribution for total
plant fluoride emissions.
95
-------
TABLE 32. SEVERITY DISTRIBUTION SUMMARY
Emission point
Total plant
Species
Fluoride (as F)
Particulate
Ammonia
Minimum
0.04
0.04
0.008
S
Maximum
2.9
2.9
0.59
Percentage
S > 0.05
90
90
52
of plants
S > 0.1
10
10
0
mass of each emission species emitted. A comparison with total
particulate and SOX emissions on a state-by-state and national
basis can be made. Table D-l in Appendix D shows the state emis-
sion burdens for the five criteria pollutants as reported in the
National Emissions Data System (NEDS) (72) . Table D-2 in Appen-
dix D is an updated version of the NEDS data as computed by
Monsanto Research Corporation under contract with EPA (73).
Table D-2 was used for computations shown in Tables 33 through
39, which are presented and discussed later in this report.
a. Phosphoric Acid and Superphosphoric Acid Plants—
Total emissions from WPPA and Superphosphoric acid manufacture
are shown in Table 33. These were calculated by multiplying each
emission factor at an emission point (Table 20) by the 1975 total
annual production for the two chemicals: 6,291,000 metric tons
for WPPA and 506,000 metric tons for Superphosphoric acid.
The masses of emissions for criteria pollutants at WPPA (particu-
lates and SOX) and Superphosphoric acid (particulaes) plants
were calculated on a state-by-state basis for comparison with
each state's total emissions burden. The resulting percentage of
state burden for the industries and the contribution to the
national burden are shown in Tables 34 and 35. The total mass of
fluoride on a state-by-state basis is also included in the tables
for completeness.
(72) 1972 National Emissions Report; National Emissions Data
System (NEDS) of the Aerometric and Emissions Reporting
System (AEROS). EPA-450/2-74-012, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
June 1974. 422 pp.
(73) Eimutis, E. C., and R. P. Quill. State-by-State Listing of
Source Types that Exceed the Third Decision Criterion,
Special Project Report. Contract 68-02-1874, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, July 7, 1975. pp. 1-3.
96
-------
TABLE 33. TOTAL ANNUAL MASS OF EMISSIONS FROM WET-PROCESS
PHOSPHORIC ACID AND SUPERPHOSPHORIC ACID PLANTS
(metric tons per year)
Emission point
Total
fluoride Particulate SOx
Wet process phosphoric acid:
Rock unloading
Rock transport
Wet scrubber system
Gypsum pond
Superphosphoric acid:
Wet scrubber
0
0
62
160 to 16,000
3.7
912
281
342
0
5.7 to 28
0
0
198
0
0
TABLE 34. WPPA INDUSTRY CONTRIBUTIONS TO STATE
AND NATIONAL ATMOSPHERIC EMISSIONS
State
Arkansas
California
Florida
Idaho
Illinois
Iowa
Louisiana
Mississippi
North Carolina
Texas
Utah
Number
of
plants
1
5
13
3
4
1
4
1
1
2
1
Total 1975
state
production.
10^ metric tons
35
140
3,384
350
260
155
1,060
142
470
250
45
Mass of
metric
Total .
fluoride
1.2 to 84
4.9 to 350
122 to 8,400
12 to 840
9 to 660
5.4 to 390
37 to 2,700
4.8 to 345
17 to 1,140
8.4 to 625
1.6 to 114
emissions.
tons/yr
Particulate
8.4
35
855
88
66
40
260
35
118
63
11.5
SOX
1.1
4.5
108
11.2
8.3
5.0
34
4.5
15
8.0
1.4
Percent of
state and
national
emissions
Particulate SOX
0.006 0.0005
0.004 0.0002
0.4 0.006
0.15 0.02
0.006 0.0002
0.02 0.001
0.07 0.015
0.02 0.002
0.02 0.0007
0.01 0.0004
0.02 0.0005
United States
36
6,291
222 to 16,000
1,540
200
0.01
0.0003
Total state and national emissions data used in this calculation are given in Appendix C as obtained
from References 72 and 73. State emission summary data were available only for criteria pollutants,
not for fluoride.
The range of fluoride emissions was based on wet scrubber emission factor (0.010 g/kg PjOo) plus gypsum
pond emission factor range (0.025 to 2.5 g/kg PaOg).
(30) 1972 National Emissions Report. EPA-450/2-74-012, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, June 1974. 422 pp.
(31) Eimutis, E. C., and R. P. Quill. State-by-State Listing of Source Types that Exceed the Third
Decision Criterion, Special Project Report. Contract 68-02-1874, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, July 7, 1975. pp. 1-3.
97
-------
TABLE 35. SUPERPHOSPHORIC ACID INDUSTRY CONTRIBUTIONS
TO STATE AND NATIONAL ATMOSPHERIC EMISSIONS
State
Florida
Idaho
Louisiana
North Carolina
Texas
Utah
United States
Number
of
plants
3
2
1
1
1
1
9
Total 1975
state
production,
10^ metric tons
220
38
7.3
142
13
.20
506
Mass of emissions,
metric
Total
fluoride
1.6
0.28
0.53
1.0
0.10
0.15
3.7
tons/yr
Particulate
12.1
2..1
4.0
7.8
0.70
1.1
28
Percent of
state and
national
particulate
emissions''
0.005
0.004
0.001
0.002
<0.001
0.001
<0.001
Based on upper limit emission factor of 0.055 g/kg
Total state and national emissions data used in this calculation are given in
Appendix C as obtained from References 72 and 73. State emissions summary data
were available only for criteria pollutants, not for fluoride.
b. Normal Superphosphate and Triple Superphosphate Plants--
The annual mass of emissions from all superphosphate plants in
the United States is given in Table 36. A comparison with the
total particulate and SOX emissions in the United States in 1975
is included.
The mass of emissions from superphosphate plants on a state-by-
state basis was also calculated, and resulting values were com-
pared to each state's emissions burden. Tables 37, 38, and 39
show the results of this analysis.
c. Ammonium Phosphate Plants—
Mass emissions for each type of pollutant were found by multiply-
ing average emission factors developed previously in this report
by 1975 total production of 2.767 x 106 metric tons of P?05.
These values are approximately 4,150 metric tons of particulate,
105 metric tons of fluoride, and 190 metric tons of ammonia.
The mass of particulate emissions from ammonium phosphate plants
on a state-by-state and national basis was compared to state and
national emissions of particulates from all sources. State-by-
state particulate emissions were estimated by apportioning
national emissions according to the statewise plant capacity
distribution in Appendix A. Table 40 shows the results of this
comparison.
In 1975 an estimated 4,150 metric tons of particulates were
emitted from ammonium phosphate manufacture, while in 1972 nation-
wide particulate emission loading from all sources was 17,872,000
metric tons (72). Thus, the ammonium phosphate industry
98
-------
TABLE 36. ANNUAL MASS OF EMISSIONS FROM SUPERPHOSPHATE
PLANTS IN THE UNITED STATES
(metric tons per year)
Mass of emissions
Emission source category
Fluoride SO,, as SO2 Particulate
NSP plants:
Rock unloading
Rock feeders
Mixer and den 44
Curing building 830
Total plant 874
GTSP plants:
Rock unloading
Rock feeders
Reactor, granulator, screens, cooler, dryer 110
Curing building 16
Total plant 126
ROP-TSP plants:
1,700
1,700
1,854
81
15
45
90
231
Cone mixer, den, storage
Rock feeders
Rock unloading
Total plant
Total superphosphate industry
60
60
1,060 1,700
96
8
42
146
2,231
NOTE.—Blanks indicate species not emitted from this source category.
TABLE 37. CONTRIBUTION TO STATE PARTICULATE EMISSIONS
BURDENS DUE TO EMISSIONS FROM NSP PLANTS
State
State production,
metric tons/yr
P2Q5
Particulate
emissions,
metric tons/yr
Percent of state
particulate burden
Alabama
Arkansas
Florida
Georgia
Illinois
Indiana
Kentucky
Maryland
Michigan
Mississippi
Missouri
Nebraska
New York
North Carolina
Ohio
Pennsylvania
South Carolina
Tennessee
Texas
Utah
Virginia
Washington
U.S. total
29,600
3,480
44,700
66,800
58,700
8,130
10,400
6,390
6,970
3,480
8,710
6,970
6,390
43,000
6,390
10,500
25,600
16,800
22,600
1,740
45,300
6,390
439,040
124
15
188
280
246
34
44
27
29
15
37
29
27
180
27
44
•107
71
95
7
190
2_7
1,843
0.00006
0.000009
0.00008
0.0001
0.00007
0.00002
0.00002
0.00004
0.00001
0.00001
0.00001
0.00001
0.00001
0.00008
0.000009
0.00001
0.00009
0.00004
0.00001
0.00003
0.0001
0.00001
0.0014
99
-------
TABLE 38. CONTRIBUTION TO STATE PARTICULATE EMISSIONS
BURDENS DUE TO EMISSIONS FROM ROP-TSP PLANTS
State
State production,
metric tons/yr
Particulate
emissions,
metric tons/yr
Percent of state
particulate burden
Florida
Idaho
Missouri
North Carolina
Utah
U.S. total
477,200
17,730
26,240
65,440
10,500
597,110
116
4
6
16
3
145
0.00005
0.000002
0.000002
0.000007
0.000001
0.0001
TABLE 39. CONTRIBUTION TO STATE PARTICULATE AND SOX EMISSIONS
BURDENS DUE TO EMISSIONS FROM GTSP PLANTS
State
production,
Mass of emissions,
metric tons/yr
Percent of
state burden
State
Florida
Idaho
Mississippi
North Carolina
Utah
U.S. total
metric tons/yr
706,200
22,510
73,490
83,290
13,410
898,900
Particulate SOX
181
6
19
21
3
230
1,313
42
137
155
25
1,672
Patticulate
0.00007
0.00002
0.00001
0.00001
0.000001
0.00018
SOX
0.0007
0.0007
0.0005
0.00007
0.00009
0.003
100
-------
TABLE 40. ESTIMATED MASS OF PARTICULATE EMISSIONS
FROM AMMONIUM PHOSPHATE PLANTS
Percent
of national
State production
Alabama
Arizona
Arkansas
California
Florida
Idaho
Illinois
Iowa
Louisiana
Michigan
Minnesota
Mississippi
Missouri
North Carolina
Texas
Utah
Washington
U.S. total
2
<1
1
2
43
5
2
5
24
<1
1
3
2
2
6
1
<1
100
Particulate emissions,
metric tons Contribution
From ammonium From all to total
phosphate plants sources (72) emissions, %
72
9
38
99
1,770
221
96
192
988
21
53
117
71
78
247
55
23
4,150
1,178,642
72,684
137,817
1,006,452
226,460
55,499
1,143,027
216,493
380,551
705,921
266,730
168,355
202,438
481,026
549,408
71,693
161,937
17,872,000
<0. 1
<0.1
-------
max
source (where X equals 0), increases to some maximum value, x
as X increases and then falls back to zero as X approaches
infinity. Therefore, a plot of x~ versus X will have the appear-
ance illustrated in Figure 43.
Figure 43
-x, X2
DISTANCE ROM SOURCE
X"/F as a function of distance
from an elevated source.
For fugitive emissions where the stack height is zero, the value
of x/F is a maximum at the source and decreases with distance
downwind according to Figure 44.
JL
F
"1.0 X0.05
DISTANCE FROM SOURCE
Figure 44.
General distribution of x/F as a function
of distance for a ground level source.
The value for the population density around a representative
plant is determined by averaging county population densities in
which actual plants are located. However, because the population
patterns within a given county may vary significantly, the actual
population density in the immediate vicinity of individual plants
may be lower than this average. Conclusions, therefore, should
not be drawn with regard to actual environmental impacts at
individual plant sites.
Due to uncertainties inherent in sampling and dispersion modelino
methodologies, the number of persons around a representative
plant exposed to a x"/F ratio greater than 0.05 is reported in
addition to x"/F > 1.0. The mathematical derivation of the af-
fected population calculation is presented in Reference 69.
a. Phosphoric and Superphosphoric Acid Plants--
The county population density around average WPPA and superphos-
phoric acid plants is 46.1 persons/km2. The affected population
values for those emission species and sources where the ratio of
x to F exceeds 0.05 and 1.0 are given in Table 41. Affected popu-
lation values for SOX were zero and are not shown in the table.
102
-------
TABLE 41. AFFECTED POPULATION VALUES FOR EMISSIONS FROM WET
PROCESS PHOSPHORIC ACID AND SUPERPHOSPHORIC ACID PLANTS
Affected population, persons
Fluoride Particulates
Emission source x/F>0.05 y/F>1.0 x/F>0.05 x/F>1.0
Wet process phosphoric acid:
Rock unloading 0 0 64 0
Rock transfer and charging '00 2 0
Wet scrubber 159 0 00
Gypsum pond 5,532 0 0 0
Superphosphoric acid:
Wet scrubber 28 0 00
In calculating affected population values for fluoride emissions
from a typical gypsum pond, it was assumed that no one lived
within 2,000 m of the edge of the pond, or 2,600 m of the center
of the pond. The value of x"/F drops below 1.0 at 1,300 m from
the center of_the pond, resulting in no affected population.
The value of x/F drops below 0.05 at 6,700 m from the center of
the pond, resulting in an affected population value of 5,532
persons.
b. Normal Superphosphate and Triple Superphosphate—
Affected population values for emissions from average superphos-
phate plants are shown in Table 42 for those emission points with
at least one pollutant which has source severity greater than or
equal to 0.05. For those emissions with source severity less
than 0.05, there is no population affected by a ground level con-
centration for which x~/F is greater than or equal to 0.05.
TABLE 42. AFFECTED POPULATION VALUES FROM SUPERPHOSPHATE PLANTS
_____ Affected population, persons
Y/F>0.05 x/F>1.0
Emission source Particulate Fluoride SQx Paniculate Fluoride SOx
NSP:
Mixer and den 0 529 0 0
Curing building 519 13,021 0 539
ROP-TSP:
Cone mixer, den, curing building 0 1,178 0 0
Rock unloading -" 50
GTSP:
Reactor, granulator, dryer, cooler, screens
Curing building
Rock unloading
0
0
15
1,356 307
161
0
0
0
0
0
0
NOTE.—Blanks indicate no emission of the species for the source.
103
-------
c. Ammonium Phosphate Plants—
Results of affected population calculations for the average
source are presented in Table 43. The average population density
was 82 persons/km2.
TABLE 43. AFFECTED POPULATION VALUES FROM
AMMONIUM PHOSPHATE PLANTS
Affected population, persons
Fluoride
Emission source
Particulate
X/F>0.05 x/F>1.0 y/F>0.05 iT/F>1.0
Ammonia
X/F>1.0
Total plant stack emission
285
288
41
G. AIR POLLUTION CONTROL TECHNOLOGY
1. Phosphoric Acid and Superphosphoric Acid
Environmental and economic concerns have prompted use of control
devices in most facets of the WPPA and superphosphoric acid
industry, with the exception of volatile emissions from the
gypsum pond. The problem of pollutant abatement in the industry
is generally approached by using add-on devices. Process modifi-
cations are not employed because of the delicate balance of
operating conditions required to produce filterable gypsum
crystals. Process technology has been developed to recover
fluoride and gypsum byproducts, offering a more economically
attractive way for the WPPA industry to reduce wastes.
The following sections discuss various controls and byproduct
recovery processes currently in use to reduce air pollutant
levels.
a. Dust Control in Raw Materials Handling Operations—
Enclosed operation and baghouses are typical methods of control
at ground phosphate rock unloading stations. Satisfactory con-
trol of dust emissions from unloading hopper-bottom railroad cars
or trucks is achieved by baghouses which realize high efficiency
in collection of this size particle (60% to 30% of the rock is
less than 74 pm (24). Efficiencies are reported to be greater
than 99% (74).
Feed hoppers, storage bins, and conveyors are also enclosed to
reduce particulate emissions and moisture contamination of the
rock. When transport of ground rock from storage bin to feed
hopper is accomplished by pneumatic conveyors, a cyclone separa-
tor and baghouse are located at the destination for control of
bulk material and discharged dust.
(74) Seinfeld, J. H. Air Pollution: Physical and Chemical
Fundamentals. McGraw-Hill Book Co., Mew York, New York,
1975. 523 pp.
104
-------
Future rock grinding operations may utilize a wet grinding
circuit rather than the current dry grinding practice. Wet
grinding, because it also means wet rock receipt and storage,
leads to a reduction in particulate emissions as well as energy
savings by eliminating a rock drying step.
b. WPPA Wet Scrubber Systems —
Of the available types of pollution control, wet scrubbers have
been the exclusive choice for treatment of contaminated process
vapors generated in the digester, filter, and evaporator. These
scrubbers combine the ability to absorb gaseous fluorides and
remove particulates by impaction on the liquid droplets. Prob-
lems in scrubber efficiency result from deposition of hydrated
silica within water nozzles or scrubber packing, which affects
liquid-vapor contact.
Crossflow packed scrubbers provided high absorption capabilities
and tend to operate free from plugging when preceded by a spray
section (28) . When gases enter the spray section, hot vapors are
cooled, high concentrations of fluorides and particulates are
reduced, and reaction takes place between the water and silicon
tetrafluoride in the gas.
SSiFij + 2H20 — > 2H2SiF6 + Si02 (30)
The silica (SiO2) precipitates in the form of a hydrated gel
[Si(OH),J.
Si02 + 2H20 — *> Si(OH)u (31)
When fluoride and particulate loading is substantially reduced,
gas passes through the more efficient stage, a cross-flow packed
scrubber, where the remaining hydrogen fluoride and particulates
are removed (28) . The crossflow design, with scrubbing spray
normal to the direction of the gas flow, washes precipitates off
the packing to prevent plugging. The collected deposits are near
the front of the packed bed, which is more heavily irrigated to
reduce solids buildup (75) . Overall efficiencies for a spray-
crossflow packed scrubber have been reported to be greater than
99% (31) , A diagram of this scrubber design is presented in
Figure 45. (31) .
Although venturi scrubbers provide effective contact and gas
absorption, they have a major disadvantage in that a high pres-
sure drop (2.5 kPa to 12.4 kPa) and corresponding high energy
requirement are necessary to meet the given standards for emis-
sions (15) . A venturi may be used instead of a spray tower
upstream from the packed scrubber described in the previous para-
graph, or in conjunction with a cyclonic spray tower.
(75) Environmental Engineers' Handbook, Volume 2, Air Pollution,
B. G. Liptak, ed. Chilton Book Co., Radnor, Pennsylvania,
1974. 1340 pp.
105
-------
.PRIMARY
CAS INLET
Figure 45. Spray-crossflow packed scrubber (31).
Important factors observed in efficiencies of control devices are
composition and temperature of scrubbing water. Gypsum pond
water contains 3,000 ppm to 10,000 ppm fluorine. The partial
pressure of the hydrogen fluoride in the pond water makes effi-
cient recovery of fluorides in the contaminated gas stream
difficult (17, 64). The mass transfer process may even become
inoperative at higher temperatures. To combat this effect, some
industries use fresh water in the last stage of the scrubber to
reduce gaseous fluorides to an acceptable level.
The temperature influence on scrubber outlet concentrations is
depicted in Figure 46 (76).
c. Superphosphoric Acid Wet Scrubber—
As in WPPA plants, superphosphoric acid plants treat exhaust air
with wet scrubbers to remove particulates and gaseous fluorine
compounds. The type of wet scrubber used in this application,
however, is different from the WPPA choice because of a lower gas
flow rate. A water-induced venturi scrubber, shown in Figure 47,
is the typical choice (31).
(76) Specht, R. C., and R. R. Calaceto. Gaseous Fluoride Emis-
sions from Stationary Sources. Chemical Engineering Prog-
ress, 63(5):7884, 1967.
106
-------
10.0
I"
00
O
I—t
H-
8
t
i.o
0.1
1 2 4 6 8 10 20 40 60 80100
INLET CONCENTRATION, x 0.028 mg F/m^ (15°C)
Figure 46. Inlet concentration versus outlet
concentration at scrubber discharge
temperatures for a cyclonic spray tower (76)
Reprinted from Chemical Engineering Progress by courtesy
of the American Insititute of Chemical Engineers.
SPRAY
MULE
SEPARATOR
MX
•ATER
OUTLET
Figure 47. Water-induced venturi scrubber (31)
107
-------
The gas stream to the scrubber is from a combination of sources:
barometric condenser, hot well vents, and product cooler tank.
The enclosed system is maintained at a slight negative pressure
to induce inward leakage at openings in access ways and equip-
ment, thus eliminating potential fugitive emissions. Scrubbers
installed to handle the exhaust streams are of nominal capacity,
about 4.2 m3/s, regardless of plant size (77). Because of the
low gas flow rate and availability of large amounts of gypsum
pond water, scrubbing requirements for superphosphoric acid
plants can be met with the venturi ejector without use of mech-
anically more complicated packed and conventional venturi
scrubbers (31).
The water-induced venturi does not depend on gas flow for motive
power. The ejector venturi uses a large liquid spray under high
pressure to induce air flow through the throat section, where
intimate gas-liquid contact occurs. This unit is followed by a
gas-liquid separation chamber to prevent entrainment of the con-
taminated liquid droplets in the exhausted gas. Efficient
separation is achieved by a cyclonic section, which also removes
remaining particulates. An alternative is a packed or cyclonic-
packed scrubber in the separator vessel.
Scrubber efficiency is increased with higher liquid-to-gas ratios
and with increasing nozzle pressure. Plant data indicate that
these installations are 99% to 99.8% efficient (31).
2. Normal Superphosphate and Triple Superphosphate
Superphosphate production and storage facilities utilize a
variety of devices including wet scrubbers, cyclones, and bag-
houses to control emissions of particulates, fluorides, and
combustion gases (31, 32).
