WATER POLLUTION CONTROL RESEARCH SERIES • 12020 FPD 09/71
Inorganic Fertilizer and Phosphate
Mining Industries-Water Pollution
and Control
ENVIRONMENTAL PROTECTION AGENCY • RESEARCH AND MONITORING
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Environmental
Protection Agency, through inhouse research and grants
and contracts with Federal, State, and local agencies,
research institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Monitoring, Environmental
Protection Agency, Room 801, Washington, B.C. 20460.
-------
INORGANIC FERTILIZER AND PHOSPHATE MINING INDUSTRIES
WATER POLLUTION AND CONTROL
Prepared by
Battelle Memorial Institute
Richland, Washington
for the
Environmental Protection Agency
Grant No. 12020FPD
September, 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.75
-------
EPA REVIEW NOTICE
This report has been reviewed by the Water Quality Office,
EPA, and approved for publication. Approval does not
signify that the contents necessarily reflect the views
and policies of the Environmental Protection Agency, nor
does mention of trade names or commercial products con-
stitute endorsement or recommendation for use.
ii
-------
ABSTRACT
A state-of-the-art survey was made of the water pollution
problems which result from the production of inorganic fer-
tilizers and phosphate rock. Information required to com-
plete the study was obtained through an extensive literature
search, questionnaires sent to the major fertilizer pro-
ducers, and visits to selected production plants. Ninety
eight plants representing thirty three different companies
were surveyed. Production figures since 1940 and estimates
of production through 1980 were accumulated for phosphate
rock and the major fertilizer products. The specific pro-
duction operations which are the principal generators of
contaminated waste waters were identified, and the waste
water volumes and compositions for each operation were de-
termined wherever possible. The capability of current
technology to treat and control the contaminated waste
waters generated by the fertilizer industry was evaluated.
Problem areas where additional research and development
effort is needed to provide adequate control of waste water
discharge were identified.
This report was submitted in fulfillment of Grant Number
12020FPD under the partial sponsorship of the Water
Quality Office, Environmental Protection Agency.
iii
-------
TABLE OF CONTENTS
Page
PART I INORGANIC FERTILIZER INDUSTRY
LIST OF FIGURES iv
LIST OF TABLES vi
INTRODUCTION 1
SUMMARY 3
DESCRIPTION OF THE INORGANIC FERTILIZER INDUSTRY 9
NITROGEN FERTILIZER INDUSTRY 14
PHOSPHATE FERTILIZER INDUSTRY 22
POTASH INDUSTRY 31
WATER USE IN THE FERTILIZER INDUSTRY 39
FERTILIZER INDUSTRY PROCESSES AND OPERATIONS 45
NITROGEN FERTILIZER PRODUCTION PROCESSES 45
PHOSPHATE FERTILIZER PRODUCTION PROCESSES 61
POTASH PRODUCTION PROCESSES 80
AQUEOUS PROCESS EFFLUENTS 87
CLASSIFICATION OF AQUEOUS EFFLUENTS 87
AQUEOUS EFFLUENT VOLUMES AND COMPOSITIONS 91
ENVIRONMENTAL EFFECTS OF AQUEOUS EFFLUENTS 117
EFFECTS OF AQUEOUS EFFLUENTS ON RECEIVING
WATERS 117
PARAMETERS FOR EVALUATING AQUEOUS EFFLUENTS 119
ANALYSIS OF AQUEOUS EFFLUENTS 133
CONTROL AND TREATMENT OF AQUEOUS EFFLUENTS 135
CLASSIFICATION OF TREATMENT METHODS 135
TREATMENT OF NITROGEN FERTILIZER EFFLUENTS 140
TREATMENT OF PHOSPHATE INDUSTRY EFFLUENTS 142
POTASH INDUSTRY EFFLUENTS 148
ECONOMICS OF EFFLUENT CONTROL 149
RESEARCH AND DEVELOPMENT REQUIREMENTS FOR REQUIRED
CONTROL OF AQUEOUS EFFLUENTS 157
REFERENCES 161
-------
TABLE OF CONTENTS (Cont.)
PAGE
PART II PHOSPHATE ROCK MINING AND BENEPICIATION
INTRODUCTION 169
SUMMARY 171
THE PHOSPHATE ROCK INDUSTRY 173
THE NATURE OF PHSOPHATE MINERALS
AND DEPOSITS 174
MINING PHOSPHATE ROCK 175
BENEFICIATING PHOSPHATE ROCK 179
EFFLUENTS FROM MINING AND BENEFICIATION 191
ENVIRONMENTAL AND POLLUTION EFFECTS OF EFFLUENTS
FROM MINING AND BENEFICIATION 195
CONTROL AND TREATMENT OF AQUEOUS EFFLUENTS 197
ECONOMIC ASPECTS OF EFFLUENT CONTROL 199
TECHNICAL ASPECTS OF IMPROVED EFFLUENT
CONTROL 199
REVIEW OF PRIOR RESEARCH RELATED TO
PHOSPHATE SLIMES DISPOSAL 201
RECOMMENDATIONS FOR RESEARCH TO DEVELOP
IMPROVED MEANS FOR DISPOSAL OF EFFLUENTS
FROM PHOSPHATE MINING AND BENEFICIATION 204
COORDINATION OF RECOMMENDED RESEARCH 207
REFERENCES 209
LITERATURE NOT CITED 213
APPENDIX A
OUTLINE OF VARIOUS BASIC PROCESSES AND
VARIATIONS TRIED BY TVA FOR PARTIAL DEWATER-
ING AND SETTLING OF SLIME 217
APPENDIX B
OUTLINE OF DEWATERING, CHEMICAL AND PHYSICAL
BENEFICIATION OF FLORIDA PHOSPHATE SLIMES
BY USBM 220
APPENDIX C
IMPROVEMENT IN FILTRATION RATES EFFECTED
BY FLOCCULANTS 223
ACKNOWLEDGEMENTS 225
vi
-------
LIST OF FIGURES
No. Title
1 FERTILIZER CONSUMPTION IN UNITED STATES
2 SYNTHETIC AMMONIA PRODUCTION IN THE UNITED STATES 16
3 LOCATION OF MAJOR AMMONIA PLANTS IN THE
UNITED STATES 18
4 CONSUMPTION OF NITROGEN FERTILIZERS IN THE
UNITED STATES 20
5 LOCATION OF AMMONIUM NITRATE PLANTS IN THE
UNITED STATES 21
6 LOCATION OF MAJOR UREA PLANTS IN THE UNITED STATES23
7 FERTILIZER PRODUCTS OBTAINED FROM PHOSPHATE ROCK 25
8 CONSUMPTION OF PHOSPHATE FERTILIZERS IN THE
UNITED STATES 26
9 LOCATION OF MAJOR AMMONIUM PHOSPHATE PLANTS
IN THE UNITED STATES 29
10 PHOSPHORIC ACID PRODUCTION IN THE UNITED STATES 30
11 LOCATION OF MAJOR PHOSPHORIC ACID PLANTS IN THE
UNITED STATES 32
12 POTASH PRODUCTION AND CONSUMPTION IN THE
UNITED STATES 34
13 LOCATION OF POTASH PRODUCERS IN THE UNITED STATES 35
14 WATER USE BY THE FERTILIZER INDUSTRY(SIC 2871) 40
15 ESTIMATED WATER USE BY THE ENTIRE FERTILIZER
INDUSTRY 43
16 TYPICAL FLOWSHEET FOR AMMONIA PRODUCTION 49
17 TYPICAL FLOWSHEET FOR AMMONIUM SULFATE
PRODUCTION USING SYNTHETIC AMMONIA 52
18 FLOWSHEET FOR AMMONIUM NITRATE PRODUCTION 55
19 TYPICAL FLOWSHEET FOR UREA PRODUCTION - TOTAL
RECYCLE AND PRILLING TOWER 60
20 WET PROCESS PHOSPHORIC ACID FLOWSHEET 65
21 FLOWSHEET FOR PHOSPHORIC ACID PRODUCTION BY
THE ELECTRIC FURNACE PROCESS 69
22 NORMAL SUPERPHOSPHATE FLOWSHEET 72
23 FLOWSHEET FOR TRIPLE SUPERPHOSPHATE PRODUCTION 74
vii
-------
LIST OF FIGURES (Continued)
No. Title Page
24 FLOWSHEET FOR PRODUCTION OF DIAMMONIUM
PHOSPHATE 76
25 AMMONIATED SUPERPHOSPHATE FLOWSHEET 79
26 TYPICAL FLOWSHEET FOR RECOVERY OF POTASSIUM
CHLORIDE FROM SYLVINITE BY THE FLOTATION
BENEFICIATION PROCESS 81
27 TYPICAL SOLUTION-RECRYSTALLIZATION FLOWSHEET
FOR RECOVERING POTASSIUM CHLORIDE FROM
SYLVINITE ORE 83
28 FLUORIDE EVOLUTION DURING CALCINATION OF
PHOSPHATE ROCK 102
29 FLOWSHEET FOR LIME NEUTRALIZATION OF GYPSUM
POND WATER 144
30 TVA FLOWSHEET FOR TREATING PHOSSY WATER 147
31 TYPICAL GYPSUM POND CONSTRUCTION COSTS 151
32 CAPITAL AND OPERATING COSTS FOR OIL SEPARATORS 152
33 TYPICAL CONSTRUCTION COSTS FOR HOLDING AND
SETTLING PONDS 154
34 GEOGRAPHICAL LOCATION OF PHOSPHATE ROCK
OPERATIONS 176
35 GENERAL PLAN FOR CONVENTIONAL ROOM-AND-PILLAR
MINING 178
36 GENERALIZED FLOWSHEET OF A PHOSPHATE ROCK
MINING AND BENEFICIATION PLANT 180
37 DISTRIBUTION OF PRODUCTS DERIVED FROM LAND-
PEBBLE PHOSPHATE ROCK, PERCENT 181
38 GENERALIZED FLOWSHEET OF A WASHER PLANT FOR
RECOVERING FLORIDA PEBBLE PHOSPHATE 182
39 GENERALIZED FLOWSHEET OF A FLOTATION PLANT FOR
RECOVERING FLORIDA PHOSPHATE 183
40 GENERALIZED FLOWSHEET OF A WASHER PLANT FOR
RECOVERING NORTH CAROLINA PHOSPHATE 184
41 GENERALIZED FLOWSHEET OF A FLOTATION PLANT FOR
RECOVERING NORTH CAROLINA PHOSPHATE 185
42 FLOWSHEET OF A WASHING PLANT FOR TENNESSEE
BROWN-ROCK PHOSPHATE 186
Vlll
-------
LIST OF FIGURES (Continued)
No.
43
44
45
46
47
48
Title
TYPICAL FLOWSHEET OF A WESTERN PHOSPHATE ROCK
OPERATION
VIEWS OF A TYPICAL FLORIDA WASHER PLANT
VIEWS OF FLORIDA SLIME PONDS AND DIKES
VIEW OF WATER DRAINAGE FROM SLIME PONDS FOR
RECIRCULATION TO PLANT
VIEW OF A SLIME POND SHOWING GROWTH OF WATER
PLANTS
VIEWS OF AMERICAN CYANAMID ' S NEW STAGE-FILLING
Page
187
189
193
194
196
TECHNIQUE FOR SLIME DISPOSAL 198
-------
L I S T OF TABLES
NO. Title
1 PRINCIPAL STRAIGHT AND MIXED FERTILIZERS
PRODUCED IN THE UNITED STATES 12
2 AMMONIA CONSUMED IN UNITED STATES IN FERTILIZER
APPLICATIONS 17
3 WATER REQUIREMENTS FOR PRODUCTION OF VARIOUS
FERTILIZERS 41
4 PROBABLE, AQUEOUS WASTE STREAMS FROM FERTILIZER
PROCESSING 92
5 TYPICAL AMMONIA PLANT WASTE WATER VOLUME AND
COMPOSITION 94
6 COMPOSITION OF AMMONIA PLANT COOLING WATER
SLOWDOWN 95
7 COMPOSITION OF UREA PLANT BOILER WATER BLOWDOWN 98
8 TYPICAL UREA PLANT WASTE WATER VOLUME AND
COMPOSITION 99
9 COMPARISON OF AQUEOUS PROCESS WASTE STREAMS FROM
NITROGEN FERTILIZER PLANTS 100
10 TYPICAL EQUILIBRIUM COMPOSITION OF GYPSUM
POND WATER 103
11 FLUORIDE DISTRIBUTION IN WET PROCESS PHOSPHORIC
ACID PRODUCTION 105
12 TYPICAL PHOSSY WATER COMPOSITION AND VOLUME 108
13 COMPOSITION OF FURNACE GRADE PHOSPHORIC ACID
PLANT COOLING WATER BLOWDOWN 109
14 TYPICAL LIMITS FOR POLLUTION PARAMETERS 122
15 TEMPERATURE RANGES FOR FRESHWATER FISHES 127
16 TYPICAL COST DATA FOR CHROMATE REMOVAL FROM
COOLING WATER BLOWDOWN 153
x
-------
LIST OF TABLES (Continued)
No. Title
17 ESTIMATED PRODUCTION OF MARKETABLE PHOSPHATE
ROCK IN THE UNITED STATES 173
18 VARIATIONS IN PHOSPHATE ROCK COMPOSITIONS 174
19 PHOSPHATE ROCK MINES IN THE UNITED STATES 177
20 GENERALIZED REAGENT CONSUMPTION IN CONCENTRATING
PHOSPHATE BY FLOTATION 190
21 WATER DISTRIBUTION FOR A TYPICAL FLORIDA PLANT
OPERATION 191
22 APPROXIMATE MINERALOGIC AND CHEMICAL COMPOSITION
OF PHOSPHATE SLIME SOLIDS 192
23 CAPITAL AND OPERATING COSTS FOR SLIMES DISPOSAL 200
xi
-------
PART I INORGANIC FERTILIZER INDUSTRY
INTRODUCTION
This study was carried out for the Environmental Protec-
tion Agency by Battelle Memorial Institute under Grant
No. 12020 FPD. The basic purpose of the study is to
provide a state-of-the-art survey of water pollution
resulting from inorganic fertilizer manufacturing opera-
tions and phosphate mining. It is intended that the
study provide the following information relevant to
the fertilizer industry.