Particulate emissions from ground rock unloading, storage, and
transfer systems are controlled by baghouse collectors. Cloth
filters have reported efficiencies of over 99.9% for particles
smaller than 75-pm (Appendix B). Collected solids are recycled
to the process.
(77) Frazier, A. W., E. F. Dillard, and J. R. Lehr. Chemical-
Behavior of Fluorine in the Production of Wet Process Phos-
phoric Acid. Presented at the American Chemical Society
Annual Meeting, Chicago, Illinois, August 24-29, 1975.
16 pp.
108
-------
Emissions of silicon tetrafluoride, hydrogen fluoride, and parti-
culate from the production area and curing buildings are con-
trolled by scrubbing the offgases with recycled water. Wet
scrubbing combines the ability to remove particulate by impaction
on the surface of liquid droplets with the ability to absorb
gaseous fluoride compounds into the liquid phase. Exhausts from
the dryer, cooler, screens, mills, and curing building, where
heavier loadings of particulate may be present, are sent first to
a cyclone separator and then to a wet scrubber.
Gaseous silicon tetrafluoride in the presence of moisture reacts
as follows:
3SiF4 + 2H20 —» Si02 + 2H2SiF6 (32)
The silica is present as a gelatinous mass of polymeric silica
which has the tendency to plug scrubber packings. The use of
conventional packed countercurrent scrubbers and other contacting
devices with small gas passages for controlling silica is there-
fore limited. Scrubber types that can be used within this
restriction are 1) spray tower, 2) cyclonic scrubbers, 3) venturi
scrubbers, 4) impingement type scrubbers, 5) jet ejector
scrubbers, and 6) spray-crossflow packed scrubbers.
Spray towers are not capable of the high efficiencies (greater
than 95%) required for compliance with present regulations. They
find use, however, as precontactors for fluorine removal at
relatively high concentration levels (greater than 3,000 ppm) .
Air pollution control techniques vary from plant to plant depend-
ing on particular plant designs. The effectiveness of abatement
systems for the removal of fluoride and particulate varies from
plant to plant depending on a number of factors. The effective-
ness of fluorine abatement is determined by 1) inlet fluorine
concentration, 2) outlet or saturated gas temperature, 3) com-
position and temperature of the scrubbing liquid, 4) scrubber
type and transfer units, and 5) effectiveness of entrainment
separation (16, 31). Control effectiveness is enhanced by
increasing the number of scrubbing stages in series and by using
fresh water scrub in the final stage. Reported efficiencies for
fluoride control range from less than 90% to over 99% depending
on inlet fluoride concentrations and the system employed. An
efficiency of 98% for particulate control is achievable (31).
3. Ammonium Phosphate
Emission control technology applied to DAP production serves
three purposes: recovery of ammonia, recovery of particulate
MAP and DAP, and prevention of pollutant emissions of ammonia,
fluorides, and particulates. Common practice in the industry is
to combine emission points for emission control: reactor and
ammoniator-granulator, dryer and cooler, and product sizing and
109
-------
material transfer. Reactor and ammoniator-granulator emissions
are vented directly to a wet scrubber system, while emissions
from remaining sources pass through cyclone collectors for prod-
uct recovery and recycle before passing to a wet scrubber system.
The chemistry for ammonia recovery is identical to the process
chemistry discussed earlier: Ammonia is scrubbed from off gases
with excess phosphoric acid where it reacts to form ammonium
phosphates which are retained in the scrubbing liquor. Silicon
tetraf luoride, the primary gaseous fluoride emission species, is
scrubbed from offgases according to reactions in Equations 33 and
34.
2HF + SiF^^HaSiFe (33)
+ 4H20 <=± Si(OHK + 2H2SiF6 (34)
All ammoniation-granulation plants have some form of pollution
control equipment, but a complete characterization of emission
control practices of the industry is not available (5) . Combined
requirements for particulate collection and gas absorption for
ammonia recovery and fluoride emission control permit application
of a wide variety of scrubber types for DAP service. Devices
applied to DAP emission control include
• Spray towers
• Venturi scrubbers
• Impingement scrubbers
• Spray-crossf low packed bed scrubbers
Spray towers provide the interphase contacting necessary for gas
absorption by dispersing scrubbing liquid in the gas phase as a
fine spray. Several types of spray towers are in general use.
The simplest consists of an empty tower equipped with liquid
spray nozzles at the top and a gas inlet at the bottom. Scrub-
bing liquor sprayed into the gas stream falls by gravity through
the upward flowing contaminated gas. A disadvantage of this
device is entrainment of scrubbing liquid aerosols into the exit
gas stream.
Cyclonic spray towers eliminate excessive droplet entrainment by
using centrifugal force to remove droplets. Figure 48 presents
schematic diagrams of one- and two-stage cyclonic spray tower
scrubbers. Gas enters the scrubber tangentially and scrubber
liquor is directed parallel to gas flow, providing crossflow
contacting of gas and liquid streams (11, 43).
Venturi scrubbers (Figure 49) are particularly well suited for
streams with high solids or silicon tetrafluoride loadings
because of their high solids handling capacity and self-cleaning
characteristics. A venturi provides a high degree of gas-liquid
mixing, but relatively short contact time and cocurrent flow
110
-------
CLEAN GAS OUT
CORE BUSTER DISK
SPRAY MANIFOLD
GAS IN
SCRUBBER SCRUBBER
LIQUOR OUT LIQUOR IN
SCRUBBER
LIQUOR INLET
GAS INLET
GAS OUTLET
Figure 48. Cyclonic spray tower scrubbers (11, 45).
SCRUBBER
LIQUOR INLET
GAS INLET
GAS OUTLET
SCRUBBER
LIQUOR OUTLET
Figure 49. Venturi scrubber (11)
111
-------
limit absorption capabilities. Scrubbing liquor is introduced at
high velocity through a nozzle upstream of the venturi throat,
and water velocity pulls flue gas through the venturi. Entrained
scrubbing liquor requires a mist eliminator. The cyclone in
Figure 49 is used to remove mists. In application to DAP emisr-
sions, venturi scrubbers are often used as the initial component
of a multiple scrubber system (11).
Although impingement scrubbers are primarily particulate collec-
tion devices, they also possess some absorption capability. The
Doyle scrubber pictured in Figure 50 is most commonly used by
the fertilizer industry.
GAS INLET
GAS OUTLET
AIR LOCK RELEASE
CONE
OVERFLOW
WEIR BOX
SPRAY
aiMINATOR
SCRUBBING
LIQUID
LIQUID INLET
Figure 50. Doyle impingement scrubber (11) .
Effluent gases are introduced into the scrubber as shown in
Figure 50. The lower section of the inlet duct is equipped with
an axially located core that causes an increase in gas stream
velocity prior to its impingement on the scrubbing liquor sur-
face. Effluent gases contact the pool of scrubbing liquid at a
high velocity and undergo a reversal in direction. Solids
impinge on the liquid surface and are retained, while absorption
of gaseous fluorides is promoted by interphase mixing generated
by impact. Solids handling capacity is high; however, absorption
capability is very limited (11).
The spray-crossflow packed bed scrubber shown earlier in
Figure 45 consists of two sections—a spray chamber and a packed
bed—separated by a series of irrigated baffles. Both spray and
packed sections are equippped with a gas inlet. Effluent streams
with relatively high fluoride concentrations—particularly those
rich in silicon tetrafluoride—are treated in the spray chamber
112
-------
before entering the packing. This preliminary scrubbing removes
silicon tetrafluoride, thereby minimizing bed plugging. It also
reduces packed stage loading and provides some solids handling
capacity. Gases low in silicon tetrafluoride can be introduced
directly to the packed section.
The spray section consists of a series of countercurrent spray
manifolds with each pair of spray manifolds followed by a system
of irrigated baffles. Irrigated baffles remove precipitated
silica and prevent formation of scale in the spray chamber.
Packed beds of both cocurrent and crossflow design have been
tried; crossflow design has proven to be more dependable. Cross-
flow design operates with the gas stream moving horizontally
through the bed while scrubbing liquid flows vertically through
the packing. Solids tend to deposit near the front of the bed
where they can be washed off by a cleaning spray. The back
portion of the bed is usually operated dry to provide mist
elimination.
Spray-crossflow packed bed scrubbing is effective from a gas
absorption standpoint, but it is less effective for collecting
particulate; hence, it is used as a "tail gas" or secondary
scrubber following a particulate scrubber. Packed scrubbers are
seldom used as primary scrubbers due to their tendency to plug
with gelatinous silicon or DAP (45).
Equipment commonly used for primary scrubbing includes Venturis
and cyclonic spray towers, while cyclonic spray towers, impinge-
ment scrubbers, and spray-crossflow packed bed scrubbers are
used as secondary scrubbers (11, 43, 45). Primary scrubbers
generally use 20% to 30% P20s phosphoric acid as scrubbing
liquor principally to recover ammonia (45). Secondary scrubbers
generally use gypsum pond water principally for fluoride control.
Throughout the industry, however, there are many combinations
and variations. Some plants use reactor-feed concentration
phosphoric acid (40% P20s) in both primary and secondary scrub-
bers, and some use phosphoric acid near the dilute end of the
20% to 30% PaOs range in only a single scrubber (31, 43). Exist-
ing plants are equipped with ammonia recovery scrubbers on the
reactor, ammoniator-granulator, and dryer, and particulate con-
trols on the dryer and cooler. Additional scrubbers for fluoride
removal are common but not typical. Only 15% to 20% of installa-
tions contacted in an EPA survey were equipped with
spray-crossflow packed bed scrubbers or their equivalent for
fluoride removal (11).
Emission control efficiencies for DAP plant control equipment
have been reported as:
113
-------
Ammonia 94% to 99% (11, 45)
Particulates 75% to 99.8% (45, 71)
Fluorides 74% to 94% (11)
Fluoride emissions and the need for controlling them could be
eliminated from DAP production if fluorides were removed from
phosphoric acid raw material. As shown earlier in Table 16,
furnace phosphoric acid has very little (less than 1 ppm) fluo-
rine content, but essentially all ammonium phosphates are
currently produced from WPPA. Furnace acid is not used primarily
because it costs 29% more per metric ton of P205 to produce than
WPPA (78) .
Particulate collection efficiency of dry cyclones increases as
gas flow rate increases. However, increasing exhaust gas flow
rate also increases gas flow rate through the dryer. It has been
reported that additional dust is emitted from the discharge end
of the dryer when gas velocity exceeds 112 m/min (46). One way
to increase gas velocity in the cyclone, but not in the dryer, is
to install an open duct in the exhaust line between the cyclone
and dryer and cooler discharge as shown in Figure 51. Gas
velocity through the dryer and cooler can then be regulated by
means of the damper.
H. BYPRODUCT RECOVERY
Fluorine compounds volatilized during production of phosphate
fertilizer materials are being considered as a valuable resource
for production of fluosilicates, fluorides, and hydrofluoric acid
(63). Fluorine is recovered from gas effluent streams as a weak
solution of fluosilicic acid by the following reaction sequence:
Phosphate rock + acid —> HF (35)
4HF + Si02 —*• SiFij + H20 (36)
3SiF4 + 2H20 —»• 2H2SiF6 + Si02 (37)
Calcium fluoride contained in the rock reacts with acid to form
hydrogen fluoride. This hydrogen fluoride in turn reacts with
silica present in the rock to form silicon tetrafluoride. Sili-
con tetrafluoride vapor dissolves readily in an aqueous scrubbing
solution to form fluosilicic acid. Silica formed during absorp-
tion of silicon tetrafluoride is removed by filtration and the
product is a solution of 17% to 25% fluosilicic acid (63).
Systems recover the acid at concentrations of 25% or less, a
constraint which results from a rapid increase in vapor pressure
(78) Environmental Considerations of Selected Energy Conserving
Manufacturing Process Options, Vol. 13, Phosphorus/Phos-
phoric Acid Industry Report. EPA-600/7-76-034m (PB 264
279), U.S. Environmental Protection Agency, Cincinnati,
Ohio, December 1976. 96 pp.
114
-------
FMESM AIM
IN
TO SCRUBBER
Figure 51. Cyclone gas velocity control (46).
at higher concentrations. The small amount of silica-containing
liquid waste generated is normally consumed as a filler in fertil-
izer production.
A number of plants in the phosphate fertilizer industry are
currently practicing recovery techniques. Approximately 60% of
NSP plants recover fluorine as a weak solution of fluosilicic
acid utilizing two-or three-stage wet scrubbing systems.
Between 10% and 20% of WPPA plants recover fluorine during
evaporation-concentration of the phosphoric acid. Two systems
available for fluosilicic acid recovery are inventions of the
Swenson Evaporation Co. and Swift & Co. (22, 27). The Swenson
system involves condensation of evaporator vapors and flash
evaporation to produce an approximately 15% solution. In the
Swift process, a weak solution of fluosilicic acid scrubs the
fluoride-containing vapors from the evaporator and flows to a
recirculation tank. Fluosilicic acid (about 18% to 20%) is bled
from the tank, and water is added to the recycled solution to
maintain the required concentration of acid for scrubbing.
115
-------
An alternative method of fluorine recovery is removal of fluo-
silicate salts prior to concentration of the approximately 30%
P20s acid. One procedure involves addition of sodium carbonate
to the filtered solution of weak acid and subsequent precipita-
tion of sodium fluosilicate.
Process modifications to recover fluoride byproducts reduce emis-
sions from the WPPA scrubber and gypsum ponds by removing fluo-
ride from process streams. The emission factor developed for the
scrubber system at WPPA plants recovering fluoride byproducts was
one-half the factor for plants not practicing recovery techniques;
116
-------
SECTION 5
WATER EFFLUENTS
Because of the integrated nature of the phosphate fertilizer
industry, considering the wastewater handling practices of the
industry as a whole is necessary. Wastewater arising from differ-
ent manufacturing operations are often combined for treatment at
one location. The integrated character of the industry can be
seen in Table 44. Over 70% of the plants produce only one type
of phosphate fertilizer material, while 30% of all the plants
consist of multiunit operations. However, more than 80% of phos-
phate fertilizer production occurs at multiunit plants.
TABLE 44. DESCRIPTION OF PHOSPHATE FERTILIZER COMPLEXES
IN THE UNITED STATES BY UNIT OPERATIONS
Unit operations
at plant site
WPPA
NSP
DAP
WPPA, SPA
WPPA, NSP
WPPA, TSP
WPPA, DAP
WPPA, TSP,
WPPA, SPA,
WPPA, NSP,
NSP, TSP,
WPPA, SPA,
WPPA, NSP,
Total
DAP
DAP
TSP
DAP
TSP, DAP
TSP, DAP '
Number of
plants
5
61
23
3
1
2
10
6
2
1
1
4
2
121
Percent
of total
4.1
50.4
19.0
2.5
0.8
1.7
8.3
5.0
1.7
0.8
0.8
3.3
1.7
100
WPPA—wet process phosphoric acid.
SPA—superphosphoric acid. .
NSP—normal superphosphate.
TSP—triple superphosphate (includes both
granular and run-of-pile).
DAP—diammonium phosphate (some plants
also make monoammonium phosphate).
The remainder of this section considers wasewater handling prac-
tices, gypsum pond characteristics, effects of line treatment,
and potential environmental effects of those plants that do dis-
charge wastewaters.
117
-------
A. SOURCES OF WASTEWATER
Two basic wastewater source types exist in a phosphate fertilizer
plant—point and nonpoint. Point sources are those which origi-
nate as a definite wastewater stream from a particular process.
Nonpoint sources originate from random leaks or from large areas
within a plant. Point sources for each of the five basic proc-
esses are discussed first, below, followed by a general discus-
sion of nonpoint sources for the entire plant.
1. Point Sources
Point sources of wastewater generated at phosphate fertilizer
plants can be divided into three general classes:
• Contact process water
• Noncontact cooling water
• Steam condensate
Contact process wastewater refers to any water which, during
manufacturing or processing, comes into direct contact with or
results from production or use of any material, intermediate prod-
uct, finished product, byproduct, or waste product.
a. Phosphoric Acid—
Sources of contact process wastewater from WPPA production
include wet scrubber liquor, gypsum slurry water, and barometric
condensers (Figure 52). Recycled gypsum pond water is used in
the wet scrubber system to remove particulates, fluorides, and
phosphates from the gas streams. This reservoir of contaminated
process water also supplies the water requirements for transfer-
ring waste gypsum to a disposal area and for operation of baro-
metric condensers. Acid sludge underflow, generated in acid
clarification, contains substantial amounts of phosphate and is
normally disposed of by blending into a dry fertilizer (usually
TSP); it does not enter the pond system.
Once-through or recirculated noncontact cooling water is used to
control the exothermic reaction when concentrated sulfuric acid
is diluted. Cooling water may be either recirculated gypsum pond
water or a separate nonprocess stream that is recycled or dis-
charged. Significant quantities of steam are used in WPPA produc-
tion. In many plants, the steam is used on a once-through basis.
Uncontaminated steam condensate is discharged to the receiving
waters without treatment. Contaminated steam condensate, such
as that from barometric condensers and vacuum ejectors, is dis-
charged to the gypsum pond.
Wastewater streams at phosphoric acid plants are contaminated to
varying degrees by quantities of phosphoric acid, fluorides,
sulfates, and gypsum.
118
-------
IN
COOLING WATER
0 TO 20 m /metric ton P.O,
—5 TO 6 m / metric ton P.O.
5 TO 6 m / metric ton P.O.
— AIR STREAM
— AQUEOUS STREAM
STEAM
CONTAMINATED WATER FROM GYPSUM POND
STEAM JET
EJECTOR
CONCENTRATED
PHOSPHORIC ACID
8 TO 16 x 10 m / metric ton
metric ton P.O,
Figure 52. WPPA production (17).
119
-------
b. Superphosphoric Acid—
Superphosphoric acid plants are located at fertilizer complexes
producing WPPA. As a result, water usage requirements are
supplied for the most part by the existing water recycle system.
Process wastewater streams at Superphosphoric acid plants come
from the barometric condensers, steam jet ejectors, and wet scrub-
bers. These streams pick up quantities of phosphoric acid and
fluorides and are returned to the gypsum pond for reuse at the
phosphate fertilizer complex. Noncontaminated steam condensate
may be segregated into a separate nonprocess water system and
recycled or discharged.
c. Normal Superphosphate
The only process wastewater stream generated at NSP plants is the
wet scrubber liquor used to reduce the level of fluoride gases
and particulate matter evolved from the mixer, den, and conveyors
(Figure 53). Scrubber liquor is discharged to a water contain-
ment or pond system and reused. Nearly two-thirds of the NSP
plants presently practice fluorine recovery, thereby eliminating
or greatly reducing the need for a pond. In this system, fluo-
rine in the exhaust gas stream is recovered as a weak solution of
fluosilicic acid. NSP plants recoverying fluosilicic acid con-
sume the small amount of silica-containing liquid waste generated
as a filler in fertilizer production and report no discharge of
wastewater.
CLARIFIED OR
CONTAMINATED WATER
0.9 TO 1. OmJ
metric ton P-05
SULFURIC ACID
PHOSPHATE ROCK
SCRUBBER
TO ATMOSPHERE
I
V
DEN
•"-
STACK
0.9 TO 1.0 m3
metric ton P.O,
PRODUCT TO CURING
Figure 53. NSP production (17).
d. Triple Superphosphate
The wet scrubber liquor is the only process wastewater stream
generated at TSP production units (Figures 54 and 55). Recycled
gypsum pond water is used in the scrubber system to reduce the
level of fluoride gases and particulate matter evolved during
fertilizer production and storage.
e. Ammonium Phosphate—
At ammonium phosphate plants, substantial quantities of ammonia
are volatilized from the acid neutralizer, ammoniator-granulator,
and dryer. Process economics require that ammonia be recovered.
120
-------
CLARIFIED OR CONTAMINATED WATER
0.7T00.8nnmetrictonP205
TO
ATMOSPHERE
CONTAMINATED
WATER
2 TO 4 x 10"2 m3/metric ton
GTS POUT
Figure 54. ROP-TSP production (17)
CLARIFIED OR
CONTAMINATED WATER
0.9 TO 1.0m3
metric ton P.O.
PHOSPHORIC ACID
PHOSPHATE ROCK
SCRUBBER
TO ATMOSPHERE
MIXER
DEN
• -
STACK
___
CONTAMINATED W
n o in i n n
ROP-TSP TO CUR ING
metric ton P-O,
Figure 55. GTSP production (17).
Weak (28% P20s) phosphoric acid is used as the scrubbing liquor
and is recycled back to the ammoniator-granulator (Figure 56)-
Phosphoric acid scrub solution is consumed in the process and
therefore results in no effluent. However, the phosphoric acid
scrub solution contains a small percentage of fluoride (1% to 3%),
and optimum scrubber operation for ammonia recovery results in
stripping of some of the fluoride from the acid. Secondary wet
scrubber systems are occasionally used to further remove fluo-
rides, particulates, ammonia, and combustion products from the
neutralizer, granulator, dryer, cooler, and screening operations.
This secondary scrubber system uses water as a scrubber liquor
and is therefore a wastewater source. Scrubber effluents are con-
tained in a water recycle system.
121
-------
PHOSPHORIC ACID-
AMMONIA
SCRUBBER
WATER
VAPOR
WATER
500 to 3.000 gal/ton of product
—GAS DISCHARGED
SCRUBBER
•WASTEWATER
PRENEUTRALIZER
EXHAUST GAS
OTHER MATERIALS'
(OPTIONAL)
AMMONIATOR
GRANULATOR
-DAP PRODUCT
RECYCLE FINES
Figure 56. DAP production (4).
2. Nonpoint Sources
In phosphate fertilizer plants, various nonpoint sources can con-
tribute to wastewater handling requirements.
a. Leaks and Spills--
In any plant/ a certain number of valve and pump leaks as well as
random spills can be expected. These leaks and spills are col-
lected as part of the housekeeping procedure and, where possible,
reintroduced directly to the process or contained in the contami-
nated water system. Spillage and leaks therefore do not normally
represent a direct contamination of plant effluent streams that
flow directly to natural drainage.
b. Runoff—
Rainfall runoff from a plant can collect quantities of contami-
nants from the ground and buildings at the production facility.
Drainage from gypsum piles and mined-out areas at a phosphate
fertilizer complex also may be a significant contributor to the
overall water handling requirements of a plant. Runoff and drain-
age are collected and treated before discharge, if necessary, or
sent to the contaminated water system for containment. Non-
contaminated waters are kept segregated where possible and
discharged without treatment.
c. Seepage—
The potential exists for chemical and radiological contamination
of groundwaters as a result of seepage from gypsum stacks and
large process water cooling ponds. Existing data is inconclusive
and is insufficient to determine the possible extent of this
122
-------
contamination. The potential impacts due to seepage need to be
determined on a site specific basis. Seepage can be reduced or
prevented if it is a problem by lining ponds and underlaying
gypsum piles with an impervious material.
3. Gypsum Pond
a. Gypsum Pond Characteristics—
The gypsum pond is an integral part of the wastewater treatment
scheme at a typical phosphate fertilizer complex. The pond
serves as a settling basin for gypsum (a byproduct of WPPA) and
other waste solids, and it functions as a reservoir for recycling
process water and cooling water. The size of the gypsum pond at
a WPPA plant is approximately 2.23 x 10~3 km2/metric ton P205/
day (20)• Gypsum ponds are located adjacent to the plant com-
plex; they are, in many cases, abandoned phosphate rock mine
pits.
Clarified gypsum pond water can be recycled for use in scrubbers
and barometric condensers and for slurrying waste gypsum cake
from the WPPA filtration process. With each recycle, the level
of dissolved contaminants in the water increases. After 3 yr to
5 yr of recycle, impurities in pond waters approach equilibrium
concentrations (20) which are a function of pH, temperature, and
other chemical factors, and are maintained by volatilization and
precipitation of impurities. Typical equilibrium concentrations
are shown in Table 45 (17, 20).
TABLE 45. TYPICAL EQUILIBRIUM COMPOSITION
OF GYPSUM POND WATER (17, 20)
Concentration, Radioactivity,
Contaminant g/m3 pCi/t" t
Phosphorus pentoxide 6,000 to 12,000
Fluoride 3,000 to 10,000
Sulfate 2,000 to 4,000
Calcium 350 to 1,200
Ammonia 0 to 100
Nitrate 0 to 100
Silica 1,600
Aluminum 100 to 500
Iron 70 to 300
"8Ra 60 to 100
8The typical pH range is 1.0 to 1.8.
Picocuries per liter; 1 picocurie equals
0.037 becquerel.
At pH less than 2, it is estimated that 80% of the phosphate pres-
ent exists as phosphoric acid, the remaining 20% being the H2PO<4~
anion (20). The major equilibrium of fluoride componds as depict-
ed in a model developed by Environmental Science and Engineering,
Inc., is shown in Figure 57 (20). Data collected by remote
sensing indicate that fluoride emissions from the gypsum pond
123
-------
ATMOSPHERE
SOLUBLE
Fe AND Al
COMPLEXES
SiO,
H2SiFfl>>
:(AI,Fe) F.
3
CaF2
SEDIMENT
(Na,K>2 SiF&
Figure 57. Major gypsum pond equilibrium (20).
consisted entirely of hydrogen fluoride. The silicon tetra-
fluoride concentration was below the detectable threshold of
0.5 ppb (66). In addition to predominant compounds, fluosilicic
acid (I^SiFg) and hydrogen fluoride (HF) , small amounts of fluo-
ride will be present in the water as soluble and insoluble alumi-
num and iron complexes.
b. Seepage Control from Gypsum Piles—
Natural soil from the surrounding area provides the base for
dikes surrounding gypsum ponds. Gypsum is used to increase the
height of the dike. A drainage ditch surrounds the perimeter of
the area to control contaminated water seepage through earth and
gypsum.
Design of the ditch is dependent on area geology and impoundment
water level. Figures 58 and 59 show examples of dike (64) and
seepage ditch construction. Water effluent collected is pumped
from a low collection point in the ditch back into the pond.
SLOPE NO CREATE!?
THAN 2:1
MINIMUM 6 m
TOP
mi
FREEBOARD.
/MINIMUM 1.5m
/ i
DRAINAGE
DITCH
OUTSIDE TOE
BERM
-8 m MINIMUM
!_|—CORE DITCH,
| MINIMUM DEPTH 1 m
SLOPE NO GREATER THAN 2:1
rinSIDE TOE
JBERM, 8 m MINIMUM
BORROW PIT
Figure 58. Recommended minimum cross section of dam (64).
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 506, by courtesy of Marcel Dekker, Inc.
124
-------
SEEPAGE DITCH
RETURN TO GYPSUM r
POND BY PUMP
OUTSIDE OF PLANT
—APPROXIMATELY
3 m WIDE BY
ABOUT 1 m DEEP
SEEPAGE
DITCH
SURFACE DRAINAGE
DITCH EXTERNAL TO
THE PLANT
Figure 59. Gypsum pond water seepage control
(17) .
c. Lime Treatment of Gypsum Pond Effluents—
Double or triple lime treatment of gypsum pond effluents is the
only wastewater control technology used by the phosphate fertil-
izer industry, and it is practiced at only those plants that
still discharge effluents. A schematic diagram of a two-stage
lime treatment plant is shown in Figure 60.
4. LOW PRESSURE STEAM
HOT WATER
TANK
SUMP
i—] n
MILK OF
^-J I IMF I
MILK OF
~~Q STORAGE
POND WATER
O TO GYPSUM POND
pHC - pH CONTROLLER
LC - LEVEL CONTROLLER
TO RIVER
OR
PROCESS UNITS
Figure 60. Two-stage lime treatment plant (17) .
At least two stages of liming are required; the first treatment
raises pH from less than 2 to about pH 3.5 to about pH 4.0 (20).
125
-------
As pH increases, availability of fluoride ions increases, as
illustrated in Figure 61 (20). Calcium fluoride (CaF2) precipi-
tates according to the following reaction (20):
Ca
++
2F~
CaF2
(38)
o
5
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
pH
Figure 61. Species predominance diagram for 0.4 M
hydrogen fluoride solution (20).
Another reaction also occurs, resulting in deposition of silica
and calcium fluoride (20):
H2SiF6 + 3CaO + H20 —* 3CaF2 + 2H20 + Si02
(39)
The second stage of lime treatment raises pH to greater than 6.0,
with calcium phosphates precipitating via the following reactions
(20) :
2H3P04 + CaO + H20 —* Ca(H2POi4)2 + 2H20
Ca(H2P04)2 + CaO + H20 —* 2CaHP04 4- + 2H20
Additional calcium fluoride will also precipitate.
(40)
(41)
Results of neutralizing a sample of gypsum pond water to a pH of
5.1 are given in Table 46 (28).
126
-------
TABLE 46. REACTION OF GYPSUM POND WATER WITH LIME (28)
Calcium carbonate added
kg/ro3
Oc
6.0
9.0
12.0
13.2
15.0
18.0
Percent of
theoretical
0
50
75
100
110
125
150
pH of b
filtrate
1.8
3.2
3.4
4.8
5.1
5.1
5.1
Chemical composition
of filtrate, g/m3
Phosphorus
pentoxide
2,000
1,650
1,410
590
580
580
580
Calcium
oxide
1,400
1,200
1,100
1,100
1,100
1,100
1,100
Sulfate
2,760
2,500
2,300
2,600
2,700
2,600
2,600
Fluoride
2,900
1,000
70
20
20
30
30
Calcium carbonate required to react with fluorine and phosphate.
Measured with Beckman glass electrode pH meter, Model H-2.
C0riginal gypsum pond water.
Laboratory data for phosphorus and fluoride removal at pH values
over 5 are presented in Table 47 (17).
TABLE 47. LABORATORY DATA FOR PHOSPHORUS AND
FLUORIDE REMOVAL AT HIGHER pH (17)
pH
5.5
6.0
6.5
7.0
7.5
8.0
8.5
9.0
9.5
10.0
Phosphorus,
Laboratory
500
330
200
120
20
3
1.2
g/m3
Plant
42
24
18
14
12
8
6
3
1.2
Fluoride,
Laboratory
13
8.5
6.8
5.8
5.2
4.8
4.6
g/m 3
Plant
17
14
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
NOTE.—Blanks indicate data not available.
226Ra is also precipitated by lime treatment with increasing pH
as shown in Table 48.
TABLE 48. REMOVAL OF 226Ra BY LIME TREATMENT (17)
pH
pCi/S.
8.
2.0
1.5
4.0
0 to 8.5
91
65
7.6
0.04
"Double lime" treatment does not reduce nitrogen levels, although
at high pH (greater than 9.0) significant ammonia loss to ambient
air can occur (17). To date there is no proven means of
127
-------
economically removing ammonia nitrogen from aqueous solutions
having low concentrations in the range of 20 to 60 g/m3. The
best control method is keeping the ammonia contaminant level low
by preventing its entry into the main contaminated water system.
This is accomplished to a great extent by scrubbing emissions
from the ammonium phosphate production unit with a weak solution
of phosphoric acid that is subsequently consumed in the process.
The main disadvantage of the liming operation for continual use
is the high cost involved. Because the buffering capacity of the
gypsum pond water is high at pH 1.0 to pH 3.0, large amounts of
lime are required to raise the pH initially to 3.0 relative to
the amount required to raise the pH from 3.0 to 6.0 (20). An
additional disadvantage is the deposition of calcium fluoride on
the lime particles, rendering them chemically inactive. The use
of high intensity agitators is required to prevent this from
happening.
An investigation was conducted specifically to evaluate the reduc-
tion in radionuclide levels in wastewaters by various lime treat-
ment processes (14). In the initial treatment laboratory tests,
process pond water was obtained from a Florida wet process facil-
ity, and four bases (quick lime, limestone, hydrated lime, and
dolomite) were added to 4 £ of process water in different amounts
to .increase the pH. After vigorous agitation, the solutions were
allowed to settle, and the resultant supernatant liquids were
filtered and analyzed for their soluble 226Ra concentrations.
The results as presented in Table 49 show that in all treatment
cases the soluble 226Ra concentration was reduced by more than
99.7%, even though the final pH ranged from 4.0 to 8.0. This
large reduction is attributed to the amount of readily available
sulfate ions in the process water enabling large-scale coprecipi-
tation of calcium-radium sulfate.
TABLE 49. LABORATORY PROCESS WATER TREATMENT STUDY (14)
Treatment
Untreated process water
Calcium oxide (quick lime)
Limestone rock
Slaked lime (hydrated lime)
Dolomite
Amount
of base
added , g
0
70
500
50
500
Resultant
PH
2.0
7.9
4.6
8.0
4.0
Dissolved
226Ra,
pCi/i
75. 8a
0.15
0.11
0.07
0.16
a6.7 pCi/fc undissolved.
Subsequently, field studies were conducted at several WPPA facil-
ities to verify the effectiveness of lime treatment as observed
in the laboratory (14). Results at four plants are presented in
Table 50.
128
-------
TABLE 50. EFFECT OF LIME TREATMENT ON RADIOACTIVITY
REMOVAL FROM EFFLUENTS FROM A WPPA PLANT (14)
Sample
PH
Total
"6Ra,
pCi/t
Plant A—Field
Untreated process water
Outfall (after double liming)
2.0
9.1
82.3 1
4.54
Plant A — Field
Untreated process water
Limed once
Prior to second liming
Outfall (after second liming)
' 1.8
4.4
4.3
7.1
55.6
1.20
1.5
1.8
Plant B— Field
Untreated process water
After first liming0
Prior to second liming
Outfall (after double liming)
Untreated nonprocess water
Nonprocess water after liming
Nonprocess water outfall
2
4.5
6
8b
"b
~b
86.2 1
74.0
0.90
0.45
1.38
2.6
0.88
Plant C~Field
Process water
Outfall (after single liming)
1.9
6.6
55.2
2.55
Total uranium.
zsmj iibbfj
Survey Number 1
,086 48 ,
1.09 NDa
Survey Number 2
411 24
b b
39.7 5.2
16.8 0.98
Survey Number 1
,769 98.8
736 33.4
67.8 3.17
0.26 ND
0.28 ND
0.96 ND
0.34 ND
Survey Number 1
676 35.1
0.26 ND
pCi/4
2 3B()
1,045
0.52
394
b
39.5
16.8
1,825
734
68.1
0.33
0.39
0.75
0.42
661
0.28
Total
ZZBTh
2.5
0.44
3.4
b
ND
0.32
3.92
6.15
ND
0.1
ND
0.13
ND
0.86
ND
thorium,
zsoTh
70
0.57
101
5^52
0.71
393
4.3
1.32
0.13
ND
0.79
1.32
8.6
ND
pci/e
zazTh
4.5
0.04
3.2
b
ND
0.11
6.33
7.5
ND
ND
ND
0.07
ND
4.1
ND
None detected. Not measured.
CThese concentrations are high because of the large suspended solids load of 23.5 g/i. The
dissolved concentrations in picocuries per liter were 5.2 for 226Ra, 12.8 for 2sltU, 0.52 for 235U,
and 12.9 for 238U.
Field survey number 1 at Plant A was conducted very early in the
rainy season prior to the initiation of large-scale effluent
treatment. Field survey number 2 was performed late in the rainy
season after almost continuous lime treatment for over 2 mo. A
comparison of process water from survey number 1 to survey number
2 shows a 32% decrease in 226Ra concentration during the second
survey. This is probably due to the combination of dilution of
the process water by the influx of surface rain runoff and the
removal of the radioactive material by treatment and discharge of
approximately 10,000 cubic meters of water per day.
Results for every plant show that treatment with lime is highly
efficient (greater than 94%) in removing 226Ra from the dis-
charged process water, in good agreement with removal efficien-
cies observed in the laboratory experiments. Lime treatment also
proved to be extremely effective in removing uranium and 230Th
from treated process water, with removal efficiencies of at least
96% and 99%, respectively, in the four cases noted.
Therefore, although primarily designed for pH, phosphorus and
fluoride control, not for removal of radionuclides in the efflu-
ent, treatment with lime was observed to be highly effective in
removing 226Ra, uranium, and thorium from the effluent discharge.
These results are attributed to the following factors (14):
129
-------
• Process water contains a large concentration of sulfate
and phosphate ions to enable ready compound formation.
• Neutralization by an agent such as lime not only allows
for the reduction of solubility of several compounds but
provides an ample supply of calcium ions to enable the
large-scale formation of calcium sulfate.
• The relative insolubility of radium sulfate makes it
readily coprecipitate with calcium sulfate.
• Uranium and thorium probably precipitate along with
calcium sulfate and other components through substitution
for calcium in formed compounds.
• Settling provides the opportunity for the precipitated
compounds to be removed from the effluent and not be
discharged as suspended solids.
B. POTENTIAL ENVIRONMENTAL EFFECTS
1. Wastewater Disposal and Treatment Practices
Information about the extent of wastewater disposal and/or con-
tainment practices utilized by the phosphate fertilizer industry
was obtained through industrial contacts. A summary of the waste-
water handling practices is presented in Table 51. Contacts with
over 70% of the plants in the industry revealed that nearly 75%
have no discharge of process wastewater. Of the 15 plants that
reported a discharge, 12 reported a discharge of treated process
water only when necessitated by excessive rainfall. Several of
these had not treated or discharged water for several years. In
actual practice, discharge of contaminated process water from the
recycle pond system is held to an absolute minimum due to the
treatment cost involved.
Wastewater discharge practices have been restricted due to
recently promulgated EPA regulations. Beginning July 1, 1977,
and effective when each plant's wastewater discharge permits are
subject to renewal, discharge of process wastewater pollutants to
navigable waters is allowed only under the following conditions
(79) :
• Process wastewater impoundment facilities must be con-
structed to contain precipitation from the 10-yr, 24-hr
rainfall event as established by the U.S. National
Weather Service.
(79) 40 CFR 418, Fertilizer Manufacturing Point Source Category,
Subpart A—Phosphate Subcategory. Federal Register, 41(98)
20582-20585, 1976.
130
-------
TABLE 51. WATER EFFLUENT DISPOSAL AND CONTAINMENT PRACTICES
FOR THE PHOSPHATE FERTILIZER INDUSTRY
[percent of plants specified (number of plants)]
Wet process
•phosphoric
acid plants
Process water discharged continuously:
Treated
Untreated
Discharge of treated process water only j
when necessitated by excessive rainfall '
No discharge of process water reported
Insufficient information .
Total
Pond system onsite for water containment
and reuse :
Continuous discharge fron pond system
Discharge only when necessitated by
periods of excessive rainfall
No discharge fron pond system reported
Treat pond system with line to precipitate
fluorides and other contaminants
Uncertain
Ho pond system onsite
Information regarding wastewater
handling system incomplete
Recover fluosilicic acid
Number of plants contacted
Number of plants in industry
Percent of industry surveyed
7
3
38
52
0_
100
90
7
38
45
3
0
0
10
28
29
36
81
<*>a
(i)
*
(ID
(15)
JO)
(29)
(26)
(2»a
(11)
(13)
(1)
(0)
(0)
(3)
(8)
Superphosphoric
acid plants
0
0
44
56
0_
100
89
0
44
44
11
0
0
11
0
9
9
100
(0)
(0)
f
(4)T
(5)
(0)
(9)
(8)
(0)
(4)
(4)
(1)
(0)
(0)
(1)
(0)
Industry
Normal
superphosphate
plants
0
0
6
88
6
100
36
0
6
23
13
6
3
61
51
66
77
-------
• Process wastewater must be treated and discharged when-
ever the water level due to catastrophic precipitation
events equals or exceeds the midpoint of the surge capacity.
• When such a discharge must occur, the pollutant concen-
trations must have 30-day average values of less than
35 g/m3 of total phosphorus and 25 g/m3 of fluoride.
2. Effluent Parameters
Wastewater from the manufacture of phosphate fertilizer materials
originates from many point and nonpoint sources. The quantity
and characteristics of a given plant effluent are dependent on
the types of processes present at a complex, plant-to-plant vari-
ations in process design and operation, equipment age, level of
maintenance, plant drainage and collection system, and wastewater
treatment methods. As a result, it is difficult to define aver-
age effluent parameters that are truly representative of the
industry as a whole. The approach taken in this study is to pre-*
sent available water discharge data for a representative number
of the phosphate fertilizer complexes that report a discharge.
Justification for this approach is as follows:
• Thirteen of the fourteen ammonium phosphate plants that
were found in the study to discharge wastewater are
located at fertilizer complexes producing phosphoric
acid. The one exception uses excess process water to
irrigate pasture land. No other information is available
concerning this plant. Another plant reporting a dis-
charge of treated wastewater and not located at a phos-
phoric acid complex was expected to discontinue ammonium
phosphate production in early 1977 and was not included
in the survey results.
• All superphosphoric acid plants are located at complexes
producing WPPA.
• Fifteen of the sixteen TSP plants are located at fertil-
izer complexes producing WPPA. The one exception was
expected to close during calendar year 1976 or early 1977.
• The two NSP plants that reported a discharge of process
water when necessitated by excessive rainfall are located
at complexes producing phosphoric acid.
Available wastewater discharge data on file as of October 1976 at
the Florida Department of Environmental Regulation in Winter
Haven were collected and are presented in Table 52. Nonprocess
water from a phosphate fertilizer plant may include any of the
following: noncontact cooling water from the phosphoric acid
132
-------
TABLE 52. WASTEWATER DISCHARGE DATA FOR PHOSPHATE FERTILIZER PLANTS
u>
Plant
code
A
B
ce
D6
E
f
G
Design
production
capacity,
metric tona
Pads/day
1,900
1,270
272
535
381
9O7
363
Type of
discharge Reported treatment
Process (pond) Double limed.
Nonprocess :
1
2
b c
Pond Neutralize discharge.
Nonprocess Lime treat if necessary.
Nonprocess Lime treat if necessary.
Process (pond)" Double limed.
Seepage (pond)
Nonprocess
Nonprocess ' Allowed to settle before discharge.
Process (pond) Neutralize and discharge.
pB,
range
(average)
5.6/11.7
(8.8)
6.2/11.7
(9.2)
2.4/9.0
(7.4)
3.5/10.5
/7.1
(6.7)
/7.5
(6.6)
5.6/11.4
3.6/6.2
5.5/10.2
/8.0
(5.4)
/10.54
(3.37)
Flow rate,
mVs
0.41/0.21
0.29/0.61
0.018/0.031
2.0/
0.36/0.91
0.36/
0.18/0.44
0.28/
0.0013/
0.000020/
0.027/
0.025/0.11
0.45/0.69
Yearly
Fluoride,
g/m3
20.1/49.0
4.38/14.0
9.3/175
28.5/225
5.0/56
1.8/
8. 2/12. S
6.3/
39.2/183
3,381/6,500
4.0/42.5
1.66/5.1
8.4/27.0
average/daily maximum
Total
suspended
solids,
g/nr>
21.8/71.0
3.7/18.0
19.3/128
68/267
10/15
6.6/
8.5/14.1
4.3/
172/
20. 8/
W
15.5/58
21.9/215
Phosphorus ,
g/m3
43.1/330.0
0.8/12.1
46.3/780
25/55
15.3/18.9
11.2
21.0/37.2
12. 1/
52. 9/
39/
4.9/
2.56/4.8
15.3/128
Ammonia-
nitrogen, 226Ha,
g/n3 pci/t Date
32.6/82 2.6/6.2 7/75
f
7/75
7/75
7.4/149 Date
S/75
-. - 6/77
4/75
8.B/
3.6/
7/75
6/75
to 6/76
to 6/76
to 12/75
d
to 4/76
to 5/78
to 4/76
1973
to 7/76
to S/76
Blanks indicate information not available. All wastewater enters pond system. New wastewater handling system now used. Early 1970's.