1. Determination of the types and magnitude of
water pollution which result from the produc-
tion of inorganic fertilizers.
2. Identification of the specific process opera-
tions which are the principal generators of
contaminated waste waters.
3. Evaluation of current technology for treating
and controlling the contaminated waste waters
generated by the fertilizer industry.
4. Identification of critical problem areas where
additional research and development efforts
are needed to provide adequate control of waste
water discharge.
The survey covers all phases of the inorganic fertilizer
industry including the mining of phosphate rock. The
production of phosphoric acid and urea were also surveyed
to provide a comprehensive study of the fertilizer industry,
Part I of this report covers the production of fertilizer
materials, while Part II covers the mining and beneficia-
tion of phosphate rock.
The literature has been reviewed to obtain pertinent
information. A great deal of information is available
in the literature on fertilizer production, production
processes, and plant operations. Unfortunately, the
information available in the literature on pollution
and its control in the fertilizer industry is relatively
limited. To obtain the information required to complete
-------
the study, questionnaires were submitted to the principal
fertilizer producers, and visits were made to a selected
number of fertilizer plants to discuss water pollution
and its control with plant operating personnel. Through
use of the questionnaires and plant visits, data were
accumulated from 41 different plants representing 33
different companies. Since most plants prepare more
than one fertilizer product the total number of produc-
tion processes surveyed totaled 98. The following list
summarizes the number of sources surveyed for each major
fertilizer product.
No. of Plants
Product Surveyed
Ammonia 16
Ammonium Nitrate 15
Ammonium Sulfate 3
Urea 14
Wet Process Phosphoric Acid 13
Furnace Phosphoric Acid 4
Normal Superphosphate 11
Triple Superphosphate 7
Ammonium Phosphates 11
Ammoniated Superphosphate 3
Potash 1
Much of the information on water pollution and its control
in the fertilizer industry which is discussed in subse-
quent sections of this report was obtained from the plant
survey. Information obtained from the literature is
referenced. Data obtained through the plant survey are
not identified as to source.
-------
SUMMARY
The purpose of the study covered in this report was to
provide a state-of-the-art survey of water pollution
resulting from inorganic fertilizer manufacturing operations
The survey covered all phases of the inorganic fertilizer
industry including phosphoric acid production and phosphate
rock mining and beneficiation. Information needed to
complete the study was obtained through a literature
review, questionnaires sent to major fertilizer producers,
and visits to selected fertilizer production operations
for discussions with operating personnel.
The fertilizer industry is undergoing continual change
and^water quality standards are also changing. Therefore,
it is recommended that this survey be updated at regular
intervals to provide an accurate profile of water pollu-
tion control in the fertilizer industry.
Fertilizer consumption in the United States is increasing
rapidly. In 1970 U.S. consumption of fertilizers was
approximately 36,000,000 short tons. It is estimated
that by 1980 fertilizer consumption will increase to
about 70,000,000 short tons annually. Nitrogen fertilizers
constitute the largest share of the fertilizer market,
and phosphate and potash fertilizers the bulk of the re-
maining market. The importance of nitrogen fertilizers
is expected to increase in the future. The types and
grades of fertilizers used are continually changing,
and the emphasis is on the production of higher nutrient
content materials. Production processes are undergoing
continual change to accommodate the changing market re-
quirements .
The^fertilizer industry consumes large volumes of water.
It is estimated that the gross water requirements of
the industry exceeded one trillion gallons in 1970. Water
recycle is widely practiced by the fertilizer industry
industry, and 75-80% of the gross water requirements
were supplied by recycled water. Recycling of water
should increase in the future, and the ultimate objectives
of many fertilizer producers is to achieve complete closed
loop water reuse. Of all the water used by the industry,
70-80% is for cooling purposes.
-------
A great many different types and grades of fertilizers
are produced in the United States. Many of these products
are prepared by the blending of the principal fertilizer
materials to give the desired compositions. The number
of different straight and mixed (nonblended) fertilizers
produced on a large scale are relatively few. They are:
Ammonia Phosphate Rock
Ammonium Nitrate Normal Superphosphate
Ammonium Sulfate Triple Superphosphate
Urea Ammonium Phosphates
Nitrogen Solutions Ammoniated Superphosphate
Potassium Chloride
Phosphoric acid, while not used directly as a fertilizer
to any extent, is an indispensable intermediate in phos-
phate fertilizer production.
Each of the major fertilizer materials can be produced
by two or more processes or process variations. It was
impossible within the limits of this study to evaluate
each process, and the possible variations, for its effect
on the environment. Instead, a typical or standard pro-
cess was defined for each major fertilizer product. A
typical flowsheet was prepared for each product and the
specific operations in the flowsheet which generate con-
taminated waste waters were identified. Using the typical
flowsheets and literature and plant survey data, typical
waste stream volumes and compositions were determined
for each of the major fertilizer products.
Waste water discharges from a fertilizer plant can be
broken down into four general types:
Cooling water
Steam condensate
Process Effluents
Sanitary wastes
The process effluents can be further subdivided into
five general classes:
By-product streams
Scrubber solutions
Process spills
Equipment cleaning (wash) solutions
Barometric condenser water
-------
One or more of these waste streams will usually be gen-
erated in any fertilizer production operation.
A review of the fertilizer industry reveals that inorganic
materials are the major pollutional problems encountered
in fertilizer plant waste water. Conventional pollutional
parameters such as BOD, COD, and coliform are relatively
unimportant because of the low organic content of most
fertilizer plant waste waters. The principal contaminants
found in most fertilizer plant waste waters are plant
nutrients such as nitrogen (as ammonia and nitrate) and
phosphates, and toxic materials such as fluoride, chromate,
and elemental phosphorus. Thermal pollution is also
a problem which must be considered. The only organic
materials which present a major disposal problem are
lubricating oils. Urea, monoethanolamine, and organic
biocides are present in certain waste waters but do not
present a major disposal problem at the present time.
On an overall basis nitrogen in nitrogen fertilizer plant
waste waters and phosphate and fluoride in phosphate
fertilizer plant waste waters present the major contam-
inants which must be controlled by the fertilizer industry.
Nonfertilizer additives such as pesticides and herbicides,
which may be added to certain fertilizer formulations,
were not considered in this study.
In the production of nitrogen fertilizers most plants cur-
rently rely on disposal by dilution for control of plant
effluents. Treatment of these process wastes is usually
restricted to the use of separators for oil removal and
settling ponds for solids removal and cooling. Chemical
treatment, using sulfur dioxide and lime, is usually pro-
vided for chromate removal from cooling water blowdown.
While disposal by dilution is currently acceptable in many
areas for disposing of nitrogen fertiliser plant waste
waters, it is not acceptable in some locations (and will
become less acceptable in other areas in the future).
Methods for nitrogen removal (principally ammonia and
nitrate) are needed by the industry to treat the waste
waters when disposal by dilution is inadequate or unaccept-
able. Gas stripping is used for ammonia removal in some
plants, but does not provide the degree of ammonia removal
which may be required. A study supported by the EPA
(Grant No. 12020EGM) is currently underway at the Tyner,
Tennessee plant of Farmers Chemical Association, Inc.,
to evaluate the use of ion exchange to remove ammonia
and nitrate from fertilizer plant waste waters.
-------
Phosphoric acid and phosphate fertilizer plant waste
waters normally require treatment for fluoride and phosphate
removal. When a gypsum pond is available, all contaminated
waste waters normally discharge to the gypsum pond. Treat-
ment of the pond water discharge by lime neutralization
reduces the fluoride and phosphate concentrations to
acceptable levels under current water quality standards.
When a gypsum pond is not available, other methods for
treating phosphate fertilizer plant waste water are needed.
Fluoride is the principal problem, and is usually handled
by recovering the fluoride as a salable by-product.
Data on the cost of providing adequate control and treat-
ment of fertilizer plant waste waters is relatively scarce.
Information obtained from the plant survey shows that
expenditures for treating nitrogen fertilizer plant waste
waters ranges from 0 to $0.78 per ton of product. These
costs reflect the comparatively low level or lack of
treatment provided these waste streams. The cost of
treating the waste waters from phosphoric acid and phos-
phate fertilizer plants ranges from $0.90-$3.00 per ton
of P2°e product and averages about $1.90 per ton of P2°5*
Several problem areas have been identified where additional
research and development effort is required to help the fer-
tilizer industry meet current and future water quality stan-
dards. The critical areas include:
1. The definition of realistic water quality standards.
2. The development and application of suitable processes
for ammonia removal from waste waters.
3. The development and application of processes for re-
moving other nitrogen compounds from waste waters.
4. Development of methods for utilizing by-product gypsum
from wet-process phosphoric acid production.
5. Development of an improved process for removal and
possible recovery of fluoride from waste waters.
6. Development of an improved process for removal and
possible recovery of phosphate from waste waters.
7. Determination of the mechanisms involved in phosphate
removal by lime neutralization.
-------
8. Development of reliable, continuous, monitoring methods
for principal contaminants found in fertilizer plant
waste waters.
Study in these critical areas will help provide the fertilizer
industry with the technology needed to maintain adequate control
over waste waters generated by the industry.
-------
DESCRIPTION OF THE INORGANIC FERTILIZER INDUSTRY
The fertilizer industry in the United States is an industry
undergoing continuing change and expansion. Production
and use of fertilizers have experienced tremendous growth
especially over the past thirty years. The types and
grades of fertilizers produced are continually changing.
Natural fertilizer materials have been almost completely
supplanted by manufactured products. New mixed fertilizers
are continually being introduced. New processes are
being developed to meet changing product requirements,
and the entire fertilizer industry is becoming much more
complex.
At least 13 elements, in addition to carbon, hydrogen,
and oxygen, have been definitely identified as essential
to plant growth. Most of these elements are required
in^relatively small amounts. Three elements, however,
(nitrogen, phosphorus, and potassium) are required in
large quantities to sustain good plant growth. Supplying
adequate quantities of these three primary nutrients
is the basic purpose of the fertilizer industry. Of
the three elements, nitrogen is required in the greatest
quantity and phosphorus in the smallest quantities. Some
authorities predict that, regardless of plant needs,
the three nutrients will eventually be used on a 1:1:1
basis.
Production and consumption of fertilizers to supply the
three primary nutrients has grown very rapidly. Figure
1 shows the consumption of fertilizer materials in the
United States since 1940. (1,2,3,4,5) Between 1940 and
1970 fertilizer consumption increased by over 400%. Based
on a number of estimates, it appears that fertilizer
consumption in the United States should increase by
about 100% between 1970 and 1980. (1,2,6,7,8) Nitrogen
consumption is expected to grow at the fastest rate and
phosphate consumption at the slowest rate. United States
production of fertilizer materials (except potash) nor-
mally exceeds consumption since exports usually exceed
imports. This trend is expected to continue in the future.
In the case of potash it is likely that consumption will
exceed domestic production for a number of years because
of the large-scale importation of Canadian potash.
-------
100
50
V)
z
o
o
31
in
u.
o
oo
SE
o
I
z
o
H*
»-
o.
=3
00
z
o
u
oc.