More recent effluent data supplied by plant personnel. Zero discharge of process water from pond system practiced.
'Last discharge of process water occurred in 1975; as of October 1976 discharge of process water only during period of excessive rainfall.
Discharge of pond water only in period of extreme rainfall. No discharge of pond water required in previous 5-yr period.
All water enters pond system. Modifications to existing wastewater handling system completed December 1977. Will discharge from pond
only during period of heavy rainfall after double limining in future.
-------
production unit, cooling tower blowdown from an associated sul-
furic acid plant, rainfall runoff, drainage from mined-out areas,
washdown waters, and spills.
3. Source Severity
For water effluents source severity compares the concentration
of a particular pollutant after discharge and dilution in the
receiving body with an estimated allowable concentration denoted
as the hazard factor. The concepts of hazard factor and severity
are used as a basis for comparison of the relative impacts of a
large number of source types. The hazard factors used in this
evaluation may be changed as better health effects data becomes
available.
In determining the source severity of a plant, the discharge quan-
tity is compared to the receiving body flow rate times the hazard
factor according to Equation 42.
v
V
S =
VR + VD F
(42)
where S = source severity for a particular pollutant
V_ = wastewater effluent flow rate, m3/s
C_. = concentration of particular pollutant, g/m3
V_ = volumetric flow rate of receiving body above plant
discharge, m3/s
F = hazard factor for particular pollutant, g/m3
Hazard factors for individual pollutants are given in Table 53
(80-82).
A value of 1.00 g/m3 was used for the ammonia-nitrogen present in
the wastewater effluent because at a pH of 7 or lower, nearly
100% of the ammonia-nitrogen exists in the ionized form.
(80) Quality Criteria for Water. EPA-440/9-76-023, U.S. Environ-
mental Protection Agency, Washington, D.C., 1976. pp. 16-21,
(81) Manual of Treatment Techniques for Meeting the Interim
Primary Drinking Water Regulations. EPA-600/8-77-005, U.S.
Environmental Protection Agency, Cincinnati, Ohio, May 1977.
73 pp.
(82) Eimutis, E. C., J. L. Delaney, T. J. Hoogheem, S. R. Archer,
J. C. Ochsner, W. R. McCurley, T. W. Hughes, and R. P. Quill
Source Assessment: Prioritization of Stationary Water
Pollution Sources. EPA-600/2-77-107p, U.S. Environmental
Protection Agency, Washington, D.C., December 1977. 119 pp.
134
-------
TABLE 53. HAZARD FACTORS (80-82)
Hazard
Effluent species factor
Fluoride 0.19
Total suspended solids 25
Phosphate-phosphorus 0.10
Ammonia-nitrogen (ionized form) 1.00
Discharge data presented in Table 52 were used to calculate
source severity values. Source severities for individual phos-
phate fertilizer complexes are presented in Table 54. Only one
set of eight measurements for 226Ra contamination in discharged
process waters was available. The low severity determined for
this case along with the information presented in Table 48 for
radium precipitation with increasing pH suggest that the severity
due to this- contaminant will remain extremely low in effluent
streams treated with lime to remove fluorides and phosphates.
Source severities for fluoride, phosphorus, and to a lesser
degree ammonia-nitrogen are in a number of cases greater than
1.0. This is due to the low flow rate (1 m3/s to 6 m3/s) of
the receiving bodies (83). By comparison, the mean flow rate of
the Ohio River at Greenup, Kentucky, is 3,210 m3/s (84).
In addition to the effects from normal wastewater discharge,
there is a potential danger from dike failure around a gypsum
pond. Such failures have occurred in the past and have resulted
in large fish kills when untreated pond waters were discharged
directly into surface streams. Dikes are now constructed to pre-
vent this from happening; thus, there is no way to evaluate the
chances of future dike failures.
(83) Water Resources Data for Florida, Water Year 1975.
Volume 3—West-Central Florida Surface Water, Ground Water,
Quality of Water. USGS-WRD-FL-75-3 (PB 259 493), U.S.
Department of Commerce, Tallahassee, Florida, July 1976.
1249 pp.
(84) Water Resources Data for Kentucky, Water Year 1975. USGS-
WDR-KY-75-1 (PB 251 853), U.S. Department of Commerce,
Louisville, Kentucky, January 1976. 348 pp.
135
-------
u>
en
TABLE 54. SOURCE SEVERITIES FOR WASTEWATER DISCHARGES
AT INDIVIDUAL PHOSPHATE FERTILIZER COMPLEXES
Plant
code
A
B
C
D
B
F
G
Receiving body
North Prong, Alafia River
North Prong, Alafia River
Peace River
Peace River
North Prong, Alafia River
Skinned Sappling Creek
to Alafia River
Thirty Mile Creek to
North Prong, Alafia River
Type of
discharge
Process (pond)
Nonprocess
Combined
-------
SECTION 6
SOLID RESIDUE
A. SOURCES OF SOLID RESIDUE
Solid residue wastes are generated at phosphate fertilizer plants
in the form of sludges and other slurries. These suspensions are
sent to the gypsum pond or other settling basin where solids
settle. The settled mass is either left in the pond, dredged
for use in extending the dike, or recovered as a resource.
There are three sources of solid residue in the phosphate fertil-
izer industry:
• Gypsum from the filtration of wet process phosphoric acid.
• WPPA sludge.
• Wet scrubber liquor.
Gypsum (CaSO^ *2H20) , a byproduct in WPPA manufacture, is formed
by reaction of phosphate rock with aqueous sulfuric acid:
Ca3(P04)2 + SHaSOi, + 6H20 +± 2H3P04 + 3 (CaSOt, »2H2O) (43)
Reactant slurry flows from the acidulator to the filtration unit,
where phosphoric acid is drawn off by vacuum filtration, leaving
gypsum cake on the filter. Cake is washed with weak phosphoric
acid to recover its residual acid and then rinsed from the filter
screens with recycled pond water. Gypsum slurry flows to the
gypsum pond for solids settling. In areas where land stability
or availability prevents the use of ponds, gypsum cake from the
filters is transported by conveyor to gypsum piles.
The quantity of gypsum produced in a WPPA plant ranges from
4.6 to 5.2 metric tons of gypsum/metric ton Pz°5 produced (24,
64). As a rule of thumb, approximately 1,360 m^ of gypsum will
be accumulated yearly per metric ton of P205 produced per day
(24) .
A second source of solid residue is phosphoric acid from which
impurity-bearing minerals settle out in the clarifier to form
acid sludges. Phosphate rock salts which contribute to acid
sludge formation include fluorine, iron, aluminum, silicon,
137
-------
sodium, and potassium salts. Table 55 shows an analysis of
solids collected at various stages of WPPA acid production (22).
TABLE 55. ANALYSIS OF SOLIDS FROM WPPA (22)
Reprinted from Phosphoric Acid, Volume I, A. V. Slack,
editor, p. 694, by courtesy of Marcel Dekker, Inc.
Analysis, weight percent
Phosphorus Alu- Fluo-
Solids from pentoxide Calcium Sulfate minum Iron rine Silica
32% P20s acid (feed
to evaporators) 1.9 14.8 38.9 0.3 0.2 19.9 10.3
54% Pads acid from
evaporators 6.8 12.9 29.0 5.1 0.3 22.0 5.3
54% PaOs acid from
storage 38.9 3.3 4.7 1.5 9.6 12.9 6.1
Fluosilicates, fluorides, silica, cryolite [(Na or K)3A1F6],
sulfates, unreacted phosphate rock, and various other combina-
tions of impurities as complex salts have been identified in
acid sludge (22).
Acid sludge is separated from acid in the clarification process.
Separated solids can be either dried and used as a fertilizer or
sent to the gypsum pond. Effluent from the clarification process
ranges from 0.7 m3 to 3.2 m3/metric ton Pz05 (17).
The third source of solid residue wastes is wet scrubber liquor.
Wet scrubbers are used throughout the phosphate fertilizer
industry to remove particulates and fluorides from exhaust gas
streams. Recycled gypsum pond water is used as scrubbing
solution. After passing through the wet scrubber, solution is
recycled back to the gypsum pond for solids settling.
At ammonium phosphate plants, for example, scrubber liquor going
to the gypsum pond contains about 10 g of solid residue per kilo-
gram of P205. This solid residue (20) is primarily silicon
hydroxide (Si[OHJi»). The solids value is calculated on the basis
of a filtered-to-concentrated phosphoric acid ratio of 1:1,
assuming that all the fluorine from the acid goes to the exhaust
gas stream as silicon tetrafluoride and that 85% of silicon
tetrafluoride is collected in the scrubbing system. These solids
will be deposited in the gypsum pond.
Although solid residue values for wet scrubber systems at other
phosphate fertilizer operations do not exist, they should be
similar to those for ammonium phosphate plants.
138
-------
B. POTENTIAL ENVIRONMENTAL EFFECTS
Approximately 99% of solid residue wastes generated at phosphate
fertilizer plants are stored in ponds, stacked in piles, or
stored in mining pits at the plant site. The remaining 1% is
sold as a raw material for various products.
Rainfall drainage from gypsum piles is collected in a ditch and
recycled to the gypsum pond. Therefore, under normal conditions
there will be no adverse environmental effect due to solid
residues. The only concern due to these wastes is the large
amount of land area required to store gypsum and the unsightly
appearance of 30-m piles of gypsum.
To date, there are no data with which to evaluate potential
effects on groundwater due to leaching from gypsum piles. Since
gypsum wastes contain mainly calcium sulfate and lesser quanti-
ties of phosphates and fluorides, any potential adverse effect
should be minimal.
There are no data available to estimate air emissions from
gypsum piles due to wind erosion. However, this effect is
minimal; layers of clay are applied to the surface of the gypsum
for added strength when the material is used for dikes. Also,
gypsum is listed as a nuisance dust with a corresponding inhala-
tion TLV of 10 mg/m3 of air (70).
C. CONTROL TECHNOLOGY
1. Disposal Practices
Waste gypsum produced in a WPPA plant ranges from 4.6 to 5.2 met-
ric tons gypsum per metric ton of ^2^5 produced (24, 64).
Approximately 1,360 m3 of gypsum will be accumulated yearly per
metric ton of PaOs produced per day so that at least 2,230 m^
of land area per daily metric ton ?205 should be reserved for
gypsum disposal.
In the United States and other locations, three disposal prac-
tices are currently used: 1) gypsum ponds and piles, 2) aban-
doned mine pits, and 3) sea disposal. In the United States, more
than 90% of the plants use gypsum ponds to collect slurry.
Initially, two or more areas are converted to lagoons by means
of low dikes provided with proper outfalls for potential effluent
discharge. As one area becomes filled, the gypsum stream is
diverted to the second area, and the first section is allowed to
dry out sufficiently to support mechanical equipment. The dike
is then increased in height using deposited gypsum as raw mater-
ial, and the procedure is repeated. Existing gypsum piles range
in height from 30 m to 36 m (17, 24).
139
-------
In western states where poor land stability or availability pre-
vents using gypsum ponds, gypsum cake from the vacuum filters is
transported by conveyor to gypsum piles.
The second disposal technique is practiced primarily in Florida.
Instead of constructing gypsum ponds, abandoned phosphate rock
surface mines are used as gypsum ponds and for other solid
residue disposal. The only potential environmental hazard from
this disposal technique is possible leaching of fluorides, phos-
phates, and 226 Ra into groundwater systems. The potential for
such leaching to occur is presently unknown.
A third disposal technique, used by less than 2% of phosphate
fertilizer plants in the United States but more widely used
throughout Europe, is practiced at plants located in coastal
areas. After removal from the vacuum filters, gypsum is slurried
with about a tenfold quantity of seawater or cooling water. It
is then pumped into the ocean, or, in a few cases, discharged
into major rivers (64).
Seawater is a better solvent for gypsum than freshwater. Solu-
bility of gypsum in seawater is about 3,500 g/m3 as compared to
about 2,300 g/m3 in fresh water. The solids content of the
gypsum slurry is below 5%, low enough for quick dispersion and
dissolution in ocean water (64).
2. Resource Recovery
Several approaches have been taken in seeking commercial uses for
waste gypsum and its associated solid residues. In 1975, approx-
imately 30 x 106 metric tons of gypsum waste were generated by
the phosphate fertilizer industry (85). Of this total, about
90 x 103 metric tons were applied to calcium-deficient soil in
the southern states for peanut growing. Gypsum was also used for
improvement of alkali soils in California and for land reclama-
tion in coastal areas.
Because gypsum waste, often referred to as phosphogypsum (86),
contains varying quantitites of phosphoric acid, it also serves
as a light fertilizer.
(85) Personal communication with John Sweeney, U.S. Bureau of
Mines, Tallahassee, Florida, 26 September 1977.
(86) Murakami, K. By-product Recovery, As Raw Material for
Plaster and Cement - Japanese Practice. In: Phosphoric
Acid, Volume I, A. V. Slack, ed. Marcel Dekker, Inc.,
New York, New York, 1968. pp. 519-523.
140
-------
Waste gypsum has been used for wallboard. In the United States,
however, the dihydrate process for phosphoric acid production
produces a gypsum waste high in phosphoric acid which results in
poor quality wallboard. Also, there is some concern about
possible low-level radiation effects from wallboard made of
uranium- and radium-containing gypsum wastes.
In Europe and Japan where the hemihydrate process is more com-
monly used, the resulting gypsum waste is purer, containing less
phosphoric acid and uranium. More of this gypsum waste is used
for wallboard. In England, where only the standard dihydrate
process is used, special purification methods make the byproduct
suitable for wallboard. This purification step is more econom-
ically feasible in England than in the United States because
natural (and purer) gypsum is not as abundant in England as it is
in the United States (27).
Another possible use for gypsum is in cement and other road top-
pings. However, the phosphoric acid and other phosphates retard
setting and lower the strength of the hardened body. Fluorine
compounds reduce setting time and lower the concrete strength,
but these effects are small compared to the effects of phosphate
contamination (86). In Florida, there are further concerns over
public exposure to low-level radiation from road surfaces con-
taining gypsum wastes or from road base material containing
phosphate rock mining overburden.
Gypsum can be reacted with ammonia and carbon dioxide to form
ammonium sulfate and calcium carbonate. This is an old and well-
known practice applied to natural gypsum, but there has been
relatively little application to waste gypsum. Only a few plants
in India, Japan, and Europe use this technology (27).
Another potential resource recovery method is treating waste
gypsum with silica at high temperatures to produce sulfuric acid.
Furthermore, the additional product of calcium silicate could be
used for cement. Although the method is technically feasible,
the high water content of gypsum, the corrosive effect of fluo-
rides, and the adverse effect of P2®5 content on cement quality
are all major drawbacks. Moreover, due to the price and avail-
ability of sulfur in the United States, this technology is not
yet economically feasible (27).
While several potential resource recovery methods are technically
feasible, less than 1% of the gypsum waste in the United States
is utilized because its recovery is not economically feasible
and its disposal does not pose an environmental hazard. The
remaining quantity is stored in piles near the plants.
141
-------
SECTION 7
GROWTH AND NATURE OF THE INDUSTRY
A. PRESENT TECHNOLOGY
The recent trend in WPPA manufacture has been toward larger
capacity, enclosed producing units with closer control of opera-
ting variables. Single, multicompartment tanks have replaced
the earlier multiple tank systems and increased capacities from
the older design capabilities of 180 to 270 metric tons P205 per
day. Today a modern plant can produce 450 to 1,100 metric tons
P20s daily (17). Improved engineering design and materials of
construction have decreased capital and operating costs per unit
capacity and have improved overall operating efficiency in WPPA
manufacture. Recent production rates for WPPA are shown in
Figure 62.
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
f'ROJECTED
1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980
PRODUCTION YEAR
Figure 62. WPPA production trend (3).
NSP was, for many years, the major agricultural source of phos-
phate nutrient. In 1947 NSP accounted for over 90% of the total
domestic supply. Since the mid 1950's, however, the popularity
of NSP has undergone a sharp decline, and only in the past few
years has the rate of decline started to moderate. Production
142
-------
has fallen steadily from 1,150,000 metric tons (P205) in 1960 to
439,000 metric tons in 1975 (9, 33). NSP consumption data are
shown in Figure 63. The number of plants manufacturing NSP has
shown a similar drop, from an estimated 200 plants located
throughout the United States in the mid 1960's to 66 plants in
1975. The major reason for this decline is the poor economics of
converting phosphate rock to a lower analysis material (NSP) with
the associated increased cost of transportation per metric ton of
nutrient—as compared with the production and distribution of the
same phosphate values via more concentrated products.
1,900
1.800
1.700
1.600
1.500
1,400
1.300
1,200
1.100
1,000
900
800
700
600
500
400
300
200
1966
__„ "2% GROWTH RATE
,^ PROJECTED
NSP
PROJECTED
1ft to 5% DECLINE
1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982
YEAR
Figure 63.
Superphosphate fertilizer consumption
from 1966 to 1982 (29).
The simplicity of the NSP production process will act as a
moderating influence in the continued decline of NSP output. NSP
can be manufactured in small, inexpensive plants with low produc-
tion costs per ton of P20s since calcium sulfate (CaSO4) formed
in the acidulation of the rock is not separated from the final
product as in WPPA manufacture. The process is simple and easy
to operate and requires less sulfuric acid per metric ton of
P205 than does WPPA production.
TSP consumption in the United States has undergone a very rapid
growth during the past quarter century. Recent consumption data
143
-------
are shown in Figure 64. Production has shown a fivefold increase
in the period from 1950 to 1975. In 1975 an estimated 1,496,000
metric tons (100% P20s basis) were manufactured (9).
In the period from 1950 to 1965 TSP, with its higher P205 content,
took over much of the market lost by NSP. Since 1966, TSP has
typically represented 25% to 35% of the total annual domestic
P2O5 fertilizer supply. At the present time, TSP is the second
leading source of fertilizer phosphate (9). Although TSP produc-
tion has maintained a moderate growth rate, it has declined in
importance relative to ammonium phsophate because the latter has
grown at a much faster rate.
The market for ammonium phosphate has expanded at the expense of
the declining NSP and TSP market as shown in Table 56. From zero
in 1950, the market share for ammonium phosphates grew to 14.3%
in 1960 and 46.1% in 1970. Annual production and capacity data
are shown in Figure 64.
5,000
Figure 64.
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80
YEAR
Ammonium phosphate capacity, production, and plant
utilization projections to 1980 (7, 9-11, 34).
144
-------
TABLE 56. DISTRIBUTION OF PHOSPHATIC FERTILIZER MATERIALS (47)
Percentage of fertilizer market
Normal superphosphates
Triple' superphosphates
Ammonium phosphates
Others3
Total
1950
85
15
0
0
100
1960
46.4
35.7
14.3
3.6
100
1970
13.5
26.9
46.1
13.5
100
1980
4.1
20.2
58.1
17.6
100
2000
1.5
19.5
50.7
28.3
100
Growth in other phosphatic fertilizers is primarily super-
phosphoric acid, a main supplier to the liquid fertilizer
market.
During a study of the ammonium phosphate industry to determine
water discharge practices, several plants were found that no
longer produced ammonium phosphates in 1977. A list of these
plants is presented in Table 57. Had these plants been closed in
1975, nationwide capacity would have been 500,000 metric tons
P205 lower, but production would have remained the same. Plant
utilization for 1975 would have been 63% instead of the reported
56%; the mean annual plant capacity would have been 116,000
metric tons P20s instead of 103,000 metric tons P205; the average
annual utilization rate would have been 74% instead of 73%; and
all severities would have been 14% greater.
TABLE 57. PLANTS IDENTIFIED AS NO LONGER OPERATING IN 1977
1975 Capacity,
Company Location 103 metric tons
Farmland Industries, Inc.
Gardinier, Inc.
Kaiser Aluminum & Chemical
Kaiser Steel
Mississippi Chemical Corp.
Pennzoil Co.
Standard Oil of California
USS Agri-Chemicals
Total
Joplin, MO
Helena, AR
Wendover, UT
Fontana , CA
Pascagoula, MS
Hanford, CA
Fort Madison, IA
Kennewick , WA
Richmond, CA
Cherokee , AL
10 plants
84
45
14
14
139
Unknown
73
27
36
68
500
B. EMERGING TECHNOLOGY
The higher energy requirement for production of thermal process
phosphoric acid has caused investigation of processes to improve
the purity of WPPA. The wet process requires about one-fifth
of the energy per ton of product required in the thermal process
(78). Because of the high cost and uncertainty of electric
power, immediate expansion of the thermal process is not foreseen
(78). Growth in the marketing areas of thermal acid will likely
be met by improved quality WPPA. Although new cleanup processes
do not produce food-grade acid, the improved quality acid can be
used in detergent and animal feed applications.
145
-------
A chemical method for purification of 32% wet process acid in-
volves neutralization and precipitation of impurities, producing
acid of detergent phosphate specifications. The two-stage neu-
tralization process generates three vapor streams and two filter
cake effluents (78) . One plant, Olin Corp. in Joliet, Illinois,
uses this process to produce sodium phosphates.
A second method of cleanup involves solvent extraction. The
P205 content of the impure aqueous solution of phosphoric acid
made by the wet process is extracted with an immiscible organic
solvent; e.g., n-butanol. Impurities are left behind in the
aqueous layer, and regeneration of the phosphoric acid is
accomplished by contacting the organic phase with fresh water
(22).
Fluorine liberated in the production of phosphate fertilizers
could constitute a major supply of fluorine as deposits of
fluorspar and cryolite are depleted. More restrictive and expen-
sive fluoride control requirements are increasing emphasis on
the potential value of waste fluorides, and increased effort in
recovery of salable byproducts is expected (22).
The market for TSP for the near future is expected to remain
relatively constant primarily due to the tremendous growth of
ammonium phosphate. Currently the major source of fertilizer
phosphate in the United States, ammonium phosphates are produced
by reacting phosphoric acid with ammonia. Eighty-four percent
of ammonium phosphate production is in the form of DAP. The
increased use of DAP is attributable to several factors. It has
a high water solubility, high analysis (18% N and 46% PaOs), good
physical characteristics, and low production cost. In addition,
the phosphate content of DAP (46%) is as high as that of TSP, so
that by comparison the 18 units of nitrogen can be shipped at
no cost.
The most likely new phosphate material to become available in
the next few years is ammonium polyphosphate (APP) made from
merchant-grade WPPA. Its market potential is based upon the
likelihood that its production economics can be competitive with
those of DAP and that it will be useful as a base for liquid and
suspension fertilizers. The nutrient analysis for APP (12-57-0)
is higher than that of DAP, and APP has demonstrated good stor-
age and handling properties. APP has been made in pilot plant
studies at TVA by reacting merchant-grade WPPA in a two-step
reaction system. The second stage is a pipe reactor in which a
melt is formed. The melt passes through a vapor disengagement
vessel and is discharged into a pugmill for granulation (87).
(87) Phillips, A. B. New Products for the Future. In: TVA
Fertilizer Bulk Blending Conference, Tennessee Valley
Authority Bulletin Y-62, Muscle Shoals, Alabama,
August 1973. pp. 23-27.
146
-------
C. INDUSTRY PRODUCTION TRENDS
The fertilizer segment of the phosphoric acid industry is cur-
rently facing a domestic overcapacity due to a number of new
plants that have either come on stream or are expected within the
next year or two. If U.S. agricultural commodity exports are
expanded and sustained on a long-term basis, the upswing could
help ease the overcapacity. Between 1974 and 1980, wet acid
capacity will have increased by 56% while, over the same time
period, total domestic demand is expected to grow by 44% to 54%.
This results in an annual increase in production of 4% to 7%
(Figure 62) .
A shift from furnace phosphoric acid to WPPA is evident for pro-
duction of sodium polyphosphates (used in detergents) and some
animal feeds. The high energy and pollution control requirements
for thermal phosphoric acid have also stimulated this trend.
Growth in these areas would somewhat lessen the overcapacity
problems caused by recent lower-than-expected demand for
phosphate fertilizers.
Superphosphoric acid has a number of advantages over the more
dilute 54% P20s phosphoric acid, and growth in this area is
promising. The foremost advantage of super acid is shipping
economy. A 7% to 10% average annual growth rate between 1977
and 1980 is projected in Figure 65.
a
o
600
500
400
300
;200
I
100
Q.
OC.
UJ
a.
•=>
PROJECTED
1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980
PRODUCTION YEAR
Figure 65. Superphosphoric acid production trend (3).
Total production of superphosphates and other phosphatic fertil-
izer materials in 1975 was 4,896,000 metric tons (100% PzOs), a
slight increase over the 4,870,000 metric tons produced in 1974
(9). Production of NSP accounted for 9% of the total, a decline
147
-------
of 4% from the preceding year. TSP, on the other hand, has main-
tained a nearly constant share of the phosphatic fertilizer
market of 25% to 35% for the 9-yr period dating from 1967 to 1975
(29). The future growth in the phosphate fertilizer industry is
shown in Figure 63.
Declining demand for ROP-NSP and ROP-TSP is forecast (29). This
prediction is based on an expected continued growth of mechanical
blends of granulated concentrates (DAP, GTSP) and a further
decline of N-P-K fertilizers produced at ammoniator-granulator
plants (5). N-P-K ammoniator-granulator plants are primary con-
sumers of ROP superphosphate materials. In addition, a shift in
raw material usage for N-P-K plants is expected. Consumption of
superphosphoric acid as a preferred phosphate source is expected
to expand at the expense of NSP and TSP. The more concentrated
superphosphoric acid (70% to 72% P20s) provides economy by mini-
mizing shipping costs. An annual decline of from 1% to 5% is
predicted for NSP fertilizers.
GTSP will continue as a preferred phosphate source in low- and
no-nitrogen mixed fertilizer blends. TSP production (within the
limits of current technology) serves as the most convenient means
of "disposing" of high-sludge-containing phosphoric acids. Pro-
duction of sludge acids will maintain the incentive for steady or
expanded output of GTSP.
As a result of all the preceding factors, TSP production is
expected to experience an average annual growth rate of from 1%
to 3%. The growth in GTSP production will be at least partially
offset by a decline in ROP-TSP production.
Ammonium phosphate popularity as a fertilizer material is pro-
jected to result in continued growth of production and share of
the phosphatic fertilizer market until the 1980's (Figure 64).
From 1975 to 1980, production is projected to grow at an annual
pace of approximately 7.5%, while capacity is estimated to
decline at an annual rate of approximately 0.5% over the same
period. The net result will be an increase in plant utilization
rate from 56% to 83%. Ammonium phosphate's share of the phos-
phate fertilizer market is projected to be 58.1% in 1980 and
50.7% in 2000 (47) as shown in Table 56.
148
-------
REFERENCES
1. Riegel's Handbook of Industrial Chemistry, Seventh Edition,
J. A. Kent, ed. Van Nostrand Reinhold Co., New York, New
York, 1974. pp. 551-569.
2. Handbook of Chemistry and Physics, 49th Edition, R. C.
Weast, ed. The Chemical Rubber Co., Cleveland, Ohio, 1968.
p. B-187.
3. Stowasser, W. F. Phosphate-1977. Publication No. MCP-2,
U.S. Department of the Interior, Bureau of Mines, Washing-
ton, D.C., May 1977. 18 pp.
4. Fullam, H. T., and B. P. Faulkner. Inorganic Fertilizer and
Phosphate Mining Industries—Water Pollution and Control
(PB 206 154). Grant 12020 FPD, U.S. Environmental Protec-
tion Agency, Cincinnati, Ohio, September 1971. 225 pp.
5. Rawlings, G. D., and R. B. Reznik. Source Assessment:
Fertilizer Mixing Plants. EPA-600/2-76-032c, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Car-
olina, March 1976. 187 pp.
6. Inorganic Chemicals 1976. M28A(76)-14, U.S. Department of
Commerce, Washington, D.C., August 1977. 30 pp.
7. Hargett, N. World Fertilizer Capacity-Computer Printout.
Tennessee Valley Authority, Muscle Shoals, Alabama, 1976.
20 pp.
8. TVA Plans Early Closure of Furnaces; Cities Switch to Wet-
Process Phosphoric. Chemical Marketing Reporter, 209(3),
1976.
9. Inorganic Fertilizer Materials and Related Products.
M28B(75)-13, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
10. Harre, E. A., M. N. Goodson, and J. D. Bridges. Fertilizer
Trends 1976. Bulletin Y-lll, National Fertilizer Develop-
ment Center, Tennessee Valley Authority, Muscle Shoals,
Alabama, March 1977. 45 pp.
149
-------
11. Final Guideline Document: Control of Fluoride Emissions
from Existing Phosphate Fertilizer Plants. EPA-450/2-77-
005 (PB 265 062), U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, March 1977. 277 pp.
12. Atmospheric Emissions from Wet-Process Phosphoric Acid Manu-
facture. AP-57 (PB 192 222), U.S. Department of Health,
Education, and Welfare, Raleigh, North Carolina, April 1970.
86 pp.
13. Lowenheim, F. A. Phosphorus Compounds, Inorganic. In:
Encyclopedia of Industrial Chemical Analysis, Volume 17,
F. D. Snell and L. S. Ettre, eds. John Wiley & Sons, Inc.,
New York, New York, 1973. pp. 142-144.
14. Guimond, R. J. , and S. T. Windham. Radioactivity Distribu-
tion in Phosphate Products, By-Products, Effluents, and
Wastes. ORP/CSD-75-3, U.S. Environmental Protection Agency,
Washington, D.C., August 1975. 30 pp.
15. Phosphoric Acid, Volume I, A. V. Slack, ed. Marcel Dekker,
Inc., New York, New York, 1968. 1159 pp.
16. Heller, A. N., S. T. Cuffe, and D. R. Goodwin. Inorganic
Chemical Industry. In: Air Pollution, Volume III: Sources
of Air Pollution and Their Control, A. C. Stern, ed. Aca-
demic Press, New York, New York, 1968. pp. 221-231.
17. Martin, E. E. Development Document for Effluent Limitations
Guidelines and New Source Performance Standards for the
Basic Fertilizer Chemicals Segment of the Fertilizer Manu-
facturing Point Source Category. EPA-440/l-74-011-a (PB 238
652), U.S. Environmental Protection Agency, Washington,
D.C., March 1974. 170 pp.
18. Caro, J. H. Characterization of Superphosphate. In:
Superphosphate: Its History, Chemistry, and .Manufacture.
U.S. Department of Agriculture, Washington, D.C., December
1964. pp. 272-284.
19. Dahlgren, S. E. Chemistry of Wet-Process Phosphoric Acid
Manufacture. In: Phosphoric Acid, Volume I, A. V. Slack,
ed. Marcel Dekker, Inc., New York, New York, 1968.
pp. 91-154.
20. Evaluation of Emissions and Control Techniques for Reducing
Fluoride Emissions from Gypsum Ponds in the Phosphoric Acid
Industry. Contract 68-02-1330, Task 3, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
November 1976. 218 pp.
150
-------
21. Slack, A. V. Chemistry and Technology of Fertilizers. John
Wiley & Sons, Inc., New York, New York, 1967. pp. 69-97.
22. Lehr, J. R. Purification of Wet Process Acid. In: Phos-
phoric Acid, Volume I, A. V. Slack, ed. Marcel Dekker,
Inc., New York, New York, 1968. pp. 637-686.
23. Hill, W. L., H. L. Marshall, and K. D. Jacob. Composition
of Crude Phosphoric Acid Prepared by Sulfuric Acid Process.
Industrial and Engineering Chemistry, 24 (9) :1064-1068 , 1932.
24. Lutz, W. A., and C. J. Pratt. Principles of Design and
Operation. In: Phosphoric Acid, Volume I, A. V. Slack,
ed. Marcel Dekker, Inc., New York, New York, 1968.
pp. 158-208.
25. Cleanup Pays Off for Fertilizer Plant. Environmental
Science and Technology, 6 (5) :400-401, 1972.
26. Banford, C. R. IMC's New Plant Shows Off Latest ^PO^ Know-
How. Chemical Engineering, 70 (11) :100-102, 1963.
27. Legal, C. C., and 0. D. Myrick, Jr. History and Status of
Phosphoric Acid. In: Phosphoric Acid, Volume I,
A. V. Slack, ed. Marcel Dekker, Inc., New York, New York,
1968. pp. 1-89.
28. Huffstutler, K. K. Pollution Problems in Phosphoric Acid
Production. In: Phosphoric Acid, Volume I, A. V. Slack,
ed. Marcel Dekker, Inc., New York, New York, 1968.
pp. 727-739.
29. Muehlberg, P. E., J. T. Reding, and B. P. Shepherd. Draft
Report: The Phosphate Rock and Basic Fertilizer Materials
Industry. Contract 68-02-1329, Task 81, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
May 1976. 205 pp.
30. Personal communication with Ed Harre, Tennessee Valley
Authority, Muscle Shoals, Alabama, 14 April 1977.
31. Background Information for Standards of Performance: Phos-
phate Fertilizer Industry, Vol. 1 — Proposed Standards.
EPA-450/2-74-019a (PB 237 606), U.S. Environmental Protec-
tion Agency, Raleigh, North Carolina, October 1974. 140 pp.
32. Gartrell, F. E., and J. C. Barber. Pollution Control Inter-
relationships. Chemical Engineering Progress, 62 (10) :44-47 ,
1966.
151
-------
33. Final Guideline Document: Control of Fluoride Emissions
From Existing Phosphate Fertilizer Plants. EPA-450/2-77-005
(PB 265 062), U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1977. 277 pp.
34. David, M. L. , J. M. Malk, and C. C. Jones. Economic Analy-
sis of Effluent Guidelines Fertilizer Industry. EPA-230/2-
74-010 (PB 241 315), U.S. Environmental Protection Agency,
Washington, D.C., January 1974.
35. The Condensed Chemical Dictionary, Eighth Edition,
G. G. Hawley, ed. Van Nostrand Reinhold Company, New York,
New York, 1971. p. 54.
36. Chemistry and Technology of Fertilizers, V. Sauchelli, ed.
Reinhold Publishing Corp., New York, New York, 1960.
pp. 251-268.
37. Waggaman, W. H. Phosphoric Acid, Phosphates, and Phosphatic
Fertilizers, Second Edition. Reinhold Publishing Corp.,
New York, New York, 1952. pp. 308-344.
38. Himmelblau, D. M. Basic Principles and Calculations in
Chemical Engineering, Second Edition. Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey, 1967. pp. 449-454.
39. Kirk-Othmer Encyclopedia of Chemical Technology, Second
Edition, Vol. 9. John Wiley & Sons, Inc., New York,
New York, 1966. pp. 46-132.
40. Slack, A. V. Fertilizer Developments and Trends. Noyes
Development Corp., Park Ridge, New Jersey, 1968. pp. 77-274,
41. Kirk-Othmer Encyclopedia of Chemical Technology, Second
Edition, Vol. 15. John Wiley & Sons, Inc., New York,
New York, 1968. p. 260.
42. Shreve, R. N. Chemical Process Industries, Third Edition.
McGraw-Hill Book Company, New York, New York, 1967.
pp. 274-277.
43. Chopey, N. P. Diammonium Phosphate: New Plant Ushers in
Process Refinements. Chemical Engineering, 69(6):148-150,
1962.
44. Vandegrift, A. E., L. J. Shannon, E. W. Lawless, P. G. Gor-
man, E. E. Sallee, and M. Reichel. Particulate Pollutant
System Study, Vol. Ill—Handbook of Emission Properties.
APTD-0745 (PB 203 522), U.S. Environmental Protection
Agency, Durham, North Carolina, 1971. pp. 313-335.
152
-------
45. Hardison, L. C. Air Pollution Control Technology and Costs
in Seven Selected Areas. EPA-450/3-73-010 (PB 231 757),
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, December 1973. pp. 11-192.
46. Achorn, F. P., and H. L. Balay. Systems for Controlling
Dust in Fertilizer Plants. In: TVA Fertilizer Conference,
Tennessee Valley Authority Bulletin Y-78, Muscle Shoals,
Alabama, August 1974. pp. 55-62.
47. Phosphate Fertilizer Plants Final Guideline Document
Availability. Federal Register, 42 (40) :12022-12023, 1977.
48. Hoffmeister, G. Quality Control in a Bulk Blending Plant.
In: TVA Fertilizer Bulk Blending Conference, Tennessee
Valley Authority Bulletin Y-62, Muscle Shoals, Alabama,
August 1973. pp. 59-70.
49. Barber, J. C. Environmental Control in Bulk Blending
Plants. 1. Control of Air Emissons. In: TVA Fertilizer
Bulk Blending Conference, Tennessee Valley Authority
Bulletin Y-62, Muscle Shoals, Alabama, August 1973.
pp. 39-46.
50. Inorganic Fertilizer Materials and Related Products.
M28B(75)-11, U.S. Department of Commerce, Washington,
D.C., January 1976. 6 pp.
51. Inorganic Fertilizer Materials and Related Products.
M28B(75)-12, U.S. Department of Commerce, Washington,
D.C., February 1976. 6 pp.
52. Inorganic Fertilizer Materials and Related Products.
M28(76)-l, U.S. Department of Commerce, Washington, D.C.,
March 1976. 6 pp.
53. Inorganic Fertilizer Materials and Related Products.
M28(76)-2, U.S. Department of Commerce, Washington, D.C.,
April 1976. 6 pp.
54. Inorganic Fertilizer Materials and Related Products.
M28(76)-3, U.S. Department of Commerce, Washington, D.C.,
May 1976. 6 pp.
55. Inorganic Fertilizer Materials and Related Products.
M28(76)-4, U.S. Department of Commerce, Washington, D.C.,
June 1976. 6 pp.
56. Inorganic Fertilizer Materials and Related Products.
M28(76)-5, U.S. Department of Commerce, Washington, D.C.,
July 1976. 6 pp.
153
-------
57. Inorganic Fertilizer Materials and Related Products.
M28(76)-6, U.S. Department of Commerce, Washington, D.C.,
August 1976. 6 pp.
58. Inorganic Fertilizer Materials and Related Products.
M28(76)-7, U.S. Department of Commerce, Washington, D.C.,
September 1976. 6 pp.
59. Inorganic Fertilizer Materials and Related Products.
M28(76)-8, U.S. Department of Commerce, Washington, D.C.,
October 1976. 6 pp.
60. Inorganic Fertilizer Materials and Related Products.
M28(76)-9, U.S. Department of Commerce, Washington, D.C.,
November 1976. 6 pp.
61. Inorganic Fertilizer Materials and Related Products.
M28(76)-10, U.S. Department of Commerce, Washington, D.C.,
December 1976. 6 pp.
62. King, W. R., and J. K. Ferrell. Fluoride Emissions from
Phosphoric Acid Plant Gypsum Ponds. EPA-650/2-74-021, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, October 1974. 329 pp.
63. English, M. Fluorine Recovery from Phosphatic Fertilizer
Manufacture. Chemical Process Engineering, 48 (12):43-47,
1967.
64. Bowers, Q. D. Disposal as Waste Material - U.S. Practice.
In: Phosphoric Acid, Volume I, A. V. Slack, ed. Marcel
Dekker, Inc., New York, New York, 1968. pp. 505-510.
65. Huffstutler, K. K., and W. E. Starnes. Sources and Quan-
tities of Fluorides Evolved from the Manufacture of Ferti-
lizer and Related Products. Journal of the Air Pollution
Control Association, 11 (12) :682-684, 1966.
66. Preliminary Report: Remote Monitoring of Fluoride Emission
from Gypsum Ponds. EPA-69/01-4145, Task 10, U.S. Environ-
mental Protection Agency, Washington, D.C., November 1977.
35 pp.
67. Volk, W. Applied Statistics for Engineers, Second Edition.
McGraw-Hill Book Co., New York, New York, 1969. 110 pp.
68. Turner, D. B. Workbook of Atmospheric Dispersion Estimates,
Public Health Service Publication No. 999-AP-26, U.S. De-
partment of Health, Education, and Welfare, Cincinnati,
Ohio, 1969. 62 pp.
154
-------
69. Reznik, R. B. Source Assessment: Flat Glass Manufacturing
Plants. EPA-600/2-76-032b, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1976.
147 pp.
70. TLV® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1976. American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, 1976. 94 pp.
71. National Emissions Data System Point Source Listing.
SCC 3-01-030-01, 3-01-030-02, 3-01-030-99, 1976. 190 pp.
72. 1972 National Emissions Report; National Emissions Data
System (NEDS) of the Aerometric and Emissions Reporting
System (AEROS). EPA-450/2-74-012, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina,
June 1974. 422 pp.
73. Eimutis, E. C., and R. P. Quill. State-by-State Listing of
Source Types that Exceed the Third Decision Criterion,
Special Project Report. Contract 68-02-1874, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, July 7, 1975. pp. 1-3.
74. Seinfeld, J. H. Air Pollution: Physical and Chemical
Fundamentals. McGraw-Hill Book Co., New York, New York,
1975. 523 pp.
75. Environmental Engineers' Handbook, Volume 2, Air Pollution,
B. G. Liptak, ed. Chilton Book Co., Radnor, Pennsylvania,
1974. 1340 pp.
76. Specht, R. C., and R. R. Calaceto. Gaseous Fluoride Emis-
sions from Stationary Sources. Chemical Engineering Prog-
ress, 63(5):78-84, 1967.
77. Frazier, A. W., E. F. Dillard, and J. R. Lehr. Chemical
Behavior of Fluorine in the Production of Wet Process Phos-
phoric Acid. Presented at the American Chemical Society
Annual Meeting, Chicago, Illinois, August 24-29, 1975.
16 pp.
78. Environmental Considerations of Selected Energy Conserving
Manufacturing Process Options, Vol. 13, Phosphorus/Phos-
phoric Acid Industry Report. EPA-600/7-76-034m (PB 264
279), U.S. Environmental Protection Agency, Cincinnati,
Ohio, December 1976. 96 pp.
79. 40 CFR 418, Fertilizer Manufacturing Point Source Category,
Subpart A - Phosphate Subcategory. Federal Register,
41(98):20582-20585, 1976.
155
-------
80. Quality Criteria for Water. EPA-440/9-76-023, U.S. Envir-
onmental Protection Agency, Washington, D.C., 1976.
pp. 16-21.
81. Manual of Treatment Techniques for Meeting the Interim
Primary Drinking Water Regulations. EPA-600/8-77-005, U.S.
Environmental Protection Agency, Cincinnati, Ohio, May 1977.
73 pp.
82. Eimutis, E. C., J. L. Delaney, T. J. Hoogheem, S. R. Archer,
J. C. Ochsner, W. R. McCurley, T. W. Hughes, and R. P. Quill
Source Assessment: Prioritization of Stationary Water
Pollution Sources. EPA-600/2-77-107p, U.S. Environmental
Protection Agency, Washington, D.C., December 1977. 119 pp.
83. Water Resources Data for Florida, Water Year 1975.
Volume 3 - West-Central Florida Surface Water, Ground Water,
Quality of Water. USGS-WRD-FL-75-3 (PB 259 493), U.S. De-
partment of Commerce, Tallahassee, Florida, July 1976.
1249 pp.
84. Water Resources Data for Kentucky, Water Year 1975. USGS-
WDR-KY-75-1 (PB 251 853), U.S. Department of Commerce,
Louisville, Kentucky, January 1976. 348 pp.