LU
M
10
0.5
TOTAL FERTILIZER MATERIAL
_L
1940
1950
1960
1970
1980
FIGURE 1 FERTILIZER CONSUMPTION IN THE UNITED STATES (l,2,3,U,5,6,7,8)
10
-------
Fertilizers are generally divided into two major types:
straight fertilizers and mixed fertilizers. Straight
fertilizers are those fertilizer materials which contain
only one major plant nutrient. The principal straight
fertilizers produced in the United States are listed
in Table 1.
Mixed fertilizers are those materials which contain two
or more primary nutrients. Mixed fertilizers can be
prepared by the simple mechanical blending of straight
fertilizers or by reacting the fertilizer materials chem-
ically to produce the desired fertilizer products. Typical
of the mixed fertilizers produced by chemical reactions
are the ammonium phosphates and ammoniated superphosphate.
The two methods can be combined to produce a three-nutrient
fertilizer. For example, in the production of ammoniated
superphosphate potassium chloride can be added to the
reaction process to give an N-P-K fertilizer. The potas-
sium chloride does not enter into chemical reactions
but simply blends into the reaction product.
In the case of straight nitrogen fertilizers, most of
the materials are consumed in direct application and
only about 32% are consumed in mixed fertilizers.^2^)
This is due primarily to the widespread use of anhydrous
ammonia, aqua ammonia and nitrogen solutions in direct
applications. The percentage of nitrogen consumed in
mixed fertilizers is not expected to change significantly
in the future.
The situation is reversed in the case of phosphorus and
potassium. At present about 86% of the phosphorus used
in fertilizer applications is in the form of mixed fer-
tilizers, and the percentage is expected to increase
in the future. ^'y' In the case of potassium, the per-
centage consumed in mixed fertilizers is about 78%fi.but
is expected to decrease somewhat in the future.(2'9'
The process of mechanically blending straight fertilizers
to produce mixed fertilizers has undergone a significant
change in recent years. Originally, the blending was
carried out by dry mixing of nongranular straight fertil-
izers. The products obtained were unsatisfactory for
a number of reasons, and this type of blending operation
has been almost completely replaced by other processes.
11
-------
TABLE 1
PRINCIPAL STRAIGHT AND MIXED FERTILIZERS PRODUCED
IN THE UNITED STATES
Nitrogen Fertilizers
Ammonia
Ammonium Nitrate
Ammonium Sulfate
Urea
Nitrogen Solutions
Phosphate Fertilizers
Phosphate Rock
Phosphoric Acid
Normal Superphosphate
Triple Superphosphate
Potash Fertilizers
Potassium Chloride
Mixed Fertilizers
Ammonium Phosphates
Ammoniated Superphosphate
*Does not include blended fertilizers and liquid mixes,
12
-------
With the advent of granular fertilizer products, the
so called "bulk blending" process has become extremely
important. This process involves simply the mechanical
mixing of granular fertilizer materials to give the desired
fertilizer composition. The materials normally used
in the bulk-blending operation are potassium chloride,
ammonium phosphates, triple superphosphate, ammonium
sulfate, ammonium nitrate, and urea. The use of granu-
lar materials eliminates many of the disadvantages of
the original dry mix process, and as a result the produc-
tion of bulk-blended fertilizers has increased very rapidly.
This can be seen by considering the rate at which/bulk-
blend plants have increased in the United States. '
Estimated Number
of Bulk-Blend
Year Plants in U.S.
466
1536
3152
The bulk-blend plants are usually small and serve only
a limited area. They are concentrated principally in
the major consumption areas of the midwest.
Another recent development is the increased use of liquid-
mixed fertilizers. They are prepared primarily by blending
aqueous solutions of various fertilizers to obtain the
desired composition. Some liquid mix plants dissolve
combinations of solid fertilizers to obtain the required
compositions. Like bulk-blend plants, liquid mix plants
are quite small and designed to serve limited areas.
The number of liquid mix plants in the United States
is significantly smaller than bulk-blend plants, but
still represents a sizable fertilizer output.^9'
Estimated Number of
Liquid Mix Plants
Year in U.S.
1960 380
1964 720
1966 1231
13
-------
In the past the production of nitrogen fertilizers, phosphate
fertilizers and potash were relatively separate operations.
The limited amounts of mixed fertilizers produced were
made primarily by shipping potash and nitrogen fertilizers
to mixed-fertilizer plants for blending with phosphate.
Production of mixed fertilizers such as ammoniated super-
phosphate and the ammonium phosphates was almost always
carried out at phosphoric acid-phosphate fertilizer plants.
Potash production is still completely separate from other
fertilizer production. However, the recent trend is
to combine nitrogen and mixed fertilizer production at
large nitrogen fertilizer complexes.
In certain instances nonfertilizer materials, such as
herbicides or pesticides, may be added to fertilizer formu-
lations. Additions of these materials to the fertilizers
and their possible entry into fertilizer plant waste
waters was not covered in this study.
NITROGEN FERTILIZER INDUSTRY
Nitrogen is required in larger quantities for good
plant growth than any other nutrient. Since 1957 fertilizer
nitrogen consumption in the United States has exceeded
both phosphate and potash on a tonnage basis. The princi-
pal sources of fertilizer nitrogen have changed drastically
during this century. Prior to 1900 almost all fertilizer
nitrogen used in the United States came from natural
organic materials. Between 1900 and 1920 natural nitrate
deposits were an important source of fertilizer nitrogen,
and for many years by-product ammonia from coke-oven
gas was an important factor in the fertilizer industry.
With the development of the Haber-Bosch process for ammonia
production, atmospheric nitrogen became the principal
source of fertilizer nitrogen. Slack(9) ranks the various
sources of fertilizer nitrogen in the United States at
the current time as follows:
Percent of Fertilizer
Source Nitrogen Supplied
Natural Organics 1
Sodium Nitrate 2
Coal 3
Atmospheric Nitrogen 94
14
-------
The economic development of the Haber-Bosch process for pro-
duction of ammonia from nitrogen and hydrogen has resulted
in ammonia becoming the most important fertilizer material
in the United States. Not only is ammonia used on a very
large scale as a fertilizer, but it also serves as a raw
material for the preparation of almost all other nitrogen
containing fertilizers produced in the United States. The
tremendous growth in synthetic ammonia production since
1940 is shown in Figure 2.
Estimates of future production indicate that ammonia should
continue to experience the same rapid growth.(2,10) The
importance of ammonia to the fertilizer industry can be
seen by referring to Table 2 which lists the quantities
of ammonia used in various fertilizer processes in the
United States.
Total ammonia consumption in the United States in 1968
was estimated to be 12/032,000 short tons.dO) of this
amount, 80.6% was used in fertilizer applications. The
rest was used principally in urea, nitric acid, and ammonium
nitrate production for nonfertilizer applications.
Location of ammonia plants is dependent on two factors,
availability of raw materials and accessibility to the
market. Of the two, a raw material supply is the most
important*- Since atmospheric nitrogen is readily available
anywhere, a source of hydrogen is of prime importance in
ammonia production. In the United States natural gas and
petroleum fractions serve as the principal sources of hydro-
gen. As a result ammonia plants are located most heavily
in those areas where natural gas and petroleum fractions
are readily available. Louisiana, Texas, California, Iowa,
Mississippi, and Arkansas are currently the major ammonia
producing states.<2> The state of Iowa is also the largest
consumer of fertilizer nitrogen. In total, there are approxi-
mately 70 companies producing ammonia in plants located
in 30 different states. The location of the major ammonia
plants are shown in Figure 3. It is estimated that the
total ammonia production capacity of the United States ,...
as of January 1971, was 18,700,000 short tons per year.* J
An interesting facet of ammonia production has been the
trend to very large capacity plants. In 1955, 300 tons
per day was considered to be a large plant. Today 1000
tons-per-day plants are not uncommon and plants as large
as 1500 tons-per-day have been built. The economic advantage
of very large plants is slowly forcing the shut-down of
15
-------
50
o
5 10
u_
o
t/1
z
o
a
o
on
a.
0.5
CAPACITY
PRODUCTION
1940
1950
1960
1970
1980
FIGURE 2. SYNTHETIC AMMONIA PRODUCTION IN THE UNITED STATES (1,2,3,^,5,10)
16
-------
TABLE 2
AMMONIA CONSUMED IN UNITED STATES IN FERTILIZER APPLICATIONS
NH-7 Used (thousands of Short Tons)
Fertilizer Product
Direct Application NH,
Ammonium Nitrate
Ammonium Sulfate
Ammonium Phosphates
Nitrogen Solutions and
Mixed Fertilizers
Urea
Other Fertilizers
TOTAL
1960
837
1252
388
228
432
309
n
3477
1965
1877
1838
707
705
428
581
99
6235
1968
3375
2164
751
1425
560
1285
136
9696
17
-------
KENAI, ALASKA
I \
FIGURE 3 LOCATION OF MAJOR AMMONIA PLANTS IN THE UNITED STATES
-------
many older, smaller plants. The high nitrogen content
of ammonia and improved shipping procedures makes it eco-
nomical to produce ammonia in large plants and ship for
substantial distances to the markets.
As stated previously, ammonia serves as a raw material
in the production of almost all nitrogen fertili2ers. On
a tonnage basis the most important nitrogen fertilizers,
in addition to ammonia, are ammonium nitrate, ammonium
sulfate, and urea. United States consumption of these
materials in fertilizer applications is shown in Figure
4. The data presented are for total consumption of each
material and include the tonnages used in direct applica-
tion and in the preparation of mixed fertilizers.
Ammonium nitrate ranks second only to ammonia as a source
of fertilizer nitrogen. During the 1950's ammonium nitrate
was the principal source of fertilizer nitrogen but has
now been displaced by ammonia. Long range predictions
show that ammonium nitrate should continue to be a major
source of fertilizer nitrogen.<2f11) Essentially all ammon-
ium nitrate production is based on the reaction of ammonia
and_nitric acid. As a result, ammonium nitrate production
facilities are almost always located adjacent to ammonia
plants. More than 40 companies in the United States pro-
duce ammonium nitrate, and current production capacity
is estimated to be approximately 9,000,000 short tons per
year. (J- ' The locations of major ammonium nitrate plants
are shown in Figure 5. More than 85% of the ammonium nitrate
produced is consumed in fertilizer applications. Most
of the remainder is used in explosives manufacture.
For many years ammonium sulfate was a major source of fer-
tilizer nitrogen in the United States. It still is used
in substantial quantities, but production has remained
relatively constant for the past 15 years. It is expected
that production will remain at approximately the current
level in the future. Almost all the ammonium sulfate pro-
duced is used in fertilizer applications.
Ammonium sulfate production in the U.S. is based principally
on the reaction of ammonia and sulfuric acid. However,
by-product sources of ammonia (such as coke oven gas) and
sulfuric acid (spent refinery acid) account for a large
portion of the ammonium sulfate production. As a result
of the use of by-product acid and ammonia there are a large
number of relatively small plants producing ammonium sulfate.
Many of these are operated by steel companies to utilize
the ammonia from coke oven gas. Current production capacity
from all sources is estimated to be 3,000,000 short tons of
ammonium sulfate per year.'11)
19
-------
10
o:
LU
Q.
CO
z
o
a:
o
u.
o
CO
z;
o
o
i—i
i-
a.
tst
z
o
o
~ 0.1
_L
I
TOTAL N
_L
1940
1950
1960
1970
1980
FIGURE 4 CONSUMPTION OF NITROGEN FERTILIZERS IN THE UNITED STATES
20
-------
FIGURE 5 LOCATION OF AMMONIUM NITRATE PLANTS IN THE UNITED STATES
-------
The use of urea in fertilizer application is a very recent
development. In 1957 only 2% of United States fertilizer
nitrogen requirements were supplied by urea. It is estimated
that in 1971 this will increase to about 11-12%. Urea
production in the United States has increased at an average
rate of about 17% a year since 1950. At the present time
approximately 75% of the urea produced goes into fertili2er
applications, while livestock feeds and urea-formaldehyde
resins account for most of the remainder. It is expected
that urea consumption in fertilizer applications will continue
to grow at a rapid rate in the future, exceeding even the
growth rate expected of direct application ammonia.
Principal reasons for the increasing popularity of urea
are its high nitrogen content and the absence of hazards
in handling and shipping. It has the highest nitrogen
content (45%) of any solid fertilizer. These advantages
are sufficient to offset the relatively high cost of urea.
If urea costs decrease in the future as expected, then
it will become even more popular.
All commercial production of urea in the United States
is by the reaction of ammonia and carbon dioxide. As a
result, urea is almost always produced at an ammonia plant
where a supply of ammonia and carbon dioxide is available
(carbon dioxide is a by-product of hydrogen production
in most ammonia plants). This availability of essentially
free carbon dioxide is the principal reason urea production
costs are as low as they are. At the present time there
are about 35 companies producing urea in the United States.