85. Personal communication with John Sweeney, U.S. Bureau of
Mines, Tallahassee, Florida, 26 September 1977.
86. Murakami, K. By-product Recovery, As Raw Material for
Plaster and Cement - Japanese Practice. In: Phosphoric
Acid, Volume I, A. V. Slack, ed. Marcel Dekker, Inc.,
New York, New York, 1968. pp. 519-523.
87. Phillips, A. B. New Products for the Future. In: TVA
Fertilizer Bulk Blending Conference, Tennessee Valley
Authority Bulletin Y-62, Muscle Shoals, Alabama,
August 1973. pp. 23-27.
88. Background Information for Standards of Performance: Phos-
phate Fertilizer Industry, Vol. 2—Test Data Summary.
EPA-450/2-74-019b (PB 237 607), U.S. Environmental Protec-
tion Agency, Raleigh, North Carolina, October 1974. 63 pp.
89. Run of the Pile Triple Superphosphate. Contract 68-02-0232,
Test Report 73-FRT-10, U.S. Environmental Protection Agency,
Washington, D.C., September 1972. 45 pp.
90. Normal Superphosphate Plant. Contract 68-02-0232, Test
Report 73-FRT-15, U.S. Environmental Protection Agency,
Washington, D.C., June 1973. 32 pp.
156
-------
91. Granular Triple Superphosphate Storage. Contract 68-02-0232,
Test Report 72-CI-30B, U.S. Environmental Protection Agency,
Washington, D.C., June 1972. 32 pp.
92. Sanders, L. Monitoring and Control of Gaseous and Particu-
late Emissions from Fertilizer Complex. Presented at the
69th Annual Meeting of the Air Pollution Control Association
(Paper No. 76-56), Portland, Oregon, June 27-July 1, 1976.
14 pp.
93. Standard for Metric Practice. ANSI/ASTM Designation E 380-
76S IEEE Std 268-1976, American Society for Testing and
Materials, Philadelphia, Pennsylvania, February 1976.
37 pp.
157
-------
APPENDIX A
PHOSPHATE FERTILIZER PLANTS IN THE UNITED STATES IN 1975 OR 1976
Table A-l lists 1975 ammonium phosphate production figures.
Tables A-2 through A-6 describe the phosphate fertilizer plants
operating in the United States in 1975 or 1976, of which there
were 36 producing WPPA, Table A-2 (7); 9 producing superphos-
phoric acid, Table A-3 (7); 66 producing NSP, Table A-4 (7); 16
producing TSP, Table A-5 (10); and 48 producing ammonium phosphate,
Table A-6 (7, 10, 11). Plant lists were modified by MRC based on
communications with industry representatives. The company name
and location of each plant are provided in the tables, along with
plant production capacity and population density of the county
where the plant is located.
In order to have a consistent industry characterization, a conver-
sion factor from P205 to product was needed for ammonium phos-
phate production. Using U.S. Department of Commerce data for the
industry in 1975 (Table A-l), a conversion of
Mass product = Mass P205(2.49) (A-l)
was generated and used.
Pure DAP has a 4.0:1.0 mole ratio of N:P205, and MAP has a ratio
of 2.0:1.0. According to the U.S. Department of Commerce, the
ratio for DAP in 1975 was 4.25:1.0; for other phosphates it was
4.47:1.0. This suggests that more nitrogen than phosphorus was
being tied up by impurities; i.e., nitrogen was reacting with
materials other than phosphoric acid, forming either soluble or
insoluble salts. The nutrient analysis for U.S. Department of
Commerce DAP was 18-44-0, very close to the most common WPPA DAP
product (18-46-0). The analysis for other phosphates was 12-27-0.
Along with the high N:P205 ratio, this suggests the presence of
large quantities of ammonium salts other than phosphates; e.g.,
ammonium sulfate. The analysis could result from a mixture of
common MAP fertilizers: 11-48-0, 11-55-0, 13-52-0, and 16-20-0.
158
-------
TABLE A-l. 1975 PRODUCTION OF AMMONIUM PHOSPHATES (50-61)
ui
/ Production, metric tons
Diammonium phosphate
Month
January
February
March
April
May
June
July
August
September
October
November
December
Total
P»0s
156,407
160,105
168,923
180,179
207,453
197,970
174,783
193,985
215,897
232,775
210,999
211,479
2,310,955
N
63,829
65,399
76,236
90,631
84,717
81,642
71,743
79,058
87,497
95,105
86,361
85,444
967,662
Gross
320,894
362,044
378,444
392,222
455,907
461,246
430,664
464,931
507,117
537,130
469,638
473,742
5,253,979
Other ammonium phosphates
P50q
32,537
39,691
39,057
36,410
35,091
32,999
39,672
39,122
32,172
37,770
38,324
37,031
439,876
N
16,083
17,340
19,869
15,771
17,613
14,101
16,554
16,866
14,982
15,083
14,928
14,558
193,749
Gross
138,223
147,550
165,781
142,098
145,515
121,874
139,626
127,679
107,617
113,214
128,903
129,803
1,607,883
Total ammonium phosphates
Ps Os
188,944
199,797
207,980
216,589
242,544
230,969
214,456
233,107
248,068
270,545
249,323
248,510
2,750,831
N
79,912
82,740
96,105
106,402
102,330
95,742
88,297
95,924
102,479
110,189
101,289
100,002
1,161,411
Gross
459,117
509,594
544,225
534,320
601,421
583,120
570,290
592,610
614,734
650,344
598,541
603,545
6,861,862
-------
TABLE A-2. WPPA PLANTS IN THE UNITED STATES IN 1975 (7)
Company
Allied Chemical Corp., Union
Texas Petroleum Division
Beker Industries Corp.
Borden, Inc., Smith-Douglas
Division
C F Industries, Inc.
Engelhard Minerals and
Chemicals Corp.
Farmland Industries, Inc.
First Mississippi Corp.
Freeport Minerals Co.
The Gardinier Companies:
Gardinier, Inc.
Gardinier Big River, Inc.
W. R. Grace and Co., Agri-
cultural Products Group
International Minerals and
Chemicals Corp.
Mississippi Chemical Corp.
Mobil Oil Corp. , Agricultural
Chemicals Division
North Idaho Phosphate Co.
Location
Geismar, LA
Conda, ID
Marseilles, IL
Taft, LA
Piney Point, FL
Streator, IL
Bar tow, FL
Plant City, FL
Nichols, FL
Greenbay, FL
Ft. Madison, IA
Uncle Sam, LA
Tampa , FL
Helena, AR
Bar tow, FL
New Wales, FL
Pascagoula, MS
Depue, IL
Kellogg, ID
Design production
capacity,
metric tons ?205/day
(short tons P205/ldayl
454(500)
690(760)*
290(320)
600(660)
454(500)
55(60)a
1,900(2,100)
725(800)
1,090(1,200)
363(400)
1,270(1,400)
658(725)
2,087(2,300)
1,360(1,500)
136(150)
907(1,000)
1,655(1,824)
590(650)
354(390)
82(90)
County population
density,
persons/km2
46.7
1.4
36.9
37.1
49.0
36.9
46.1
180.1
46.1
46.1
30.7
29.6
466.8
22.3
46.1
46.1
44.8
16.9
2.9
(continued)
160
-------
TABLE A-2 (continued)
Company
Occidental Petroleum Corp. ,
Occidental Chemical Co.
01 in Corp., Agricultural
Chemicals Division
Pennzoil Co.
Royster Co.
J. R. Simplot Co., Minerals
and Chemicals Division
Stauffer Chemical Co. , Fertil-
izer and Mining Division
Texasgulf, Inc., Agricultural
Division
Union Oil Co. of California
United States Steel Corp. ,
USS Agri-Chemicals Division
Valley Nitrogen Producers,
Inc.
The Williams Companies,
Agrico Chemical Co.
Location
Lanthrop, CA
White Springs, FL
Pasadena, TX
Joliet, IL
Hanford, CA
Mulberry , FL
Pocatello, ID
Pasadena, TX
Salt Lake City, UT
Aurora, NC
Nichols, CA
Bar tow, FL
Ft. Meade, FL
Bakersfield, CA
Helm, CA
Donaldsville, LA
South Pierce, FL
Design production
capacity,
metric tons P20s/day
(short tons P205/day)
110(120)a
635(700)
662(730)
345(380)
658(725)
417(460)
55(60)3
381(420)
522(575)
290(320)
163(180)
181(200)
953(1,050)
476(525)
476(525)
23(25)
272(300)
535(590)
18(20)
127(140)
227(250)
1,143(1,260)
771(850)
County population
density,
j?ersons/km2
77.7
5.8
385.9
274.9
17.8
46.1
17.5
385.9
235.7
16.1
49.0
46.1
46.1
15.4
4.9
46.7
46.1
MRC estimate.
161
-------
TABLE A-3. SUPERPHOSPHORIC ACID
IN 1975 WHICH DERIVE
PLANTS IN THE UNITED STATES
THEIR PRODUCT FROM WPPA (7)
Company
Allied Chemical Corp., Union Texas
Petroleum Division
Farmland Industries, Inc.
International Minerals and Chemicals
i— '
en North Idaho Phosphate Co.
M
Occidental Petroleum Corp.
Occidental Chemical Co.
J. R. Simplot Co., Minerals
and Chemicals Division
Stauffer Chemical Co., Fertilizer
and Mining Division
Texasgulf Inc., Agricultural
Division
Design production County
capacity, population
metric tons P2Os/day Type of density.
Location (short tons PjO^/day) concentration persons/tan^
Geismar , LA
Greenbay, FL
Bonnie , FL
Kellogg, ID
White Springs, FL
Pocatello, ID
Pasadena, TX
Salt Lake City, UT
Aurora, NC
417(460)
454(500)
460(508)
36 (40)
227(250)
118(130)
104(115)
77(85)
72(79)
113(125)
275(303)
275(303)
269(297)
Submerged
combustion
Vacuum
evaporation
Vacuum
evaporation
Vacuum
evaporation
Submerged
combustion
Vacuum
evaporation
Vacuum
evaporation
Vacuum
evaporation
Vacuum
evaporation
46.7
46.1
46.1
2.9
5.8
17.5
385.9
235.7
16.1
-------
TABLE A-4. NSP PLANTS IN THE UNITED STATES IN 1976 (7)
Plant name
American Plant Food Corp.
Borden Inc., Chemical Division
Centra la Farmers Coop. , Inc.
Columbia Nitrogen Corp.
Farmers Fertilizer
Gardinier, Inc.
Georgia Fertilizer
Cilchrist Plant Food
Gold Kist, Inc.
W. R. Grace and Co.
Indiana Faro Bureau Coop. Association, Inc.
International Minerals and Chemical Corp.
Kaiser Aluminum and chemical Corp. ,
Agriculture Chemicals Division
Kerr-McGee Corp.
Layco
Lone Star Co., NIPAK, Inc., Subsidiary
Mineral Fertilizer Co.
Occidental Petroleum Corp.,
Occidental Chemical Co., Subsidiary
Ohio Valley Fertilizer
Pelham Phosphate Co.
Richmond Guano Co.
Royster Co.
Southern States Phosphate and Fertilizer Co.
Stauffer Chemical Co.
Swift and Co., Swift Chemical Co., Division
Texaco, Inc.
U.S. Steel Corp.,
USS Agri-Chemicals Division
Valley Nitrogen Producers, Inc.
Weaver Fertilizer Co., Inc.
The Williams Companies,
Agrico Chemical Co., Inc., Subsidiary-
Location
Fort Worth, TX
Houston, TX
Norfolk, VA
Russelville, KY
Streator, IL
Forkland, AL
Moultrie, GA
Texarkana, TX
Tampa, FL .
Valdosta, GA
Morris, IL
Clyo, GA
Cordele, GA
Charleston , SC
Joplin, MO
Nashville, TN b
Indianapolis, IN
Americus, GA
Augusta, GA .
Chicago Heights, IL
Florence, AL L.
Fort Worth, TX
Hartsville, SC
Spartanburg, SC
Winston Salem, NC
h
Acme, NC
Baltimore, MD
Cottondale, FL .
Jacksonville, FL
Philadelphia, PA
Lakeland, FL
Nacagodoches , TX
Midvale, UT
Ashkum, IL
White Springs, FL
London, KY
pelham, GA
Richmond, VA
Athens, GA
Chesapeake, VA
Jackson, MS
Savannah, GA
Tacoma, MA
Bartow, FL b
Birmingham, AL.
Charleston, SC
Do than, AL h
Norfolk, VA
Savannah, GA
Wilmington, NC
Omaha, HE"
Albany, GA .
Columbus, GA
Nashville, TN
Navassa, NC
Bakersfield, CA
Norfolk, VA.
Buffalo, NY
Cairo, OH" b
Charleston. SC
Fulton, IL" b
Greensboro , NC
Pensacola , . FL"
Pierce, FL b
saginaw, MI b
Walnut Ridge, AK
Production
capacity ,
metric tons
PjOs/yr
a
97070
27,200
6,000
32,700
5,440
11,600
9,070
6,350
17,200
3,630
7,260
10,900
' 14,500
13,600
13,600
12,700
8,160
a
107900
12,700
12,700
a
107900
_a
20,000
9,980
11,800
17,200
16,300
a
47340
2,720
6,350
6,350
16,300
8,160
10,900
3,630
9,980
5,440
24,500
9,980
13,600
11,800
6,350
12,700
9,070
6,350
9,070
10,900
13,000
6,350
14,000
15,000
a
137600
9,980
9,980
8,160
21,800
5,440
6,350
8,160
10,900
5,440
County
population
density,
persons/km2
316.3
385.6
1,991 .6
14.7
36.9
6.2
21.6
19.9
180.1
40.7
23.0
10.7
23.4
94.6
47.3
401.8
757.5
20.9
192.2
2,195.6
39.0
316.3
36.9
78.8
189.2
18.7
397.4
13.8
258.7
5,858.7
46.1
14.9
235.7
11.2
5.8
23.3
14.1
2,519.6
196.5
99.8
92.3
158. 7
. 93.3
46.1
221.1
94.6
37.0
1,991.6
158.7
160.6
446.1
105.7
2,195.6
401.8
167.9
15.4
1,991.6
402.5
104.3
94.6
34.7
167.9
116.7
46.1
103.3
10.6
Plant production capacity not available.
Plants closing during calendar year 1976 or
during the first half of 1977.
163
-------
TABLE A-5. TSP PLANTS IN THE UNITED STATES IN 1976 (10!
a
Production capacity,
metric tons P2Os/yr
/
-y
S
V
J
Plant name
Borden Inc., Chemical Division feoub)
CF Industries, Inc.
Engelhard M. and C.
Farmland Industries, Inc.
Gardinier, Inc.
W. R. Grace and Co.
International Minerals and Chemical Corp.
Mississippi Chemical Corp.
Occidental Petroleum Corp.,
Occidental Chemical Co. , Subsidiary
Royster Co. RoP--T£T>
J. R. Simplot Co.
Stauffer Chemical Co.
Texasgulf, Inc.
U.S. Steel Corp. ,
USS Agri-Chemicals, Division
The Williams Companies,
Agrico Chemical Co. , Subsidiary
Location
Piney Point, FL
Plant City, FL
Nichols, FL
Pierce , FL
Tampa , FL
Bar tow, FLu
Joplin, MO
Bonnie, FL
Pascagoula , MS
White Springs, FL
Mulberry, FL
Pocatello, ID
Garfield, UT
Lee Creek, NC
Fort Meade, FL
b
Pierce, FL
GTSP
29,900
190,000
0
79,000
190,000
163,000
0
125,000
114,000
70,800
0
35,000
20,800
1^,000
110,000
140,000
ROP-TSP
0
150,000
117,000
0
150,000
128,000
40,800
0
0
0
88,000
27,600
16,400
102,000
0
110,000
County
population
density,
persons/km2
49.0
180.1
46.1
46.1
180.1
46.1
47.3
46.1
44.8
5.8
46.1
17.5
235.7
16.1
46.1
46.1
For those plants which produced both GTSP and ROP-TSP, a product distribution of 56% GTSP and
44% ROP-TSP was assumed, based on total share of the market.
k
Plants closing during calendar year 1976 or during the first half of 1977.
-------
TABLE A-6.
AMMONIUM PHOSPHATE PLANTS IN THE
UNITED STATES IN 1975 (7, 10, 11)
Company
'Allied Chemical Corp.
Beker Industries
Borden Chemical Co. Lfl/n^-n)
Brewster Phosphates
CF Industries, Inc.
Conserv, Inc.
Farmland Industries, Inc.
First Mississippi Corp.
Ford Motor Co.
Gardinier, Inc.
W. R. Grace & Co.
Gulf Resources
IMC Chemicals Corp.
Kaiser Aluminum & Chemical
Kaiser Steel
Mississippi Chemical Corp.
Mobil Chemical Co.
Monsanto Industrial Chemicals Co.
Nipak, Inc.
North Idaho Phosphate Co.
Northwest Coop Mills
Occidental Petroleum Corp.,
Occidental Chemical Co. , Subsidiary
Olin Corp.
Pennzoil Co. .
Phosphate Chemicals
Royster Co.
J. R. Simplot Co.
Standard Oil Co. of California
Stauffer Chemical
Tennessee Valley Authority
Texas Gulf, Inc.
Union Oil Co. of California
USS Agri-Chemicals
Valley Nitrogen Producers
Williams Companies,
Agrico Chemical Co., Subsidiary
U.S. total
Location
Geismar, LA
Conda, ID
Taft, LA
Piney Point, FL
Luling, LA
Geismar, LA
Bonnie, FL
Plant City, FL
Nichols, FL
Pierce, FL
Joplin, HO
Fort Madison, IA
Dearborn, HI
Tampa , FL
Helena, AR
Bartow, FL
Kellogg, ID
Bartow, FL
Wendover, UT
Fontana , CA
Pascagoula, MS
Depue, IL
Trenton, MI
Kerens, TX
Kellogg, ID8
Pine Bend, MN
White Springs, FL
Lathrop, CA
Plainview, TX
Pasadena, TX
Han ford, CA
Pasadena, TX
Mulberry, FL
Pocatello, ID
Fort Madison, aIA*
Kennewick, WA
Richmond, CAa
Garfield, UT
Muscle Shoals, AL
Lee Creek, NC
Nichols, CA8
Cherokee, AL
Bartow, FL
Helm, CA
Fresno, CA
Chandler, AZ
Pierce, FL
Donaldsonvllle, LA
Production
103 metric tons
PaQB/yr
123
146
182
^1
136
45
577 /
250 I/
109 /
164 '
84
155
9
227 /
45
95 /
19
227^
14
14
139
114
16
30
24
63
275 •
16
9
209
_b
45
45-'
73'
73
27
36
51
18
92 •/
20
68
IS^
32
-b
11
42 /
687
4,926
capacity
10 3 metric tons
product/yr
306
365
454
193
340
113
1,440
624
272
408
209
386
23
567
113
238
48
567
34
34
347
283
41
75
59
156
687
41
23
522
_b
113
113
181
181
68
91
127
45
229
50
170
32
n9
_D
27
104
1,714
12,292
County
population
density,
persons/km3
46.7
1.4
37.1
49.0
96.2
46.7
46.1
180.1
46.1
46.1
47.3
30.7
1,686.3
180.1
22.3
46.1
2.9
46.1
1.2
12.9
44.8
13.0
1,686.3
10.8
2.9
92.9
5.8
77.7
13.2
385.9
17.8
385.9
46.1
17.5
30.7
14.9
151.8
235.7
31.8
16.1
49.0
31.8
46.1
26.3
26.3
4.9
46.1
46.7
1973 Gross capacities reported in Reference 10 assumed unchanged for 1975.
Capacity information unavailable.
165
-------
APPENDIX B
EMISSIONS DATA
The following tables present stack test data from individual
plants used to calculate emission factors.
TABLE B-l. WPPA PLANT SOURCE TEST DATA FOR ROCK UNLOADING5
Plant
A
B
C
Averages
Test
production,
metric tons
PaOs/hr
63.3
12.7
23.8
32.3
Controlled
particulate
emission
factor,
g/kg PaOs
0.18
0.26
0.017
0.15 ± 250%
Stack
height,
m
14
b
10
12
Emissions data on file at the Florida Depart-
ment of Environmental Regulation in Winter
Haven.
No data available.
TABLE B-2.
WPPA PLANT SOURCE TEST DATA FOR ROCK
TRANSFER AND CHARGING TO REACTOR3
Plant
A
B
D
E
Averages
Test
production,
metric tons
P20s/hr
63.3
50.6
18.5
22.8
38.8
Controlled
particulate
emission
factor.
g/kg P20s
0.006
0.012
0.062
0.10
0.045 ± 180%
Stack
height.
m
17
b
18
27
21
Emissions data on file at the Florida Depart-
ment of Environmental Regulation in Winter
Haven.
No data available.
166
-------
TABLE B-3. WPPA PLANT SOURCE TEST DATA FOR WET SCRUBBER SYSTEM
a,b
Plant
A
B
o
F
G
H
I
J
K
L
H
N
O
Averages
Test
production,
metric tons
P,05/hr
24.
23.
23.
34.
41.
14.
29.
19.
39.
6.
18.
18.
8.
19.
50.
15.
37.
24.