Total production capacity in 1970 was approximately 4,600,000
short tons of urea per year. Production capacities
the larger urea plants range up to 1600 tons per day.
Figure 6 gives the location of the major urea plants in
the United States.
PHOSPHATE FERTILIZER INDUSTRY
Up until the middle 1950's phosphate was the major fertil- '
izer nutrient consumed in the United States. Since that
time nitrogen consumption has become predominant, but phos-
phates are still used on a very large scale.
The principal source of all phosphate consumed in the United
States is phosphate rock. In general phosphate rock has
a chemical composition which can be represented by the
formula 3Ca3(PO.)2*CaX2 where x may be a halogen, hydroxyl
ion, etc. Fluorapatite (3Ca3[PO4]2*CaF9) is the most pre-
valent form, and all U.S. phosphate rock contains fluorine
in substantial amounts (3-4%) .
22
-------
KENAI, ALASKA
U)
FIGURE 6 LOCATION OF MAJOR UREA PLANTS IN THE UNITED STATES (2)
-------
The phosphate in the rock is almost completely insoluble
in soil solutions. In order to be effective in furthering
plant growth the phosphate must be soluble to some degree.
The basic objective of the phosphate fertilizer industry
is to convert the phosphate in the rock to a form that
is soluble. The degree of solubility needed is dependent
on a number of factors such as soil conditions and degree
of fertilization. The water solubility of phosphate fer-
tilizers can vary over a very wide range: from very low
solubility for ground phosphate rock to complete solubility.
The general trend in the United States is to higher solubility
products.
Phosphate rock can be converted to a variety of products
for fertilizer applications. Figure 7 shows a schematic
diagram of the major fertilizer products obtained from
rock. The relative importance of the various products
is continually changing and mixed fertilizer—both liquid
and solid—are becoming increasingly important. As a result,
the direct application of straight phosphate fertilizers
is decreasing.
Phosphoric acid, while not consumed directly in any quantity
in fertilizer applications, is an indispensable intermediate
in phosphate fertilizer production.
The principal straight phosphate fertilizers are triple
superphosphate, normal superphosphate and phosphate rock.
The major mixed phosphate fertilizers are monoammonium
and diammonium phosphates which are normally considered
as a single product in presenting production and consumption
data. Ammoniated superphosphate is also an important mixed
phosphate fertilizer but has no set composition. The com-
position can be varied by controlling the ammonia-superphosphate
ratio. Figure 8 shows the consumption in the United States
since 1940 of the principal phosphate fertilizers. '''
Estimates of consumption through 1980 are also included.f1'2'8)
The data presented illustrates very emphatically the strong
trend away from low analysis fertilizers such as phosphate
rock and normal superphosphates.
Normal superphosphate, which was the dominant source of
phosphate for many years, has decreased in consumption
since about 1950. Normal superphosphate is still used
on a fairly large scale, however, and should continue to
be used at a slowly declining but still substantial rate
for the next 20 years. All normal superphosphate is pro-
duced by the reaction of sulfuric acid and phosphate rock
24
-------
i
NORMAL
SUPERPHOSPHATE
ro
TRIPLE
SUPERPHOSPHATE
NH.
1
AMMONIATED
SUPERPHOSPHATE
SOLID MIXED
FERTILIZERS
LIQUID MIXED
FERTILIZERS
PHOSPHATE FERTILIZER PRODUCTS
FIGURE 7 FERTILIZER PRODUCTS OBTAINED FROM PHOSPHATE ROCK
-------
10
a:
-------
to give monocalcium phosphate and calcium sulfate. Pro-
duction plants are relatively small and located at the
points of consumption. The low analysis (16-20% P20s)
of normal superphosphate makes it cheaper to ship phosphate
rock to consumption areas rather than shipping the normal
superphosphate from the ore deposits. There are currently
about 150 plants producing normal superphosphate in 30
different states. Production capacity in 1969 was estimated
to be 4,600,000 short tons normal superphosphate per year,
and is expected to drop to 4,500,000 short ton per year
in 1974. CI1)
Triple superphosphate consumption in the United States
has undergone a very rapid growth during the past 30 years.
At the present time triple superphosphate is the second
leading source of fertilizer phosphate. It is produced
by reacting phosphate rock with phosphoric acid to give
monocalcium phosphate. Since no calcium sulfate is formed,
the P205 content of triple superphosphate ranges from 44-
47%, i.e., about three times that of normal superphosphate.
As a result of its high P20s content, triple superphosphate
is usually produced near the phosphate rock deposits and
shipped to the market areas for direct application or for
use in preparation of mixed fertilizers. There are currently
about 20 companies with plants in 10 states producing triple
superphosphate. Current productions capacity is estimated
to be about 5,800,000 short tons of triple superphosphate
per year. About 75% of the production capacity is located
in the state of Florida.
The ammonium phosphates have experienced an extremely rapid
growth over the past 20 years; they are currently the major
source of fertilizer phosphate in the United States. They
will probably become even more important in the future.
Diammonium phosphate is the preferred form. The ammonium
phosphates are produced by reacting phosphoric acid with
ammonia. Either wet-process or furnace grade acid can
be used, and the product composition can be varied over
a wide range by varying the ratio of reactants. Ammonium
phosphates are widely used in liquid mixed fertilizers
and in combination with other solid fertilizers. It is
common practice in many plants to add other fertilizer
materials to the phosphoric acid — ammonia reaction to produce
solid products of various compositions. Materials which
may be added include potassium chloride, sulfuric acid,
ammonium nitrate, nitric acid, and urea.
27
-------
There are more than 40 companies in the United States that
produce ammonium phosphates. The locations of the major
plants are shown in Figure 9. Current production capacity
is approximately 3,800,000 short tons of ammonium phosphates
per year. (H)
Ammoniated superphosphate is a major source of fertilizer
phosphate in the United States. The name is rather a mis-
nomer since the product usually contains potassium, other
nitrogen materials such as ammonium sulfate, and possibly
diammonium phosphate. Ammoniated superphosphate is normally
produced by reacting normal or triple superphosphate with
an ammoniating solution to form calcium phosphate and mono-
ammonium phosphate. The ammoniating solution is an aqueous
solution of ammonia/ ammonium nitrate, and possibly urea.
The solution composition can vary over a wide range depend-
ing on the product desired. Current practice favors pro-
duction of a granular material. As a result, ainmoniated
superphosphate production has been closely related to the
development of granulation techniques. Many different
granular fertilizer formulations can be prepared, based
on the ammoniation of superphosphate, by adding other fer-
tilizer materials to the reaction mixture. The materials
that may be added include anhydrous ammonia, ammonium sul-
fate, diammonium phosphates, phosphoric acid, sulfuric
acid, and potassium chloride.
With the increased use of high analysis fertilizers, phos-
phoric acid becomes increasingly important in the phosphate
fertilizer industry. Phosphoric acid is produced by two
basic processes. In the wet process, phosphate rock is
reacted with a strong mineral acid (usually sulfuric) to
form orthophosphoric acid. In the furnace process phosphate
rock is converted to elemental phosphorous which is burned
to ?20s and then reacted with water to form the acid. Both
wet-process and furnace grade phosphoric acid are used
in the fertilizer industry. However, the tonnages of furnac
grade acid consumed are relatively small. Figure 10 shows
the production of both types of acid in the United States
since 1940. Most furnace grade acid is used in nonfertilize
applications, but its use in the fertilizer industry should
increase in the future. Almost all of the wet-process
phosphoric acid produced in the United States is used by
the fertilizer industry. Total phosphoric acid production
should experience the same growth in the future that is
predicted for the phosphate fertilizer industry. Because
of the uncertainty of the future of phosphates in the de-
tergent industry it is difficult to predict the future
of furnace grade acid. However, it will probably continue.
to grow at a modest rate.
28
-------
to
I \
FIGURE 9 LOCATION OF MAJOR AMMONIUM PHOSPHATE PLANTS IN THE UNITED STATES (11)
-------
in
o
CM
o.
co
O
o
:c
CO
u_
o
CO
o
-: i
0.5
a
o
Of.
a.
o.
co
o
0.1
TOTAL
PHOSPHORIC ACID
WET PROCESS ACID
FURNACE ACID
1940
1950
1960
1970
1980
FIGURE 10 PHOSPHORIC ACID PRODUCTION IN THE UNITED STATES
30
-------
At present there are about 35 companies producing wet process
phosphoric acid in the United States and 8 companies (including
TVA) producing furnace acid. Figure 11 shows the location of
the major phosphoric acid plants in the United States. Wet
process acid production capacity in 1969 was estimated to be
5,800,000 short tons P2O5 per year; furnace grade acid capacity
was 1,700,000 short tons P2°5 Per vear- ^^
POTASH INDUSTRY
Potassium is one of the three primary nutrients required for
plant growth. The functions of potassium in the plant growth
cycle are not well defined. However, it is required in rela-
tively large amounts to assure healthy plant growth. In most
cases plant requirements for potassium exceed those for phos-rqv
phorus and sometimes approach the requirements for nitrogen.' '
Thus, potassium is used in large quantities in fertilizer appli-
cations throughout the world.
In the potassium (and fertilizer) industry the term "potash"
is normally used to refer to any potassium compound which is
used for its potassium content. Regardless of the source of
the potassium, it is fertilizer industry practice to state the
potassium content of the compound in terms of its equivalent
potassium oxide (K20) content. Pure potassium chloride, for
example, has an equivalent K2O content of 63.17%. The K2O
content of other common potassium compounds is given below.
Pure Compound Equivalent K^O Content (%)
KN03 46.6
K2S04 54.1
K2C03 68.2
KHC03 47.0
KOH 83.9
KOH (45% Solution) 37.8
KC1 63.2
31
-------
CJ
ro
• WET PROCESS PHOSPHORIC ACID
A FURNACE PHOSPHORIC ACID
FIGURE 11 LOCATION OF MAJOR PHOSPHORIC ACID PLANTS IN THE UNITED STATES (11)
-------
More than 90% of the potash consumed in the United States
each year goes into fertilizer applications. As a result,
the potash industry has experienced the same rapid growth
over the past 30 years which other segments of the fertilizer
industry have known. Data presented in Figure 12 illustrate
the rapid growth of potash production and consumption in
the United States since 1940. The relatively depressed state
of the U.S. fertilizer industry has held potash production
and consumption relatively constant since 1967. However,
long range projections show that potash consumption (primar-
ily in the fertilizer industry) should grow at a substantial
rate between 1970 and 1980. (2; The rate of growth will prob-
ably be somewhere between the rates experienced by the nitro-
gen and phosphorus based fertilizers.
Potassium chloride is the major source of the fertilizer
potassium consumed in the United States. It normally accounts
for 90-95% of the potash used in fertilizer applications.
The popularity of KC1 is the result of its ready availability,
low price, and high K20 content. Potassium sulfate supplies
most of the remaining potash requirements of the fertilizer
industry (5-6%). Potassium nitrate, potassium hydroxide,
potassium magnesium sulfate, and the potassium phosphates
are also used in fertilizer applications, but only to a very
limited extent.
Direct application of fertilizer potash is relatively limited,
and accounts for only about 22% of the potash used in the
U.S. in fertilizer applications.^/9) The bulk of the potash
is used in either blended solid fertilizers or liquid fertil-
izers. The direct application of potash is growing, however,
since plant requirements for potash can be satisfied by rel-
atively infrequent applications.
The major sources of potash in the United States are the
sylvinite deposits in New Mexico and Utah and the natural
brines at Searles Lake, California, and Bonneville, Utah.
Sylvinite, a physical mixture of halite (NaCl) and Sylvite
(KC1), accounts for about 90% of U.S. potash production.
There are currently 10 major potash producers in the United
States, seven of which operate in the Carlsbad area of New
Mexico (see Figure 13). Production capacity of the 10 pro-
ducers is reported to be approximately 3.8 million short
tons K2O/year.(2) The New Mexico operations produce the
bulk of U.S. potash output (approximately 80%). The high
grade ores in New Mexico are being slowly depleted, however,
and there is considerable interest developing in the sylvi-
nite deposits in eastern Utah. In addition, Canadian produc-
tion of potash will have a significant effect on future potash
developments in the United States.
33
-------
50
10
oc
o
CO
u_
o
to
H-
o
0.5 —
CONSUMPTION
PRODUCTION
1940
1950
1960
1970
1980
FIGURE 12 POTASH PRODUCTION AND CONSUMPTION IN THE UNITED STATES
34
-------
Uo
FIGURE 13 LOCATION OF POTASH PRODUCERS IN THE UNITED STATES
-------
With current annual U.S. consumption of potash being approxi-
mately 3 million short tons of ^O, there is considerable
overcapacity in the United States. A recent Bureau of Mines
reportd3) estimates that the domestic demand for potash
will not catch up with the overcapacity until 1974 and possi-
ble much later. The Bureau also estimates that Canadian
producers will supply the bulk of U.S. potash requirements
for the next few years.