2
9
7
0
6
4
3
4
4
0
4
4
5
3
6
3
8
9
Controlled emission factors. Recovers
j/kg PjOs byproduct Controlled fluorine emission factor
Total
fluorine
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.012
.007
.011
.0033
.0042
.0083
.0033
.0055
.0039
.035
.011
.0025
.012
.012
.009
.011
.010
.010
of
g/kg P,0
Particulate- PzOs SO* fluorine With recovery
0.045 No
0.040 No
' 0.030 No
No
No
0.036 0.0077 No
0.0011 Yes
Yes
0.053 Yes
0.17 No
Yes
Yes
NO
No
0.0038 Yes
NA
NA
0.058 NA
0.029 NA
0.054 0.038 0.032
NAC
NA
NA
NA
NA
NA
0.0033
0.0055
0.0039
NA
0.011
0.0025
NA
NA
0.009
NA
NA
NA
NA
0.0059
5
Without recovery
0
0
0
0
0
0
0
0
0
0
.012
.007
.011
.0033
.0042
.0083
NA
NA
NA
.035
NA
NA
.012
.012
NA
NA
NA
NA
NA
.012
s. Stack
height.
m
37
17
21
28
37
31
29
Data for plants A through K from plant test data on file at the Florida Department of Environmental Regulation in Winter
Haven. Data for plants L and M from material balances shown earlier in Table 19. SOx emission factors for plants N and O
taken from data available in Reference 12.
Blanks indicate no data available.
°Not applicable.
TABLE B-4. SUPERPHOSPHORIC ACID PLANT SOURCE TEST DATA FOR WET SCRUBBER SYSTEM
Plant
AA
BB
CC
Averages
Test
production,
metric tons
PaOs/hr
9.0
12.5
20.44
14.0
Controlled
g/
Total
fluorine
b
0.0036
0.011
0.0073
emission factor.
/kg PzOs
Particulate
0.011 to 0.055
Stack
height.
m
27
21
15
21
Emissions data on file at the Florida Depratment of Environ-
mental Regulation in winter Haven.
bBlanks indicate no data available.
-------
TABLE B-5. PLANT SOURCE TEST DATA FOR ROP-TSP MANUFACTURE
00
Reported controlled
emission factor
Controlled emission factor,
g/kg P205
Date
of
Plant Production rate
A 14.2
14.5
15.1
16.2
16.1
16.3
6.46
B 33. b
34.8
35.8
Cd'6 37.6
38.2
40.4
20.1
20.3
40.5
20.1
50.7
54.4
tons
tons
tons
tons
tons
tons
tons
tons
tons
tons
metric
metric
metric
metric
metric
metric
metric
metric
metric
a
c
tons
tons
tons
tons
tons
tons
tons
tons
tons
TSP/hr
TSP/hr
TSP/hr
P2Os/hr
P2Os/hr
TSP/hr
P205/hr
TSP/hr
TSP/hr
Source of emissions
Mixer, den, storage
Mixer, den, storage
Mixer, den, storage
Mixer, den, storage
Mixer , den , storage
Mixer, den, storage
Mixer , den , storage
Mixer, den, storage
Mixer, den, storage
Mixer, den, storage
Mixer, den, storage
Mixer , den , storage
Mixer, den
Mixer, den
Mixer , den
Mixer, den
Mixer , den
Mixer, den
Mixer, den
Screening, milling
and shipping
Mixer, den
Fluoride
0.321 Ib/ton P20s
0.228 Ib/ton P2Os
0.035 Ib/ton P2Os
0.181 Ib/ton P2Os
0.304 Ib/ton P2Oj
0.147 Ib/ton P2Os
0.074 to 0.345 Ib/ton P2O5
0.319 Ib/ton P2O5
0.125 Ib/ton P2Os
0.126 Ib/ton P2Os
0.125 Ib/ton P2Os
0.022 to 0.230 Ib/ton P2OS
13.9 Ib/day
17.1 Ib/day
0.66 Ib/hr
0.45 Ib/hr
0.25 Ib/hr
0.74 Ib/hr
0.58 Ib/hr
3.6 Ib/day
0.37 Ib/hr
Particulate
7.0
7.2
2.0
2.1
2.6
4.5
3.54
4.4
2.2
Ib/hr
Ib/hr
Ib/hr
Ib/hr
Ib/hr
Ib/hr
lt/hr
Ib/hr
Ib/hr
Fluoride Particulate
0.161
0.114
0.018
0.091
0.152
0.074
0.037 to 0.173
0.160
0.063
0.063
0.063
0.011 to 0.115
0.022
0.026
0.015
0.01
0.005
0.017
0.013
0.0043
0.007
0.17
0.18
0.046
0.048
0.06
0.103
0.080
0.09
0.038
analysis Reference
2/72
2/72
3/72
9/72
9/72
9/72
1972
1975
9/72
9/72
9/72
1972
1974
1974
1975-76
2/76
5/76
1975-76
5/76
1974
1975-76
88
88
88
88
88
88
88
_b
89
89
89
89
b
™w
D
~K
v?
~K
U
-.
-P
"b
Range of 35 stack tests made by plant operator not included in developing average emission factor.
Emissions data on file at the Florida Department of Environmental Regulation in Winter Haven.
CRange of 51 stack tests made by plant operator not included in developing average emission factor.
Assuming 16 hr/day plant operation to convert from pounds per day to pounds per hour and 49% P2Oj content in ROP-TSP to convert from pounds TSP to
pounds P2°5-
8Fluoride emission measurements for Plant C were not included in the statistical analysis due to the fact that a curing dryer is used in place of
the curing building.
(88) Background Information for Standards of Performance: Phosphate Fertilizer Industry, Vol. 2—Test Data Summary. EPA-450/2-74-019b
(PB 237 607), U.S. Environmental Protection Agency, Raleigh, North Carolina, October 1974. 63 pp.
(89) Run of the Pile Triple Superphosphate. Contract 68-02-0232, Test Report 73-FRT-10, U.S. Environmental Protection Agency, Washington, D.C.
September 1972. 45 pp.
-------
TABLE B-6. PLANT SOURCE TEST DATA FOR NSP MANUFACTURE
h-'
CTi
Plant
A 2.89
2.33
2.70
B 18.1
18.1
18.1
18.1
18.1
C 22.5
14.2
Production rate-
metric
metric
metric
metric
metric
metric
metric
metric
metric
metric
tons
tons
tons
tons
tons
tons
tons
tons
tons
tons
P205/hr
P205/hr
P2O5/hr
NSP/hr
NSP/hr
NSP/hr
NSP/hr
NSP/hr
NSP/hr
NSP/hr
Reported controlled
emission factor
Source
Mixer,
Mixer,
Mixer,
Curing
Curing
Curing
Curing
Mixer,
Mixer,
Mixer,
Mixer,
Mixer,
Mixer,
Mixer,
of emissions
den
den
den
building
building
building
bui Iding
den
den
den
den
den
den
den
0
0
0
0
0
0
0
0
0
4
0
Fluoride
.061 Ib/ton P2O5
.03 Ib/ton P2O5
.03 Ib/ton P2O5
.4380 Ib/hr
.6665 Ib/hr
.101 Ib/hr
.154 Ib/hr
.141 Ib/ton NSP*!
.083 Ib/ton NSP
.5 Ib/day
.12 Ib/hr
Particulate
0
0
0
0
2
0
3
0
0
0
0
0
d °
0.205 Ib/ton NSP .
0.0986 Ib/ton NSP
0.140 Ib/ton NSpd
1.5 Ib/hr 0
0.30 Ib/hr 0
Controlled emission factor. Date
g/kg P2Og of
Fluoride Particulate analysis Deference
.031
.015
.015b
.073°
. 43-uncontrolled
.lllb
. 70-uncontrolled
.017^
. 57-uncontrolled
.026b c
. 85-uncontrolled
.353
.208
.0565e
.0196
1974-75
3/76
3/76
8/74
1/75
3/76
3/76
6/73
6/73
0.52 6/73
0.247 6/73
0.35 6/73
0.156 1974
0.048 1975
a
~a
"a
^a
_a
a
_a
90
90
90
90
90
a'
"a
Emissions data on file at the Florida Department of Environmental Regulation. Assuming average plant production rate of 3 tons P2Os/hr.
C d
Using reported scrubber control efficiency of 97% for fluoride removal. Assuming a 20% P2Os content in the NSP product.
Assuming 8 hr/day operation and a 20% P2Os content in the NSP.
(90) Normal Superphosphate Plant. Contract 68-02-0232, Test Report 73-FRT-15, U.S. Environmental Protection Agency, Washington, D.C., June 1973.
32 pp.
-------
TABLE B-7. PLANT SOURCE TEST DATA FOR GTSP MANUFACTURE
Controlled emission
factor, g/kg P2O5
Reported controlled emission factor Fluo- SOx Partic-
Plant
A
B
C
D
Production rate
31.4 metric tons
P205/hr
33. 1 metric tons
rock/hr
54.4 metric tons
GTSP/hr
12. 2 metric tons
PjO5/hr
13.8 metric tons
P2O5/hr
13.7 metric tons
P2O5/hr
245 metric tons
P205/day
363 metric tons
1,815 metric tons
p2Os stored
1,815 metric tons
P2O5 stored
1,815 metric tons
^2^5 stored
4,084 metric tons
^2^*5 stored
4,085 metric tons
PjOs stored
4,084 metric tons
P2°5 stored
3,525 metric tons
PjOt, stored
277 metric tons
P205/day
274 metric tons
P205/day
428 metric tons
P^05/day
386 metric tons
P205/day
356 metric tons
P205/day
420 metric tons
P205/day
443 metric tons
P20s/day
395 metric tons
P20j/day
128,123 metric tons
rock/yr
130,137 metric tons
GTSP/yr
Source of emissions
Reactor, granulator .
cooler, dryer, screens
Rock feeder
Shipping and curing
building
Reactor, granulator.
cooler, dryer, screens
Reacto , granulator.
cool r, dryer, screens
Reacto , granulator.
cool r , dryer , screens
Reacto , granulator ,
cooler, dryer, screens
Reactor, granulator.
cooler, dryer, screens
Reactor , granulator ,
cooler, dryer , screens
Reactor, granulator.
cooler, dryer, screens
Curing building
Curing building
Curing building
Curing building
Curing building
Curing building
Curing building
Curing building
Reactor, granulator.
cooler, dryer, screens
Reactor , granulator ,
cooler, dryer, screens
Reactor, granulator.
cooler, dryer, screens
Reactor, granulator.
cooler, dryer, screens
Reactor , granulator ,
cooler , dryer , screens
Reactor , granulator ,
cooler, dryer, screens
Reactor , granulator ,
cooler, dryer, screens
Reactor, granulator.
cool er , dryer , screens
Rock feeder
Shipping and curing
building
Fluoride SOx as SO2
1.276 Ib/hr 153.6 Ib/hr3
0.278 Ib/ton P2O5
0.174 Ib/ton P205
0.182 Ib/ton P2O5
0.28 Ib/ton P205
0.17 Ib/ton P2OS
0.18 Ib/ton P2OS
0.24 Ib/ton P2OS
0.06 Ib/ton P205
0.0007 Ib/hr P205
0.0002 Ib/hr ton
P2Os stored
0.0005 Ib/hr ton
P2Os stored
0.00006 Ib/hr ton
P2O5 stored
0.00005 Ib/hr ton
P2Os stored
0.00006 Ib/hr ton
P20$ stored
0.00015 Ib/hr ton
P2Oj stored
0.537 Ib/ton P2O$
149 Ib/day
183 Ib/day
182 Ib/day
178 Ib/day
179 Ib/day
169 Ib/day
206 Ib/day
Particulate ride as SO; ulate
0.02 2.22
/•
0.3217 Ib/hr 0.004
3 Ib/hr 0.055C
0.139
O.OS7
0.091
0.14
0.09
0.09
0.12
0.03
d
0.042
f\
0.012
A
0.030
H
0.008
H
0.007°
A
0.008
.
0.013
0.269
0.25
0.20
0.22
0.23
0.20
O.18
0.24
0.04 Ib/ton 0.019
rock . •
O.01 Ib/ton 0.44
CTSP
Date of
analysis
5/76
5/76
6/72
6/72
6/72
1975
1975
9/72
9/72
9/72
6/72
6/72
6/72
5/72
6/76
1976
1975
1974
1973
1972
1971
1970
7/76
7/76
Reference
_b
b
_b
88
88
88
91
91
91
_.b
_b
91
91
91
83
88
88
88
_b
b
b
b
_b
b
_b
b
b
_b
See footnotes at end of table, p. 172.
-------
TABLE B-7 (continued)
Production rate
Source of emissions
Reported controlled eniasion factor
Fluoride SOx as SO2 Particulate
Controlled emission
factor , g/kq P2Os
Fluo-
ride
SOx
as 502
Partic-
ulate
Date of
analysis Reference
52 metric tons
GTSP/hr
18 metric tons
rock/hr
45 metric tons
rock/hr
Reactor, granulator,
cooler, dryer, screens
Rock feeder
Rock unloading
O.OS Ib/ton P205 1,866 Ib/day8 0.41 Ib/ton 0.03
P205
0.06 Ib/ton
rock
0.14 Ib/ton
rock
0.028
0.07C
10/75
10/75
G 21.1 metric tons Reactor, granulator, 0.026 Ib/ton P2Os 0 0.013 0 1/76 ' _b
P2O5/hr dryer, screens
21.1 metric tons Reactor, granulator, 0.028 Ib/ton P2°5 ° 0.014 0 1/76 _b
P2O5/hr dryer, screens
21.1 metric tons Reactor, granulator, 0.026 Ib/ton P20s 0 0.013 0 1/76 b
P205/hr dryer, screens
21.1 metric tons Rock unloading 5 Ib/hr 0.11 1/76 b
P205/hr
704 metric tons Curing building 0.19 Ib/hr 0 0.003 0 1/76 b
P20s/day
704 metric tons Curing building 0.188 Ib/hr 0 0.003 O .1/76 b
P2O5/day
704 metric tons Curing building 0.37 Ib/hr 0 0.006 0 1/76 b
P205/day
H 10.9 metric tons Reactor, granulator, 0.06 Ib/ton P2Os 0.03 9/72 88
P2O5/hr cooler, dryer, screens
10.9 metric tons Reactor, granulator, 0.18 Ib/ton P2Os 0.09 1/72 88
P2O5/hr cooler, dryer, screens
10.9 metric tons Reactor, granulator, 0.12 Ib/ton P2Os 0.06 1/72 88
P2O5/hr cooler, dryer, screens
1,316 metric tons Reactor, granulator, 0.00042 Ib/hr ton 0.033' 88
P2Og stored cooler, dryer, screens
1/316 metric tons Reactor, granulator, 0.00031 Ib/hr ton 0.025 33
P20>5 stored cooler, dryer, screens
1,316 metric tons Reactor, granulator, 0.0003S Ib/hr ton 0.02?" ee
P2O5 stored cooler, dryer, screens
SOx emission factor based on the consumption and analysis of fuel oil burned in the dryer.
Emissions data on file at the Florida Department of Environmental Regulation.
Assuming 0.42 ton of rock is consumed in the production of 1 ton of GTSP and that GTSP product contains 46% P20s by weight.
Assuming average daily production rate of 400 tons P2Os.
Assuming that GTSP dust contains 2.5% fluoride.
t
'Assuming 24 hr/day operation.
g
Assuming plant operates at design production capacity of 221 tons P2G*5/day.
?JOT£.—Blanks indicate emission factor not measured during test.
{91) Granular Triple Superphosphate Storage. Contract 68-02-0232, Test Report 72-CI-30B, U.S. Environmental Protection Agency, Washington, D.C.
June 1972. 32 pp.
-------
TABLE B-8. STACK HEIGHTS FOR NSP PLANTS (72)
Plant name
Centrala Farmers Coop., Inc.
Gardinier, Inc.
W. R. Grace and Co.
International Minerals and Chemical Corp.
Kerr-McGee Corp.
Richmond Guano Co.
Swift and Co. ,
Swift Chemical Co., Division
U.S. Steel Corp. , USS
Agri-Chemicals Division
Weaver Fertilizer Co., Inc.
The Williams Companies,
Agrico Chemical Co., Inc., Subsidiary
Location
Forkland, AL
Tampa , FL
Charleston , SC
Florence, AL
Spartanburg , SC
Baltimore, MD
Cottondale, FL
Jacksonville, FL
Richmond , VA
Bar tow, FL
Norfolk, VA
Chicago Heights, IL
Norfolk, VA
Pensacola, FL
Main stack (mixer,
den) height, m
10.7
22.3
16.8
24.4
18.3
13.7
10.7
15.5
6.1
22
18.3
24.4
15.2
24.4
TABLE B-9. STACK HEIGHTS FOR GTSP PLANTS (72)
Plant name
Borden Inc., Chemical Division
CF Industries, Inc.
Gardinier, Inc.
Occidental Petroleum Corp.,
Occidental Chemical Co., Subsidiary
Texasgulf, Inc.
The Williams Companies,
Agrico Chemical Co., Inc., Subsidiary
Location
Piney Point, FL
Plant City, FL
Tampa,. FL
White Springs , FL
Lee Creek, NC
Pierce, FL
Main stack (reactor,
granulator, cooler, dryer,
screens) height, m
61.0
57.6
38.4
30.5
32.6
42.7
TABLE B-10. STACK HEIGHTS FOR ROP-TSP PLANTS (72)
Plant name
Location
Main stack (mixer,
den, curing
building) height, m
CF Industries, Inc.
Gardinier, Inc.
Royater Co.
Texasgulf, Inc.
Plant City, FL
Tampa, FL
Mulberry, FL
Lee Creek, NC
30.5
20.7
20.4
30.5
172
-------
TABLE B-ll. SAMPLING EMISSIONS DATA FOR AMMONIUM PHOSPHATE MANUFACTURE
a ,b
Controlled emission factor,
gAg PjO,
Plant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Total
fluoride
Source of emission NH-3 Particulate (as F) Reference
Dryer
Dryer
Cooler
Dryer
Dryer
Cooler
Dryer/cooler
Dryer/cooler
Ammoniator
Dryer/cooler
Ainntoniator
Ammoniator
Dryer/cooler
Ammoniator-granulator
Dryer/cooler
Ammoniator
Dryer/cooler
Ammoniator-granulator
Ammoniator
Ammoniator
Ammoniator
Dryer/cooler
Dryer/cooler
Amnoniator
Dryer
Cooler
Ammoniator
Ammoniator
Dryer/cooler
Dryer/cooler
Ammoniator
Dryer/cooler
Ammoniator
Dryer/cooler
Ammoniator
Ammoniator
Cooler/dryer
Ammoniator
Ammoniator
Ammoniator
Dryer/cooler
0.08
0.32
0.17
0.32
0.
0.
0.
11
06
32
0.54
0.
0.
<0.
<0.
0.
1.
0.
0.
0.
0.
0.
0.
7.
2.
5.
0.
0.
0.
0
0
0.
0.
0
0.
<0.
4.
0.
03
16
02
04
37
0
19
13
43
37
40
19
5
2
3
48
46
58
37
80
36
04
1
19
0.13
0.12
4.
0.
0.
0.
6
26
13
44
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
45
•Plant
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Source of emission
Ammoniator
Dryer
Dryer/cooler
Dryer
Dryer
Granulator
Mills and screens
Cooler
Granulator
Granulator
Cooler
Cooler
Granulator
Granulator
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Total plant
Controlled emission factor,
g/kg PjOc;
NH3 Particulate
Total
fluoride
(as F)
0.89
0.12
0.61
0.
0.
0.
0.
0.
0.
0.
0.
1.
0.
1.
0.05
0.14
d
105
0.06
0.05
0.04 0.
0.
0.
38
71
05
03
SO
13
10
16
5
47
8
21
15
09
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
003
004
003
001
019
019
026
005
044
028
038
025
050
080
037
072
076
081
043
018
004
014
017
035
029
064
022
016
019
0.009
Reference
70
70
70
C
"c
~c
~c
"c
"c
~c
~c
"c
"c
~c
65
65
65
65
65
C
~C
~c
"c
~c
92
C
~c
"c
"c
"c
~c
"c
c
"c
~c
"c
~c
88
88
C
^Emissions were reported in several ways, but all have been normalized to units of grus per kilogram of P2O5. Ammonium phosphate was assumed to be
46» P2O5.
The number of plants identified as sampled exceeds the number of plants producing ammonium phosphates. Due to anonymity of sampling sites, multiple
reportings of some plant emissions could not be eliminated.
CData obtained from public files at the Florida Department of Environmental Regulation, Winter Haven, Florida, June 1976.
This value was not used in determining the average emission factor since a plant with an acidic scrubber would not release this amount of ammonia
under normal conditions.
(92) Sanders, L. Monitoring and Control of Gaseous and Particulate Emissions from Fertilizer Complex, presented at the 69th Annual Meeting of the
Air Pollution Control Association (paper No. 76-56), Portland, Oregon, July 27 to July 1, 1976. 14 pp.
-------
APPENDIX C
MASS BALANCES
As an aid in the evaluation of emission factors/ mass balances
were developed for the production of NSP and TSP (ROP-TSP and
GTSP) . Balances were performed on the basis of phosphorus pent-
oxide and fluoride contents of rock, acid, and fertilizer product.
Production statistics used in the equations were those reported
by individual plants to the Florida Department of Environmental
Regulation.
NSP
Material balances performed on the basis of phosphorus pentoxide
(P20s) and fluoride (F ) involve only the rock and fertilizer
product in NSP production.
Assumptions
• NSP product contains 20% PaOs by weight.
• Phosphate rock contains 33% P2O5 and 3.8% F~.
• Cured NSP has a fluoride content ranging from 1.41% to
2.15% (6).
The PaOs balance to establish the amount of rock required, R, to
produce 1 metric ton of NSP product is :
R = 606 kg rock
Result
The production of 1 metric ton of NSP requires the consumption of
0.606 metric ton of rock.
In order to estimate the amount of fluorine that is lost to the
atmosphere or absorbed by the scrubbing medium, the difference
between the fluoride entering with the rock and that leaving with
the product is determined. Two cases will be considered in the
following analysis: Case I considers cured NSP product with a
174
-------
fluorine concentration of 1.41%, and Case II considers a fluorine
concentration of 2.15% in the product.