Recovery of potassium chloride from sylvinite is a relatively
simple operation. The ore is mined and the KC1 is separated
from the NaCl and gangue material. No major chemical opera-
tions are involved. Potassium chloride production from
sylvinite is, therefore, essentially a mining industry rather
than a chemical processing industry.
Underground deposits of sylvinite can be mined in two ways:
1. By traditional shaft mining techniques in which
a shaft is sunk to the ore bed and crushed ore
hauled to the surface.
2. By solution mining in which the potassium chloride
is dissolved in-situ and raised to the surface
in solution form. At the present time only
shaft mining for potash recovery from sylvinite
is practiced in the United States. There is
talk in the industry, however, that at least
one U.S. producer is considering solution
mining.
Various grades of KC1 are used in the potash industry. Grad-
ing is based on purity and particle size. Current purity
designations(2) are as follows:
Grade Purity Major End Use
fertilizer 95-97% KCl Solid Fertilizers
soluble 99% KCl Liquid Fertilizers
chemical 99.95% KCl Industrial Chemicals
The use of KCl in blended solid fertilizers has placed certain
restrictions on particle size and most producers supply
the various grades of KCl in different size ranges. Typically,
a specific grade of KCl would be available in the following
size grades.(2J
36
-------
Grade Size Range
standard 100% - 20 mesh
95% + 65 mesh
coarse 100% - 14 mesh
95% + 28 mesh
granular 100% - 6 mesh
98% + 20 mesh
Two basic types of processes are being used by U.S. producers
to recover KC1 from sylvinit'e ore: flotation beneficiation,
and solution-recrystallization. Flotation beneficiation
produces a fertilizer-grade KC1 for use in solid fertilizers.
Solution-recrystallization normally produces a soluble-
grade KC1 for use in liquid fertilizers.
In the typical flotation process sylvinite ore is crushed,
then pulped with saturated brine of KCl and NaCl, clay
slimes and insoluble ore removed, and then the KC1 and NaCl
are separated by flotation. The KC1 produced contains NaCl
and small amounts of clay and iron. About 65-70% of U.S.
sylvinite production is processed by flotation.
The solution-recrystallization process is based on the facts
that KCl is much more soluble in hot water than in cold,
while NaCl is only slightly more soluble in hot water than
cold. Typically, ground sylvinite ore is treated with a
heated recycle KCl-NaCl brine. The KCl goes into solution
together with a small amount of NaCl. Solids are removed
from the solution and then it goes to crystallizers where
it is cooled to drop out the KCl. The KCl is recovered
by filtration and then washed and dried. The brine solution
is recycled, after heating, back to the dissolution operation.
Gravity separation can also be used to produce a crude grade
of KCl from sylvinite. In this process the ore is first
crushed and screened. The screened ore is then mixed with
a saturated solution of NaCl and KCl and sent to a bank
of Wilfley tables where the KCl and NaCl crystals are separated
by gravity difference. The KCl product contains approximately
50% K2O equivalent.
Recovery of KCl from natural brines is a much more complex
process than recovery from sylvinite. The Searles Lake
brine is a complex mixture of sodium and potassium salts.
The salt bed is divided into two layers and the typical
composition of brine from each layer is given below.d4)
37
-------
Major Components Upper Layer Lower Layer
KC1 5.02% 2.94%
NaCl 16.06 15.51
Na2S04 6.75 6.56
Na2CO3 4.80 6.78
1.63 1.90
KBr 0.12 0.08
The potassium is recovered primarily from the upper layer
and involves a large number of carefully controlled operations
to produce a high purity KC1 product.
The Bonneville (Salduro Marsh) brines are a complex mixture
of metal chlorides and sulfates. A typical brine composition
is shown below
Component Percent
NaCl 18.0 - 24.0
KC1 0.8 - 1.2
MgCl2 0.9 - 1.2
MgS04 0.2 - 0.3
CaSO4 0.3 - 0.4
LiCl 0.03- 0.04
The brine is first concentrated in ponds by solar evaporation
until it is saturated with KCl. The brine is then pumped
to a second series of ponds where it is evaporated to deposit
a mixture of KCl and NaCl. Concentration is continued until
carnallite (KCl*MgCl2) begins to precipitate. The brine
is then removed from the ponds and the KCl-NaCl mixture
collected and sent to the refinery. At the refinery the
KCl-NaCl mixture is separated by flotation beneficiation
to give a fertilizer grade KCl.
Because of the widespread use of KCl in different types
and grades of solid mixed fertilizers, there has been an
increasing demand for KCl of larger particles than is readily
obtained from the normal production methods. This has led
to the development of methods for increasing the particle
size of fertilizer grade KCl. Both compaction and fusion
38
-------
techniques are being used. In the fusion process finely
divided KC1 is melted and then solidified in thin layers.
The layers are crushed and screened to give the desired
particle size. Undersize particles are recycled. In the
compaction process (14> the finely divided KC1 is compacted
into thin sheets using high pressure rolls at 200-250 °F.
The KC1 sheet is then flaked, crushed, and screened.
WATER USE IN THE FERTILIZER INDUSTRY
There is very little information in the literature on the
overall water requirements of the entire fertilizer industry.
There are references to water consumption for specific plants
or processes but none that cover the entire industry. Census
of Manufacturers publications by the United States Department
of Commerce1^-5) have reported water use patterns for SIC
2871 (Fertilizers). However, this data does not include
water used in the production of ammonia, urea, phosphoric
acid and phosphate rock, all of which are large consumers
of water. The Census of Manufacturers data for SIC 2871
are plotted in Figure 14. Figures on water use for SIC
28 (Chemicals and Allied Products) which includes all of
the fertilizer materials of interest except phosphate rock,
are also included. The data show that water use in the
fertilizer industry is growing at a faster rate than for
the chemical industry as a whole. This reflects the rapid
growth of fertilizer production since 1954.
In the present study a total of 33 fertilizer producers
throughout the country have supplied information on the
water required in the production of various fertilizer materials
and intermediates. Data on both gross and intake water
requirements were obtained. Table 3 summarizes the data
on gross water requirements. There is a considerable range
in water requirements for production of a given material.
This is due to a number of factors including processing
alternatives, water availability, and air pollution control
which can require considerable water scrubbing of gas streams.
For example, in the production of materials such as the
superphosphates, ammonium phosphates, ammoniated superphos-
phate, and granular N-P-K fertilizer water requirements
were reported to vary from 0-15,000 gallons per ton of pro-
duct. This range is due primarily to varying requirements
for water scrubbing of gaseous effluents. In urea production
the wide range in water requirements reflects the type of
process used -whether once-through, partial recycle or total
recycle.
39
-------
104
LU
Oi
LU
Q.
OO
O
_i 3
LU
O
oo
z:
o
CD
1
S 102
ID
cc.
LJJ
1—
3
i n
1 U
I I I i I
«•
~ ^~~~~ ~— I
GROSS WATER USED-^>^^ x - * -
x^ " — —INTAKE WATER
/
x~-x /
^ N/ _
s
s
s -
o''
s^ +"*-*
r. R n *\ 'i uiATrr? n*;Fn ^ .S •* x ^ v ^*
>^ ox
_^**"^ >^ TWTAIfF UATFD
^^^^ ^^ ^ 1 H 1 rt i\ c. W M 1 L K — —
- o*^ '^~Q
Q^^^^ ° SIC 2871 (FERTILIZERS)
X SIC 28 (CHEMICALS AND -
ALLIED PRODUCTS)
i i I i i
1954 1958 1962 1966 1970
FIGURE 14 WATER USE BY THE FERTILIZER INDUSTRY - SIC 2871 (15)
40
-------
TABLE 3
WATER REQUIREMENTS FOR PRODUCTION OF VARIOUS FERTILIZERS
Fertilizer Product
Ammonia
Ammonium Nitrate
Ammonium Sulfate
Urea
Phosphoric Acid
(wet process)
Phosphoric Acid
(furnace)
Normal Superphosphate
Triple Superphosphate
Ammonium Phosphate
Ammoniated Superphosphate
(Granular N-P-K Fertilizers)
Potassium Chloride
Gross Water Requirements
56,000-188,000 gal/ton NH3
3,000-35,000 gal/ton NH4NO3
100-10,000 gal/ton (NH4)2SO4
20,000-90,000 gal/ton urea
20,000-40,000 gal/ton P2O5
32,000-50,000 gal/ton P2O5
0-3,000 gal/ton Product
0-5,000 gal/ton Product
0-15,000 gal/ton Product
0-5,000 gal/ton Product
20,000-40,000 gal/ton KC1
*Information supplied by numerous fertilizer producers through-
out the United States.
41
-------
Using the data presented in Table 3 and knowing the yearly
production rates for each fertilizer material, one can obtain
a rough estimate of water use by the fertilizer industry
(water used in fertilizer blending plants and in production
of minor products is quite small and should not contribute
significantly to overall water consumption by the industry).
Such an estimate has been made, using an average water con-
sumption figure for each product, and the results are shown
in Figure 15. Estimated water consumption beyond 1970 is
based on estimated fertilizer production and current water
use rates. The estimates of water consumption are signifi-
cantly higher, over corresponding periods, than those reported
by the Census of Manufacturers. This reflects the large
volume of water used in ammonia, urea, phosphoric acid and
elemental phosphorus production.
Water reuse patterns vary significnatly from material to
material and from plant to plant. Some companies reported
essentially once-through use of water, while others reported
complete recycle and entirely closed loop operation. At
the current time, based on information from the 33 companies,
water recycle accounts for 75-80% of the gross water requirements
of the fertilizer industry. As water pollution control
standards become more stringent, water recycle should increase.
This is substantiated by Census of Manufacturers data which
show that water reuse for SIC 2871 (fertilizers) increased
from 40 to 48% between 1958 and 1964. This compares with
estimates of 75-80% water recycle for 1970 for the entire
fertilizer industry.
The major portion of the water used by the fertilizer indus-
try is for cooling purposes. In 1964 it was reported that
53% of the water used was for cooling. However, based on
information obtained in the current study cooling water
accounted for between 70-80% of all water usage.
42
-------
10'
cc
UJ
2 io3
CO
o
UJ
(ft
=3
Oi
UJ
10
SIC 28 (CHEMICALS AND ALLIED x'
PROD.) x'
FERTILIZER INDUSTRY x'
x
*-.
2871 (FERTILIZERS)
1954 1958 1962 1966 1970 1974
FIGURE 15 ESTIMATED WATER USE BY THE ENTIRE FERTILIZER INDUSTRY
(Based on information obtained in current study)
43
-------
FERTILIZER INDUSTRY PROCESSES AND OPERATIONS
To be able to adequately understand the problems of aqueous
effluent generation/ control, and treatment in the fertilizer
industry a basic knowledge of the major production processes
is required. A great many different types and grades of
fertilizers are produced in the United States. Many of these
are prepared simply by blending (in bulk blending and liquid
blending plants) straight and mixed fertilizers to obtain
the desired fertilizer compositions. Blending plant operations
are quite simple and usually involve only the mixing of
the various fertilizer materials. The problems of controlling
aqueous effluents in blending plants are, therefore, minimal
compared to those encountered in the production of straight
and mixed (nonblended) fertilizers.
The number of different straight and mixed (nonblended)
fertilizers produced on a large scale in the United States
are relatively few (see Table 1). However, each of these
materials are produced by at least two different processes
or process modifications. In most cases the operations
involved in preparing the various fertilizer materials are
substantially different. Because of the great number of
different processes and operations involved it was impossible,
within the scope of this study, to evaluate each individual
operation and process for its effect on the environment.
What has been done was to define, as completely as possible
for each major fertilizer product, a standard or typical
production process. The typical process contains all of
the processing operations which can contribute to the gen-
eration of aqueous effluents. Assuming the process as defined
is typical of U.S. production methods for a given fertilizer
material, the aqueous effluents generated by the process
and their treatment and control will be typical of U.S.
practice. The typical processes are described in the follow-
ing paragraphs and generalized flowsheets are presented.
Specific waste stream flows and compositions are presented
in the following section on aqueous process effluents.
NITROGEN FERTILIZER PRODUCTION PROCESSES
Ammonia Production
On a tonnage basis, ammonia is produced on a larger scale
in the United States than any other inorganic chemical except
sulfuric acid.(2) More than 75% of the production is con-
sumed in fertilizer applications.
45
-------
Almost all of the ammonia is produced by the direct reaction
of nitrogen and hydrogen (Haber-Bosch process) in a converter.
The reaction is
This reaction is exothermic and care must be taken to con-
trol the temperature in the converter. The conversion to
ammonia is low (10-20%) and the standard practice is to
condense the ammonia from the gas stream and recycle the
unreacted nitrogen and hydrogen. Increasing the reaction
pressure increases the ammonia conversion and most ammonia
plants operate the converter in the range of 120-700 atmospheres
pressure. (16,17,18) increasing the reaction temperature
decreases the ammonia concentration at equilibrium. However,
increasing the temperature increases the rate of ammonia
formation so it is necessary to select a converter temper-
ature which gives maximum ammonia production. Most converters
operate in the range of 400-600 °C. (16 '17 /18)
The rate of reaction between nitrogen and hydrogen is very
slow regardless of temperature. Economic production of
ammonia, therefore, requires a catalyst to accelerate the
reaction rate to a useful level. The catalyst used is iron
promoted with metal oxides such as CaO, A^Og, MgO, K2O,
and SiO2 (each catalyst manufacturer has his own formula) .
Effective catalyst life is dependent on a number of factors
and excessive temperature is especially damaging. Therefore,
careful control of the temperature in the reactor must
be maintained to obtain an optimum balance of ammonia pro-
duction and catalyst life. Impurities in the feed gas can
also have a harmful effect on the catalyst. Sulfur, chlorine
and oxygen containing compounds must all be removed from
the feed to insure efficient catalyst operation.
A major factor in ammonia synthesis is the preparation of
the feed gas. Atmospheric nitrogen is readily available
and is always the nitrogen source used. Obtaining the hydro-
gen is always the major problem in an ammonia plant. A
number of different hydrogen sources have been used, includ-
ing coke oven gas, refinery gas, naphtha, fuel oil, crude
oil, and electrolytic hydrogen. In the United States at
the present time most of the hydrogen used in ammonia pro-
duction comes from natural gas.
There are four major steps in preparing the hydrogen from
natural gas: (1) removal of sulfur impurities; (2) reaction
of the gas with oxygen to form hydrogen and carbon monoxide;
(3) conversion of the carbon monoxide to carbon dioxide;
and (4) removal of the carbon dioxide and residual carbon
46
-------
monoxide from the hydrogen. In most cases the sulfur compounds
are removed from the natural gas by the gas producer. If
additional sulfur removal is required then the ammonia producer
can use ethanolamine/ iron oxide, or activated carbon to
reduce the sulfur to the required level.
Several processes have been developed for converting the
natural gas to hydrogen and carbon monoxide. In the United
States, steam reforming and noncatalytic partial oxidation
are the most widely used processes. Approximately 65% of
U.S. ammonia production uses steam reforming of natural
gas. In the steam reforming process, natural gas and steam
are reacted over a nickel catalyst at 1200-1300 °C in a
primary reformer to form carbon monoxide and hydrogen. The
reaction is
C H + x H20 •*• x CO + (x + £) H2
The reaction is only partially completed in the primary
reformer. A secondary reformer is used to convert the re-
maining hydrocarbon. Air is injected into the secondary
reformer to burn unreacted hydrocarbon and hydrogen and
to supply heat. The air used is sufficient to give a hydrogen-
to-nitrogen ratio of 3:1 in the product gas stream. The
gas from the secondary reformer contains hydrogen, carbon
monoxide, nitrogen, steam, and small amounts of carbon dioxide,
argon, and unreacted hydrocarbon.
In the noncatalytic partial oxidation process the hydrocarbon
feed and oxygen, or oxygen-enriched air, are preheated separ-
ately and then reacted at about 1400 °C and 30 atmospheres
pressure to form carbon monoxide and hydrogen:
CxHy + I °2 - x C0 + 2 H2
The product gas stream contains hydrogen, carbon monoxide,
carbon dioxide, and small amounts of nitrogen, unreacted
hydrocarbon, and argon. The oxygen for the reaction is
obtained from a liquid air plant. The nitrogen from the
liquid air plant serves as the nitrogen supply for the syn-
thesis reaction.
The oxides of carbon must be removed from the synthesis
gas before it goes to the converter. First step in the
purification of the synthesis gas is the shift conversion
of carbon monoxide to carbon dioxide. This is accomplished
by reacting the carbon monoxide with steam over an iron
oxide-chromium oxide catalyst:
47
-------
CO + H2O -*• H2 +
The reaction, which is exothermic, is usually carried out
in a series of converters at a temperature of about 400 °C
using an excess of steam. The degree of conversion desired
depends on the process which is used to remove the carbon
dioxide from the synthesis gas. It is usually preferable
to reduce the carbon monoxide to less than 0.5%.
Several methods have been developed for removing the carbon
dioxide from the synthesis gas. In the United States monoethano-
lamine (MEA) scrubbing is the preferred process. By scrubbing
the synthesis gas with a 30% MEA solution, it is possible
to reduce the CC>2 content of the gas to less than 0.5%.
The CO2 is stripped from the MEA solutions and the solution
is recycled. The CC>2 is discharged to the atmosphere, or
can be used in other processes such as urea production if
desired.
The synthesis gas from the MEA stripping process still con-
tains several hundred parts per million oxides of carbon.
In order to use the gas in the ammonia converter the carbon
oxides should be less than 10 ppm. Reducing the carbon
oxides to the required level is best accomplished by methana-
tion. This is done by heating the gas to about 300-400 °C
and passing it over a nickel catalyst where the carbon oxides
and hydrogen react to form methane and water:
CO + 3 II -> CH4 + HO
C02 + 4 H2 -» CH4 •*• 2 II2O
The methane formed does not have to be separated from the
synthesis gas since it has no affect on the catalyst. Except
for the water the gas from the methanation process is suffi-
ciently pure to enter the ammonia converter. The water
is removed from the gas when it is compressed prior to enter-
ing the converter.
A number of processes have been developed for ammonia pro-
duction based on the original Haber-Bosch process. These
processes differ primarily in the design of the converter
used and the reaction conditions (primarily temperature
and pressure) in the converter. Figure 16 is a simplified
flowsheet typical of the ammonia process as it is practiced
in the United States. In this process, natural gas free
of sulfur is compressed to about 30 atmospheres and then
mixed with high pressure steam. The resultant mixture flows
to the primary reformer where it is indirectly heated to
1200 °C in the presence of a nickel catalyst. The reaction
48
-------
PRIMARY REFORMER
SECONDARY
REFORMER
NATURA
LER
WATER
L G/
VS STEAM DR^
STEAM
1
* t t
n — — n
Cl 1)
| !
CTC
1
1
AU
/
K A
1 V M
1
^
•V-
t
1«.
~ FUE
L
*f
**,
\
CO SHIFT CONVERTER
FIRST STAGE
AIR
HEAT
RECOVERY
SECOND STAGE
RECOVERY " f
WATER
CO,
•a-1
HEAT
ABSORBER
C02 STRIPPER
PURGE
NH3 REFRIG.
METHANATOR
SEPARATOR T AMMONIA
PRODUCT
FIGURE 16 TYPICAL FLOWSHEET FOR AMMONIA PRODUCTION
(See pp 93-95 for Waste Stream Volume and Composition)
-------
products flow to a secondary reformer where compressed air
is added and a portion of the hydrocarbons and hydrogen
burned. The heat released by the combustion of the gas
raises the gas temperature and completes the conversion
of the natural gas. The resultant gas mixture is cooled
to about 400 °C and then mixed with high pressure steam.
The steam-gas mixture is fed to the shift converter where
the CO is converted to CO2 . The gas from the shift converter
is cooled and scrubbed with a recirculating solution of
MEA to remove the C02. The gas is then heated to 350 °C
and fed to the methanator where any oxides of carbon are
converted to methane. From the methanator the gas is cooled
and sent to the compressors. The gas is compressed to the
operating pressure of the ammonia converter. The compressed
gas enters the synthesis converter loop where it mixes with
the ammonia containing gas stream from the converter. The
gas stream is cooled to condense the ammonia present and
the ammonia product is removed in a liquid-gas separator.
The remaining gas stream recycles to the converter. To
prevent a buildup of inerts in the recirculating gas stream,
a small portion of the gas is continually purged from the
system. The ammonia synthesis reaction is exothermic and
a considerable amount of heat must be removed from the con-
verter loop by v/ater cooling.
Ammonium Sulfate Production
Ammonium sulfate occupies a decreasing but still important
position in the United States fertilizer industry. Almost
all of the ammonium sulfate produced in the United States
is prepared by reacting ammonia with sulfuric acid according
to the reaction
Ammonium sulfate is normally produced as a crystalline pro-
duct and control of the particle size is a major problem.
In most instances the reaction between the acid and ammonia
is carried out in the aqueous phase. The ammonium sulfate
is precipitated by controlled evaporation and the product
separated from the mother liquor by filtration or centrifu-
gation. The separated product is then dried and screened.
The continuing importance of ammonium sulfate in the ferti-
lizer industry is due primarily to the fact that at least
one of the reactants can often be obtained from certain
waste streams from other industries. Currently, about 60%
of the ammonium sulfate production in the United States
is by the recovery of ammonia, sulfuric acid or ammonium
sulfate from industrial waste or by-product streams.^ '
50
-------
The production of coke from coal by the steel industry results
in the production of a coke oven gas which contains significant
amounts of ammonia. By scrubbing the ammonia containing
gas with sulfuric acid the steel industry produces large
tonnages of ammonium sulfate.
In the petroleum and petrochemical industries significant
quantities of waste sulfuric acid are generated. By reacting
the spent acid waste with ammonia, fertilizer grade ammonium
sulfate can be produced. The production of caprolactam
results in the formation of large quantities of by-product
ammonium sulfate (about four tons per tons of caprolactam).
With the increasig production of caprolactam this is becoming
a very important source of ammonium sulfate.^"'
The ammonium sulfate produced from industrial waste streams
normally contains substantial amounts of impurities. These
impurities do not usually interfere with the use of the
ammonium sulfate as a fertilizer, although they may compli-
cate the recovery of the solid product, by crystallization,
from the aqueous phase.
Ammonium sulfate production from by-product and waste streams
is often used as a means of pollution control. This being
the case, there are no major water pollution problems which
result directly from the by-product production of ammonium
sulfate.
Substantial amounts of ammonium sulfate are produced from
manufactured ammonia and sulfuric acid (approximately 40%
of U.S. production), as opposed to recovery from waste streams.
This manufacture of ammonium sulfate can lead to significant
water pollution problems.
The major problem in reacting ammonia and acid to produce
ammonium sulfate is controlling the particle size of the
sulfate in the reaction system. A variety of reaction systems
have been developed to produce a suitable product. Regardless
of the reaction system used the basic operations required
for ammonium sulfate production are the same. They consist
of reacting the ammonia and acid in an aqueous system, evap-
orating water to precipitate the ammonium sulfate, separating
the sulfate crystals from the aqueous phase, drying the
solids, and then screening to obtain the desired size frac-
tion. A flowsheet typical of ammonium sulfate production
using synthetic ammonia is shown in Figure 17.
In this process, anhydrous ammonia is fed to a crystallizer
where it is mixed with sulfuric acid and a slurry of ammonium
51
-------
\s\
t\i
VENT
BAROMETRIC
CONDENSER
SCRUBBER
NH<
•WATER
CYCLONE
•• WATER
WASTE
CRYSTALLIZER
/
CENTRIFUGE
LIQUOR
MOTHER LIQUOR
TANK
(NH4)2S04
SLURRY
(NH4)2S04
PRODUCT
FIGUEE 17 TYPICAL FLOWSHEET FOR AMMONIUM SULFATE PEODUCTION USING SYNTHETIC AMMONIA
(See p 100 for Waste Stream Volume and Composition)
-------
sulfate crystals. Circulation of the slurry in the crystal-
lizer is designed to give a product of the desired particle
size. The reaction of sulfuric acid and ammonia is exother-
mic and the heat liberated results in the evaporation of
water and precipitation of the ammonium sulfate.
A continuous stream of slurry is removed from the crystallizer
and sent to a continuous centrifuge where the liquor is
removed. The crystal cake discharges from the centrifuge
to a gas fired dryer. In the dryer the ammonium sulfate
particles are dried to less than 0.1% moisture. The dried
product is cooled and screened to the desired particle size.
The saturated liquor from the centrifuge flows to a mother
liquor holding tank and then recycles to the reactor. The
ammonium sulfate oversize and undersize from the screens
flow to a tank where they are dissolved. The resulting
solution then returns to the mother liquor holding tank.
The offgas from the reactor consists primarily of water
vapor but does contain small amounts of ammonia and sulfate.
This stream is condensed in a barometric condenser and discharged.