Case I
Fluoride entering in the rock:
Fluoride in fertilizer product:
k* NSP) =
Thus 8.9 kg F~/metric ton NSP (44.5 kg F~/metric ton P205) is
released during the production and curing operations. Therefore,
a scrubber efficiency of 99% would result in an emission factor
of 0.445 kg F /metric ton of P205 in the product.
Case II
Fluoride entering in the rock:
Fluoride in fertilizer product:
(°'0kg5NsgF")(1'°00 kg NSP) = 21.5 kg F~
Thus 1.5 kg F~/metric ton NSP (7.5 kg F~/metric ton P205) is
released during the production and curing operations.
In this case a scrubber efficiency of 99% would result in an
emission factor of 0.075 kg F /metric ton P205 in the product.
GTSP
Statistics for rock and acid consumption taken from those
reported by Plant A:
Phosphoric acid (40% P205) 55,600 kg/hr
Phosphate rock 29,500 kg/hr
Assumptions
• Phosphate rock contains 33% P20s and 3.8% F~ by weight.
Estimates based on values reported to the Florida Department of
Environmental Regulation.
175
-------
• Phosphoric acid contains 2.0% F .
• GTSP contains 46% P205 and 2.5% F~.
The P2®5 balance to establish the corresponding rate of GTSP pro-
duction is:
/0.33 kg P205\/29,500 kg rock\ /0.40 kg P20.s\ /55,600 kg acid\
\ kg rock A hr / \ kg acid )\ hr )
kg
kg
GTSP = 69,500 kg/hr
Results
The production of 1 metric ton of GTSP requires the consumption
of 0.8 metric ton of 40% phosphoric acid and 0.42 metric ton of
rock.
In order to estimate the amount of fluorine that is lost to the
atmosphere or absorbed by the scrubbing medium, the difference
between the fluoride entering with the rock and acid and that
leaving in the production is determined.
Fluoride entering in the rock and acid:
Fluoride remaining in product:
k^ GTSP) = 25
Thus 7 kg F~/metric ton GTSP (15.2 kg F~/metric ton P205) is
released during the production and curing of GTSP.
A scrubber efficiency of 99% would then result in an emission
factor of 0.15 kg F~/metric ton of P20s in the product.
ROP-TSP
Statistics for rock and acid production taken from those reported
by Plant A:
Phosphoric acid (56% P2O5) 22,000 kg/hr
Phosphate rock 12,200 kg/hr
Assumptions
• Fifty-six percent phosphoric acid contains 1.5% F .
176
-------
• Cured ROP-TSP contains 49% P205 and 2.0% F .
• Phosphate rock contains 33% P2°5 an^ 3.8% F
The ?20s balance to establish the corresponding rate of ROP-TSP
production is:
/0.33 kg P205\/12,200 kg rock\ /0.56 kg P205V22,000 kg acid\
\ kg rock /\ hr ) \ kg acid A hr )
kg
ROP-TSP = 33,400 kg/hr
The production of 1 metric ton of ROP-TSP requires the consump-
tion of 0.66 metric ton of 56% phosphoric acid and 0.37 metric
ton of rock.
In order to estimate the amount of fluorine that is lost to the
atmosphere or absorbed by the scrubbing medium, the difference
between the fluoride entering with the rock and acid and that
leaving in the product is determined.
Fluoride entering in rock and acid:
Fluoride leaving in product:
°
(kg
kgROP-TSp
(1'00° k* ROP-TSP) = 20 Kg
Thus 4 kg F /metric ton of ROP-TSP (8.2 kg F /metric ton P20s) is
released during the production and curing of ROP-TSP.
A scrubber efficiency of 99% would then result in an emission
factor of 0.082 kg F~/metric ton of P20s in the product.
177
-------
APPENDIX D
NEDS DATA BASE
Table D-l gives the state emissions burdens for the five criteria
pollutants as reported in the NEDS (72). Table D-2 is an updated
version of the NEDS data as computed by MRC under EPA contract
(73).
178
-------
TABLE D-l. NEDS EMISSION SUMMARY BY STATE (72)
Mass of emissions, Mtric tons/yr
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Oist. Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
U.S. totals
P articulates
1,178,643
13,913
72,685
137,817
1,006,452
201,166
40,074
36,808
19,451
226,460
404,574
61,621
55,499
1,143,027
748,405
216,493
348,351
546,214
380,551
49,155
494,921
96,160
705,921
266,230
168,355
202,435
272,688
95,338
94,040
14,920
151,768
102,785
160,044
481,017
78,978
1,766,056
93,595
169,449
1,810,598
13,073
198,767
52,336
409,704
549,399
71,692
14,587
477,494
161,934
213,715
411,558
75,427
16,762,000
sox
882,731
5,874
1,679,768
39,923
393,326
49,188
168,068
209,310
60,630
897,381
472,418
45,981
54,387
2,043,020
2,050,541
283,416
86,974
1,202,827
166,664
144,887
420,037
636,466
1,466,935
391,633
50,591
1,152,373
871,235
58,014
304,851
86,596
463,736
444,310
345,979
473,020
78,537
2,980,333
130,705
36,776
2,929,137
65,761
247,833
17,354
1,179,982
753,098
152,526
17,751
447,394
272,991
678,348
712,393
69,394
28,873,000
HOX
397,068
32,757
123,871
168,989
1,663,139
147,496
155,832
58,407
46,824
664,794
369,817
44,221
48,552
974,372
1,371,233
242,524
233,987
419,142
442,817
76,741
265,204
334,379
2,222,438
311,834
172,519
448,300
148,405
101,948
88,933
67,309
489,216
199,181
572,451
412,599
85,708
1,101,470
222,687
135,748
3,017,345
46,921
521,544
49,490
426,454
1,303,801
80,998
24,286
329,308
187,923
229,598
408,525
72,572
21,722,000
Hydrocarbons
643,410
28,389
189,981
195,538
2,160,710
193,456
219,661
63,886
41,789
619,872
458,010
89,530
84,230
1,825,913
600,477
316,617
309,633
326,265
1,919,662
122,918
295,867
440,481
717,891
410,674
195,950
413,130
271,824
127,821
53,673
88,469
819,482
152,057
1,262,206
447,238
70,289
1,153,493
341,358
234,669
891,763
65,833
907,833
90,478
362,928
2,218,891
98,282
41,980
369,416
344,643
116,155
523,930
55,319
23,994,000
CO
1,885,657
167,357
815,454
843,204
8,237,667
857,781
897,580
204,227
190,834
2,695,817
2,036,010
275,566
343,720
6,412,718
2,933,780
1,440,621
1,002,375
1,189,932
5,633,827
376,196
1,261,804
1,682,218
3,243,526
1,760,749
829,094
1,854,901
611,061
569,522
215,751
256,380
2,877,319
504,249
4,881,922
1,734,398
318,679
5,205,719
1,456,627
929,247
3,729,830
283,650
4,222,168
387,356
1,469,253
6,897,748
402,527
150,510
1,548,031
1,659,117
494,214
1,582,869
303,297
91,782,000
ADJUSTMENTS TO GRAND TOTAL
The U.S. summary does not include certain source categories. The following additions
should be considered part of the U.S. grand total for a more accurate picture of
nationwide emissions.
New York point
sources
Forest wild
fires
Agricultural
burning
Structural
fires
Coal refuse
fires
Total
U.S. subtotal
(above)
U.S. grand
total
311,000
375,000
272,000
52,000
100,000
1,110,000
16,762,000
17,872,000
993,000
0
15,000
0
128,000
1,076,000
28,873,000
29,949,000
382,000
88,000
29,000
6,000
31,000
536,000
21,722,000
22,258,000
127,000
529,000
272,000
61,000
62,000
1,051,000
23,994,000
25,045,000
44,000
3,089,000
1,451,000
200,000
308,000.
5,086,000
91,782,000
96,868,000
179
-------
TABLE D-2. STATE LISTING OF EMISSIONS AS OF JULY 2, 1975 (73)
Mass of emissions, metric tons/yr
Percent of U.S. totals
Partic-
State ulate
i
2
3
«
5
6
7
a
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
ALABAMA
ALASKA
ARIZONA
ARKANSAS
CALIFORNIA
COLORADO
CONNECTICUT
DELAWARE
FLORIDA
GEORGIA
HAWAII
IDAHO
ILLINOIS
INDIANA
IOWA
KANSAS
KENTUCKY
LOUISIANA
MAINE
MARYLAND
MASSACHUSETTS
MICHIGAN
MINNESOTA
MISSISSIPPI
nissouRi
MONTANA
2002000. 0
1.53000
16340000.0
12.50000
3265000.0
2.49000
1619000.0
1.24000
5675000.0
4.33000
3156000.0
2.41000
365600.0
O.Z7900
130200.0
0.09930
2430000.0
1.86000
2331000.0
1.78000
251200.0
0.19200
2430000.0
1.85000
3584000.0
2.74000
2202000.0
1.68000
2579000.0
1.97000
33S8000.0
2.56000
1854000.0
1.42000
1651000.0
1.260UO
1038000.0
0.79200
657300.0
0.50200
802700, 0
0.61300
2804000.0
2.14000
3036000.0
2.33000
1490000.0
1.14000
2839000.0
2.17000
4975000.0
3.80000
S02
1228000.0
1.91000
222600.0
0.34700
200200.0
0.31100
205400.0
0.31900
2557000.0
3.98000
473300.0
0.73600
1227000.0
1.91000
420700.0
0.65SOO
17SSOOO.O
2.73000
1635000.0
2.54000
232000.0
0.36100
59140.0
0.09200
J714000.0
5.78000
3056000.0
4.72000
397400.0
0.61800
225000.0
0.35000
1627000.0
2.53000
58S800.0
o.moo
770700. u
1,20000
1352000.0
2.10000
3840UOO.O
5.97000
3513000.0
5.46000
S46800.0
1.32000
280300.0
0.43600
1259000.0
1.96000
177000.0
0.27500
NOX
2bl60U.O
2.27000
31990.0
0.27700
7S100.0
0.65100
77310.0
0.67UOO
796800.0
6.91000
116800.0
1.01000
152200.0
1.32000
45720.0
0,99600
410300.0
3.56000
294200.0
2.55000
40790,0
0.35400
33220,0
0,28800
66510U.O
9.77000
414400.0
3.59000
137700.0
1.19000
109900.0
0.95.500
3020UU.O
2.62000
219000.0
1.900UO
54270.0
0.47000
215100.0
1.86000
322300.0
2.79BOP
548000.0
4.7SUOO
185000,0
1.60000
87010.0
0.7SHOO
287500.0
2.49000
346SO.O
0.3UOOO
Hydro-
carbons
342100.0
1.29000
140800.0
0.53200
171100.0
0.647UO
281700.0
1.07000
1914000.0
7,24000
294400. U
1.11000
259400.0
0.98100
77510.0
0.29300
536200,0
2.03000
526700,0
1.99000
62720.0
0.23700
163600.0
0.61900
1343000. U
5.08000
6-75100.0
2.55000
400800.0
1.52000
742800,0
2.81000
274600.0
1.040011
1741000.0
6.58000
71970.0
0.27200
302300.0
1.14000
463100.0
1.75000
734000.0
2.76000
388000.0
1.47000
350200.0
1.32000
588400.0
2.22000
174200.0
0.65800
CO
372600.0
7.04000
472200.0
2.58000
178300.0
0.9/600
225800.0
1.24000
1987000,0
10.90000
105800.0
0.57900
92690.0
0.50700
24580.0
0.14500
Sf.02000.0
19.20000
705400,0
3.86000
84750. 0
0.46400
518300.0
2,84000
412500.0
2.26000
182100.0
0.9*700
90720.0
0. »9700
174600.0
0.95600
2193UO.O
1.2UOOO
'•39900,0
4 . bOOOO
614.50.0
0.33600
163400.0
0.89400
1904UO.O
l.OHOOO
299400.0
\.64000
1507UO.O
0.82500
228200.0
1.25000
268500.0
1.47000
230500.0
1.26000
180
-------
TABLE D-2 '(continued)
Mass of emissions, metric tons/yr
Percent of U.S. totals
Partic-
State ulate
27
28
Z1
30
n
42
33
it
i5
36
17
36
39
40
HI
42
43
44
45
2710,0
0.54400
782200.0
6..78UOO
SB760.0
II. 33600
146300.0
1.27000
16560.0
0.16100
264100.0
2.29000
f. 15500.0
6.03000
48410.0
0.42000
13710.0
0.11900
197600.0
1.71000
126300.0
1.09000
306500.0
2.66000
231300.0
2.00000
70570.0
0.61200
11^00000.0
Hydro-
carbons
255600.0
0.96600
36140.0
0.13700
44430.0
0.16800
786600.0
2.97000
310200.0
1.17000
1353000.0
5.11000
465100.0
1.76000
73930.0
0.26000
1244000.0
4.70000
674700.0
2.S5000
204800.0
0.77400
1331000.0
5.0600(1
93730. U
U.354UO
260500.0
0.98"iUO
31110.0
0.34400
340900.0
1.29000
1139000.0
Ib. 60000
112600.0
0.42600
25460.0
0.0963U
415200.0
1.57000
361600.0
1.37000
172800.0
0.65300
362600.0
1.37000
275200.0
1.04000
26400000.0
CO
59590.0
0.32600
28700.0
0.1S700
302UO.O
•0.16500
281400.0
1.54000
49400.0
0.27100
551600.0
3.02000
371500.0
2.03000
223^0.0
0.12200
482700.0
2.64000
200800.0
1.10000
3049UO.O
l.fe'OOO
527000.0
2.86000
29390.0
0.1&100
483900.0
2.65000
23460.0
0.12900
?00300.0
1.10000
IbOlOOO.O
8.2^000
46640.0
0.2S600
14190.0
0.07770
235100.0
1.29000
425500.0
2.33000
435100.0
2,30000
161300.0
0.88300
20670.0
0.11400
1A300000.0
181
-------
GLOSSARY
acidulator: Reaction vessel where wet process reaction occurs.
affected population: Number of people around a typical plant who
are exposed to a source severity greater than 0.05 or 1.0.
ammoniation-granulation: Process in which a chemical reaction
with ammonia is combined with the physical process of
granule formation.
batch den: Enclosed compartment in which a liquid mix of acid
and phosphate rock is held until solidification occurs.
becquerel: Unit of radioactivity equal to one disintegration per
second, 3.7 x 1010 Bq = 1 curie.
beneficiation: Combined physical and chemical process used to
concentrate the phosphate value of phosphate rock ore.
blunger (pugmill): U-shaped trough in which paddles mounted on
twin contrarotating shafts agitate, shear, and knead a
solid-liquid mixture to produce granules.
BPL: Bone phosphate of lime or tricalcium phosphate, Ca3(P04)2-
concentrated phosphoric acid (merchant-grade phosphoric acid) :
Product of wet process phosphoric acid manufacture, approxi-
mately 53% P205.
contact process water: Any water which, during manufacturing or
processing, comes into direct contact with or results from
the production or use of any raw material, intermediate
product, finished product, byproduct, or waste product.
continuous den: Slow moving conveyor belt on which the liquid
mixture of acid and phosphate rock sets into a solid form.
curing: Process by which superphosphate fertilizer material is
held for a period of time ranging from a few days to a
number of weeks during which the acidulation reaction
continues.
emission factor: Quantity of a species emitted per unit of input
or product.
182
-------
filtered phosphoric acid: Product of wet process phosphoric acid
manufacture prior to concentration, approximately 29% P205.
fugitive emissions: Gaseous and particulate emissions that are
not emitted through a primary exhaust system such as a stack.
furnace process phosphoric acid: Phosphoric acid produced by
heating phosphate rock in a furnace, burning the resulting
elemental phosphorus, and hydrating it to phosphoric acid.
gypsum: Calcium sulfate dihydrate (CaSO^»2H20); byproduct of the
wet process reaction between phosphate rock and sulfuric
acid.
gypsum pond: Liquid waste receiver with the primary purpose of
separating solid gypsum (CaS(\) from a liquid stream result-
ing from the production of phosphoric acid manufacture.
Supernatant from the pond is used as a wet scrubbing liquor
to remove fluorides from exhaust gases in ammonium phosphate
production.
hazard factor: Value equal to the primary ambient air quality
standard for criteria pollutants or to a reduced TLV; i.e.,
(TLV)(8/24)(1/100) for noncriteria emissions.
liming: Water treatment process using lime [Ca(OH)2] to neutral-
ize waters and precipitate impurities.
merchant-grade phosphoric acid: See concentrated phosphoric acid.
melt: Molten fertilizer.
normal superphosphate: Fertilizer which contains from 16% to 21%
phosphorus pentoxide (P205) prepared by reacting ground
phosphate rock with sulfuric acid.
(N-P-K): Designation of fertilizer nutrient analysis:percent
total nitrogen—percent phosphorus expressed as P205--
percent potassium expressed as K20.
run-of-pile: Solid fertilizer material of a nonuniform particle
size.
representative plant: Typical plant defined to establish a base
on which to evaluate the emissions of the industry. The
plant has average industry parameters.
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).
183
-------
superphosphoric acid: Acid produced by concentration of 54% ^2^
to about 70% P2°5'
ten-yr, 24-hr rainfall event: Maximum 24-hr precipitation event
with a probable recurrence interval of once in 10 yr (as
defined by the U.S. National Weather Service).
threshold limit value: Airborne concentration of substances
under which it is believed that nearly all workers may be
repeatedly exposed day after day without adverse effect.
triple superphosphate: Fertilizer containing 45% or more
phosphorus pentoxide (P2C>5) prepared by reacting ground
phosphate rock with phosphoric acid.
wet process phosphoric acid: Phosphoric acid produced by react-
ing sulfuric acid with phosphate rock.
184
-------
CONVERSION FACTORS AND METRIC PREFIXES (93)
To convert from
Curie (Ci)
Degree Celsius (°C)
Joule (J)
Kilogram (kg)
Kilogram (kg)
Kilometer2 (km2)
Kilometer2 (km2)
Meter (m)
Meter2 (m2)
Meter2 (m2)
Meter3 (m3)
Meter3 (m3)
Metric ton
Pascal (Pa)
Pascal (Pa)
Pascal-second (Pa-s)
Second (s)
CONVERSION FACTORS
to
Becquerel
Degree Fahrenheit
British thermal unit
Pound-mass (avoirdupois)
Ton (short, 2,000 Ib mass)
Acre
Mile2
Foot
Acre
Foot2
Liter
Foot3
Ton (short, 2,000 Ib mass)
Atmosphere
Pound-force/inch2 (psi)
Poise
Minute
Multiply by
3.700 x 1010
> = 1.8 t° + 32
9.479 x I0~k
2.205
1.102 x 10~3
2.471 x 10-1*
3.861 x ID'1
3.281
2.471 x 1Q-1*
1.076 x 101
1.000 x 101
3.531 x 101
1.102
9.869 x 10~6
1.450 x 10"1*
1.000 x 101
1.667 x 10~2
METRIC PREFIXES
Prefix Symbol Multiplication factor
Kilo
Centi
Milli
Micro
Pico
k
c
m
y
P
Example
103
10-
10-
10"
10-
2
3
6
12
5
5
5
5
5
km
cP
mg
ug
pCi
=
5
5
5
5
5
x
X
X
X
X
103
10"
10-
10-
10"
2
3
6
1
meters
poise
gram
gram
2 curie
(93) Standard for Metric Practice. ANSI/ASTM Designation E 380-
76e, IEEE Std 268-1976, American Society for Testing and
Materials, Philadelphia, Pennsylvania, February 1976. 37 pp.
185
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-019C
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
SOURCE ASSESSMENT: Phosphate Fertilizer
Industry
5. REPORT DATE
1979
May
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J. M. Nyers, G. D. Rawlings, E. A. Mullen,
C. M. Moscowitz, and R. B. Reznik
8. PERFORMING ORGANIZATION REPORT NO.
MRC-DA-895
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
Box 8, Station B
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 A
Task Final;
T AND PERIOD COVERED
5/76 - 3/79
14. SPONSORING AGENCY CODE
EPA/600/13
... SUPPLEMENTARY NOTES TERL-RTP project officer is Ronald A. Venezia. MD-62, 919/541-
2547. There are other source assessment reports in this series and in EPA-600/2-
76-032. 77-107. and 78-004 series.
ABSTRACT
report describes a study of air emissions, water effluents, and solid
residues resulting from the manufacture of phosphate fertilizers. It includes the pro-
duction of wet process phosphoric acid, superphosphoric acid, normal superphos-
phate, triple superphosphate, and ammonium phosphate. Air emissions from pro-
duction of phosphate fertilizers include particulates , fluorides , ammonia, and sul-
fur oxides. Phosphate fertilizer plants control air emissions by a combination of
cyclones, baghouses, and wet scrubbers. Material handling operations are generally
enclosed to reduce fugitive particulate emissions. Only fluoride emissions from
curing and storage at normal superphosphate plants are typically uncontrolled. Water
effluents from the production operations arise from wet scrubbers, barometric con-
densers, steam ejectors, gypsum slurry, and acid sludge. Noncontact cooling water
is normally segregated from other wastewater streams. Wastewaters are contami-
nated with phosphates , fluorides, sulfates and gypsum. Process water is discharged
to large gypsum ponds for storage and recycle; it is normally not discharged to sur-
face streams. Solid residues generated at phosphoric acid plants are gypsum from
the filtration of wet process phosphoric acid, wet process phosphoric acid sludge,
and solids suspended in the wet scrubber liquor.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI I;icld/Group
Pollution
Assessments
Industrial Processes
Fertilizers
Phosphate Deposits
Phosphoric Acids
Dust
Fluorides
Ammonia
Sulfur Oxides
Pollution Control
Stationary Sources
Phosphate Fertilizers
Particulate
13 B
14 B
13H
02A
08G
07B
11G
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
201
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
186
------- |