The gas stream from the product dryer contains ammonium
sulfate particulate matter and small amounts of ammonia.
The gas passes through cyclones to remove the solids and
is then scrubbed with water to remove the ammonia. The
fines collected in the cyclone are sent to the dissolving
tank where they are dissolved with the rejects from the
screens. The scrubber solution discharges to the aqueous
disposal system.
Ammonium sulfate crystals can present handling problems
due to caking. Careful drying can help to reduce caking;
but in some plants the practice is to coat the ammonium
sulfate particles with an inert material such as clay to
prevent caking.
Ammonium Nitrate Production
Ammonium nitrate is a major source of fertilizer nitrogen
in the United States. It makes an excellent fertilizer
material because of its high nitrogen content (35%) and
low cost of preparation. Large amounts of ammonium nitrate
are used in direct fertilizer applications. In addition,
it is widely used in mixed fertilizers, both liquid and
solid, and in nitrogen solutions.
Almost all ammonium nitrate production in the United States
is based on the reaction of ammonia with nitric acid:
53
-------
The acid-ammonia neutralization reaction is straightforward
and easily carried out. The major problems in ammonium
nitrate production relate to recovery of the solid product
from the neutralized solution. The hazardous nature of
solid ammonium nitrate adds to the recovery problems. There
are three major processes for preparing solid ammonium nitrate
crystallization, flaking, and prilling. "9) of the three,
prilling is the most important in the United States.
The basic steps in the production of solid ammonium nitrate
are reaction of the acid and ammonia, concentration of the
resultant solution, and conversion to solid ammonia nitrate.
Figure 18 is a simplified flowsheet typical of the ammonium
nitrate process and showing the major alternatives for the
recovery of the solid ammonium nitrate. Typically, anhydrous
ammonia and nitric acid are reacted in a neutralizer (the
reaction can be carried out under vacuum, at atmospheric
pressure, or at an elevated pressure). The reaction is
highly exothermic and water evaporates from the reaction
mixture to give a concentrated solution containing 80-85%
ammonium nitrate. The solution from the neutralizer flows
to a concentrator where additional water is evaporated to
raise the ammonium nitrate concentration to about 95%.
If a prilled product is desired, the solution from the con-
centrator is pumped to the top of a tall tower and sprayed
down into a rising current of heated air. As the droplets
of solution descend in the tower water is evaporated and
spherical particles of solid ammonium nitrate are formed.
The size of the particles produced is controlled by spray
nozzles, and is usually in the range of 8-16 mesh.™) The
particles from the prilling tower are further dried in a
gas fired dryer to reduce the moisture of the ammonium nitrate
to less than 0.5%. The dried product, after cooling, is
screened and then coated with an anti-caking agent such
as clay. The undersize particles from the screens are dis-
solved and recycled to the concentrator.
When crystalline ammonium nitrate is desired, the solution
from the concentrator is fed to a continuous crystallizer
similar to that used for ammonium sulfate production. In
the crystallizer the solution is cooled by vacuum evapora-
tion to produce the crystalline ammonium nitrate. The cry-
stalline product obtained is withdrawn from the crystallizer
and sent to a continuous centrifuge where the crystals and
liquor are separated. The liquor from the centrifuge recycles
to the crystallizer. The ammonium nitrate crystals are
dried to less than 0.1% water, cooled, screened, and coated
with an anti-caking agent. The undersize particles from
the screens are recycled to the crystallizer.
54
-------
VENT SCRUBBER
(OR CONDENSER)
DISSOLVED
FINES
RECYCLE
DISSOLVED
FINES
RECYCLE
FIGURE 18 FLOWSHEET FOR AMMONIUM NITRATE PRODUCTION
(See p 100 for Waste Stream Volume and Composition)
55
-------
In the ammonium nitrate processes as described, there are
several operations which generate waste streams that can
present disposal problems. In the neutralizer, a small
amount of ammonia (and nitric acid) can be lost with the
water evaporated. Condensation of the water vapor results
in a waste stream contaminated with ammonia and nitrate
ion. Operation of the concentrator can produce a similar
waste stream. In the prilling process, the air stream which
leaves the top of the prilling tower contains a mixture
of water vapor, ammonia, and nitrogen oxides. When this
gas stream is scrubbed with water a waste stream contain-
ing ammonia is generated. The offgas from the dryer contains
ammonium nitrate particles. The solids are separated by
means of a cyclone, redissolved, and recycled to the concen-
trator.
In the crystallization process, the water evaporated in
the crystallizer can contain small amounts of ammonia and
nitric acid. Condensation of the water gives a waste stream
contaminated with both ammonia and nitrate. The dryer used
in the crystallization processes present the same types
of disposal problems as are encountered with the dryer in
the prilling process.
When an ammonium nitrate solution is the desired product
the neutralization reaction is carried out in equipment
similar to that described above. The major waste disposal
problem is the aqueous waste stream from the neutralizer
condenser which is contaminated with ammonia and nitrate.
Urea Production
Urea is a major source of fertilizer nitrogen in the United
States. Its use in fertilizer applications is growing at
a rapid rate, and based on the tons of nitrogen supplied
trails only ammonia and ammonium nitrate in U.S. consumption.
All urea production in the United States is based on the
reaction of ammonia and carbon dioxide to form ammonium
carbamate which is then dehydrated to form urea:
Since the raw materials for urea production are ammonia
and carbon dioxide, urea plants are almost always operated
in conjunction with an ammonia plant. The ammonia plant
serves not only as the source of anhydrous ammonia, but
also as the source of high purity carbon dioxide which is
a by-product from the preparation of the ammonia synthesis
gas.
56
-------
The reaction between ammonia and carbon dioxide is carried
out in a pressurized reactor to form a solution containing
urea, ammonium carbamate and water. The ammonia-carbon
dioxide reaction is strongly exothermic and under the condi-
tions found in the urea reactor proceeds almost to completion.
The dehydration reaction is endothermic and under the reaction
conditions proceeds to 40-60% completion. Overall, the
two reactions are exothermic and external cooling must be
provided to maintain the reaction conditions. Both reactions
are reversible and the equilibria obtained will depend on
the pressure, temperature, and concentrations of the reactants
and products. Urea formation increases with increasing
temperature. Increasing the temperature, however, requires
an increase in operating pressure because the urea only
forms in the liquid phase. Since the pressure required
to maintain the liquid phase increases very rapidly with
temperature the reactor is normally held below 210 °C. Depend-
ing on the process being used, the reactor temperatures
vary from 170-210 °C and the pressure from 135-400 atmospheres.
The product stream from the reactor is a mixture of urea,
ammonium carbamate, water, unreacted ammonia, and carbon
dioxide. An excess of ammonia is always used so the CO2
concentration in the exit stream is low. The final steps
in the urea process involve the decomposition of the ammonium
carbamate, recovery of the urea product in usable form,
and possible recycle of the unreacted ammonia and carbon
dioxide.
Basically, there are three types of processes for urea pro-
duction which differ primarily in the way in which the uncon-
verted ammonia and carbon dioxide from the reactor are handled:
1. Once-through process - In the once-through process
no attempt is made to recycle the unreacted ammonia
and carbon dioxide. Instead, the offgas ammonia is
used in the production of other fertilizer products
or intermediates (i.e., HN03, NH4NO3, etc.).
2. Partial Recycle Process. In the partial recycle pro-
cess a portion of the unreacted ammonia is recovered
and recycled while the carbon dioxide is used in
other processes or is wasted.
3. Total Recycle Process. In the total recycle pro-
cess both the unreacted ammonia and carbon dioxide
are recovered and recycled to the urea reactor.
57
-------
The choice of which process to use depends on a number of
factors, and requires a detailed analysis to arrive at a
satisfactory decision. The once-through process has the
advantage of a relatively low capital cost. However, its
economic operation requires that the unreacted ammonia can
be used in other processes and that carbon dioxide be avail-
able at low cost. Partial Recycle and Total Recycle are
less dependent on raw material costs and on integrating
with other processes, but they have significantly higher
capital costs. The current study indicates that greater
than 50% of the urea plants in the United States use total
recycle.
After the ammonium carbamate, ammonia, and carbon dioxide
are removed, the product solution contains about 75-80%
urea. Subsequent processing of the urea solution depends
on the product form required. If the urea is used in liquid
fertilizers or nitrogen solutions, no further processing
is required except for the final blending operation. If
a solid product is desired, there are three major methods
for converting the urea solution to a solid: (1) crystalliza-
tion of the urea, (2) prilling, and (3) crystallization
followed by prilling.
The selection of the method of solidification depends primarily
on the end use for the urea and on the biuret content that
can be tolerated in the product. When urea is heated it
decomposes to form biuret according to the reaction
2 NH2CONH2 *; NH2CONHCONH2 + NH3
The decomposition reaction proceeds relatively slowly below
120 °C, but becomes quite rapid as the temperature approaches
the melting point of urea (132.7 °C). In making prilled
urea it is necessary to melt the urea. This can result
in a significant biuret content in the final product (greater
than 1%). Biuret formation is much less in making crystalline
urea and by careful control can be held below 0.1%. By
using crystallization followed by prilling, it is possible
to obtain a prilled product containing less than 0.5% biuret.
In many fertilizer applications the biuret concentration
of the urea must be kept as low as possible because of possi-
ble crop damage. This means that product end use has a
significant effect on selection of the solidification process
used. Fertilizer grade urea normally contains less than
1% biuret, but can contain up to 2.5%. Technical grade urea
for industrial purposes may contain as little as 0.1% biuret.
58
-------
Urea has a tendency to cake unless the moisture content
is kept low. Both crystalline and prilled urea are dried
to a moisture content of less than 0.3% moisture, and should
be stored in moisture-proof bags to prevent adsorption of
water. Urea frequently is conditioned with clay or similar
materials to prevent caking.
A number of companies have developed urea processes and
almost every process has the option of once-through, partial
recycle, or total recycle operation. The differences between
the various processes are involved primarily with the ways
of handling the ammonia and carbon dioxide offgases from
the ammonium carbamate decomposition. Regardless of the
process design used, the basic operation of urea production
are the same and the waste streams generated are similar.
Figure 19 presents a simplified flowsheet typical of urea
production using total recycle and a prilling tower. Anhydrous
liquid ammonia is fed to a reactor where it is reacted with
carbon dioxide to give a solution of urea, ammonium carbamate,
and water. The temperature in the reactor is about 185 *C
and the pressure 200 atmospheres. The reaction mixture
flows to a stripping section where its pressure is reduced
to about 20 atmospheres. The excess ammonia is released
from solution and part of the carbamate is dehydrated. The
gases from the stripper are washed to give a pure ammonia
gas stream and an ammonium carbamate solution. The ammonia
is recycled to the ammonia feed systems and the carbamate
solution is recycled to the reactor. The solution from
the stripper flows to a decomposer (operating at about 5
atmospheres pressure) where the rest of the carbamate dehydrates,
The resultant solution, which contains 75-80% urea, flows
to a concentrator. The vent gases from the decomposer are
condensed to form a dilute ammonium carbamate solution which
recycles to wash the gas stream from the stripper.
In the concentrator, water is evaporated to raise the urea
concentration to greater than 98%. The molten urea from
the concentrator is pumped to the top of a prilling tower
where it is sprayed downward against a stream of cold air.
The urea prills from the tower are cooled, screened and
packaged for storage.
The water evaporated from the concentrator contains small
amounts of ammonia. The water is condensed and the ammonia
is stripped from the condensate with steam. The water is
discharged to the environment and the ammonia steam gas
stream is recycled to the process.
59
-------
0\
o
UREA-HzO
CARBAMATE
NH3
CO?
H0
CC
O
<: i—
LU CJ
NH,
-»• »
CO,
AIR'
WASTE.
±
cc
LU
H20
WASTE
O
u
3;
CARBAMATE
' UREA-CARBAMATE
CD
CO
=3
CC
O
\f>
H20
SCREENS
1
COOLER
UREA
o
z
1-1 CH.
PRODUCT
—* I AIR
to
o
o.
o
z
o
DILUTE CARBAMATE
an
o
tJ
z
o
cc
-> a.
z: Q.
O 1-1
^ ac.
UREA
STEAM
WASTE
FIGURE 19 TYPICAL FLOWSHEET FOR UREA PRODUCTION - TOTAL RECYCLE AND PRILLING TOWER
(.See pp 97-99 for "Waste Stream. Yolume and Composition)
-------
The air stream from the prilling tower also contains some
ammonia. Air pollution requirements may necessitate water
scrubbing the ammonia from the air prior to discharge. The
scrubber solution may then be treated to remove the ammonia
before the water is discharged.
PHOSPHATE FERTILIZER PRODUCTION PROCESSES
Wet Process Phosphoric Acid Production
In the wet process for preparing phosphoric acid the phos-
phate rock (principally f luorapatite) is reacted with a
strong mineral acid to convert the tricalcium phosphate
to orthophosphoric acid (HsPO^j) . Hydrochloric and nitric
acids can be used, but the preferred route is the reaction
with sulfuric aicd. The overall reaction of the tricalcium
phosphate and sulfuric acid results in the formation of
phosphoric acid and calcium sulfate:
The reaction proceeds in several steps but the overall re-
action is as written above.
The calcium sulfate formed is relatively insoluble in phosphoric
acid and it precipitates from solution. Depending on reaction
conditions, the calcium sulfate can crystallize out as an
anhydrite (CaSO^j) , a hemihydrate (CaSO^j.l/^I^O) , or a dihy-
drate (CaSC>4'2H20) commonly referred to as gypsum.
Almost all phosphate rock from a given deposit will contain
fluorine and P2G5 ^n a relatively constant ratio which approxi-
mates that for f luorapatite. The fluoride content of the
rock is attacked by the sulfuric acid during acidulation
with the formation of calcium sulfate and hydrogen fluoride:
CaF2 + H2S04 -»• CaS044 + 2 HF . .
The overall reaction of fluorapatite with sulfuric acid
is, therefore,
3Ca, (POA) ,'CaF? + 10 H-SO. + 10 X H0O M .
J 4 2 (s) 2 4(aq) 2 U)
+ 6 H3P04 + 2 HF(ag) + 10 CaS04.XH20(s)
laq;
where X can be zero, 1/2 or 2 depending on the reaction
conditions.
61
-------
The hydrogen fluoride which forms will react with any silica
or silicates in the rock to form silicon tetrafluoride which
then hydrolyzes to form fluosilicic acid:
Si02 + 4 HF(ag) + SiF4 + 2 H20
3 SiF, + 2 H00 •* 2 H0SiF,. + SiO.,*-
ft / £ o f.
Depending on reaction conditions, some of the fluosilicic
acid will decompose with the volatilization of silicon tetra
fluoride, according to
H2SiP6 -> SiF4t + 2- IIF(aq).
Depending on the impurities in the rock, there are a number
of minor reactions which can occur during acidulation. If
any carbonate is present, it will react with the sulfuric
acid with the evolution of carbon dioxide:
CaC03 + H2SO4 -> CaS04* + CO2 + H20 .
If any sodium and potassium are present alkali fluosilicates
will be formed:
Na0O + H0SiF,- -> Na0SiF, + H-O ,
JL 2. b 2 D /.
These compounds can be particularly troublesome. They are
a principal cause of scaling in equipment and piping, and
are corrosive.
If hydrochloric acid or nitric acid is used to attack the
rock, the calcium by-product is soluble in the phosphoric
acid. The reactions are:
Ca3(PO4)2 + 6 HC1 -»• 3 CaCl2 + 2 I
Ca3(P04)2 + 6 HN03 -> 3 Ca(N03)2 +
The preparation of phosphoric acid by either reaction requires
a method for removing the soluble calcium salt from the
acid. Phosphoric acid processes have been developed which
utilize either hydrochloric or nitric acid but none are
used commercially in the United States.
The production of phosphoric acid by the wet process using
sulfuric acid is the principal source of fertilizer phosphoric
acid in the United States. Preparation of H3PO/ by the
wet process is strongly influenced by the nature of the
calcium sulfate by-product. The P2°5 recovery from the
62
-------
phosphate rock, the filtration rate of the calcium sulfate-
phosphoric acid slurry, and the strength of lloPO^ produced
are all related to the form of the calcium sulfate produced.
It is possible to precipitate the calcium sulfate in three
different forms and this has led to the development of three
general types of wet processes.
1. The anhydrite process
2. The hemihydrate process
3. The dihydrate process
Various modifications of the three types of processes exist
and Slack(9) has further subdivided the various processes
into the following classifications:
1. Continuous pure dihydrate crystallization (standard
U.S. process)
2. Continuous primary dihydrate crystallization with
subsequent recrystallization into hemihydrate
3. Continuous pure hemihydrate crystallization
4. Continuous crystallization of inactivated hemi-
hydrate with partial simultaneous crystallization
of dihydrate
5. Continuous primary hemihydrate crystallization and
subsequent recrystallization to dihydrate before
filtration
6. Continuous hemihydrate crystallization in strong acid,
acid separation and dilution and recrystallization
to dihydrate
7. Continuous primary hemihydrate crystallization and
secondary anhydrite recrystallization
8. Continuous pure anhydrite crystallization
9. Anhydrite crystallization with batchwise reactor
charging
10. Production of anhydrite clinker followed by leaching.
In the United States only the dihydrate process has achieved
any commercial significance, and a number of companies have
developed processes based on the precipitation of the dihydrate,
The processes vary principally in the design of the reactor
systems in which the rock and sulfuric acid are reacted
and in the method used to filter the gypsum from the product
acid. There are also variations in the types of evaporators
which are used to concentrate the phosphoric acid from the
filters.
63
-------
Regardless of the type of dihydrate process design used,
the basic operations are the same. A flowsheet typical
of the dihydrate processes and showing the basic operations
is shown in Figure 20. Typically, beneficiated phosphate
rock, containing 30-35% P2°5/ is ground to 60% -200 mesh.
The ground rock is fed continuously to a reaction system
where it is mixed with sulfuric acid and a recycle stream
of dilute phosphoric acid. The reaction system consists
of one or more digestion tanks depending on the design.
The more recent developments favor the use of a single,
large compartmented tank. In the digestion tank(s) the
rock and acid react to form gypsum (dihydrate) and phosphoric
acid. To obtain complete reaction of the rock, the rock-
gypsum-acid slurry is rapidly circulated between the diges-
tion tanks or the compartments of a single tank.
To insure the formation of dihydrate, the temperature in
the digester is maintained at about 75 °C. Since the reac-
tion between the phosphate rock and sulfuric acid is exothermic/
heat must be removed from the digester system to maintain
temperature control. The amount of heat to be removed depends
largely on the concentration of the sulfuric acid. In most
plants the concentration of the sulfuric acid used will
be between 77 and 98% with 93% acid being the norm. The
heat can be removed from the system by a number of ways
including vacuum coolers, air sparging, and dilution and
cooling of the sulfuric acid prior to use.
Retention time in the digester is in the range of 3-8 hours.
The extraction of P2°5 from the rock normally exceeds 96%
and the strength of phosphoric acid produced is in the range
of 30-32% P2°5- TnG gypsum formed must be easily filtered
and retain a minimum of
The acid slurry flows from the digester to a filter system
where the gypsum is removed. Good filter design is quite
complex, and a number of companies have developed filters
especially for the phosphoric acid industry. Typically
the filter system yields two phosphoric acid streams: a
30-32% P2C>5 product stream and a stream containing about
20% P2°5 which recycles to the digester. The gypsum cake
is thoroughly washed with water to remove residual acid.
The gypsum cake from the filter is slurried with water and
discharged to waste.
The 30-32% P2°5 product stream from the filter is sent to
an evaporator system where it is concentrated to give a
phosphoric acid product containing about 54% P2C>5. Because
of impurity precipitation and scaling, concentrator design
is quite complex and a number of companies have developed
a variety of concentrators designed to control scaling.
64
-------
PHOSPHATE
ROCK
WATER
L
VENT TO ATMOSPHERE
FLUORINE
sr.PIIRRFR
1 .
WASTE WATER
m
DIGESTION TANK
r
WATER
ACID
CONCENTRATOR
-»• FILTERS (
LJJT_J
r -
WATER
SLURRY TANK
GYPSUM SLURRY
TO GYPSUM POND
WATER
»• VENT TO ATMOSPHERE
FLUORINE
SCRUBBER
PHOSPHORIC
ACID PRODUCT
STORAGE
FIGURE 20 WET PROCESS PHOSPHORIC ACID FLOWSHEET
(See pp 103-107 for Waste Stream Volume and Composition)
-------
If the product acid is to be used fairly soon for fertilizer
production, no attempt is made to remove the impurities
which precipitate. If the acid is to be stored or shipped
before use, the solids are normally removed. This can be
done by storing the acid for several days in tanks and allowing
the solids to settle. The thickened solids, containing
a substantial amount of ^2°5' are c°Hected and normally
used in fertilizer manufacture. The clarified acid is ready
for shipment.
As mentioned previously, any fluorine in the phosphate rock
will react in the digester to form IIP and II^SiFg. Fluorine
will be released to the atmosphere as SiF4, HF, and fluoride
dust in varying amounts in the vent gases from the phosphoric
acid process. To control air pollution it is necessary that
the fluorine evolved be removed from the gas streams prior to
discharge to the atmosphere. The major source of fluorine
evolution are the reactor, the concentrator, and the filter.
It is standard practice to water-scrub the vent gases for
fluoride removal. This results in a waste water stream that
must be treated for fluoride removal before discharge to the
environment. It is normal practice in some plants to recover
part of the fluoride values from the waste water as fluosili-
cates and the resultant contaminated water is sent to the
gypsum pond.
The wet process phosphoric acid plant produces large tonnages
of gypsum (on the order of 4.5 tons of gypsum per ton of ^2Q5
product). It is general practice to store this gypsum in diked
areas called gypsum ponds. The slurried gypsum from the filter
is pumped to these ponds where the gypsum settles out. The
clarified water is recycled back to the various parts of the
process. The water in the gypsum pond slowly increases in
phosphate and fluoride content until an equilibrium concentra-
tion is reached. The pll of the water decreases to between 1
and 2. If it becomes necessary to discharge the gypsum
pond water to the environment, it must be treated to reduce
the fluorine and phosphate to acceptable levels and to control
the pH.
Electric Furnace Phosphoric Acid
Phosphoric acid produced by the electric furnace process
is of higher purity than acid produced by the v/et process.
Furnace grade phosphoric acid is used primarily in the chemical
and food industries and, at the present time, finds only
limited use in the fertilizer industry. However, with the
increased use of high analysis liquid fertilizers and the
66
-------
potential loss of the detergent market, it is quite possible
that furnace grade phosphoric acid will find substantially
increased use in the fertilizer industry in the future.
While there are a number of companies which produce furnace
grade phosphoric acid, the processes used by the various
producers are quite similar. One does not find the differ-
ences in equipment and operation in the furnace process
that occur in the wet processes.
Basically, the furnace processes involve the reaction of
phosphate rock with carbon and silica in an electric furnace
to form elemental phosphorus. The overall reaction can
be written as follows:
2 Ca3(PO4)2 + 6 Si02 + 10 C -> 6 CaSi03 + P4* + 10 CO*
The elemental phosphorus which volatilizes from the furnace
is collected and oxidized to $2G5* The P2°5 is absorbed in
water to give a concentrated orthophosphoric acid:
P4 + 502 •* 2 P205
P205 + 3 H20 -* 2 H3P04
Two side reactions of importance occur in the furnace. Any
iron oxide in the rock, silica or carbon is reduced to ele-
mental iron:
Fe2°3 +3C * 2 Fe + 3 CO
The elemental iron then reacts with phosphorus to form ferro-
phosphorus,
8 Fe + P4 •*• 4 Fe2P
The ferrophosphorus is stable under the furnace reaction condi-
tions and represents a direct loss of phosphorus. In standard
furnace operation the ferrophosphorus and calcium silicate slag
are allowed to build up in the furnace to predetermined levels
and are then tapped from the furnace in separate liquid phases.
About 10-20% of the fluorine in the rock in the furnace is con-
verted to volatile silicon tetrafluoride and leaves with the
phosphorus. The rest of the fluorine stays with the calcium
silicate slag (probably as CaF2).
67
-------
The furnace process can use phosphate rock of any grade,
but since silica is required as a raw material in the process
a low phosphate, high-silica rock is preferred. Phosphate
rock from the Tennessee area and from the Western United
States is particularly suitable for the furnace process.
A flowsheet typical of the furnace process is shown in Figure
21. Typically, ground phosphate rock is agglomerated to
give about 8 mesh particles. A variety of methods such
as pelletizing, briguetting, or nodulizing can be used to
agglomerate the rock. The rock is then blended with a controlled
amount of low ash coke and silica v/hich is low in iron and
lime. The resultant mixture is charged to the electric
furnace which is equipped with carbon electrodes. In the
furnace the components of the charge react at about 1400 °C
to form elemental phosphorus, calcium silicate slag, and
ferrophosphorus. The ferrophosphorus forms a molten layer
in the furnace with the slag layer on top of the ferrophosphorus
layer. The slag layer is tapped from the furnace about
once an hour, and the ferrophosphorus layer is tapped twice
a day.
The phosphorus leaves the furnace, through openings in the
furnace cover, with the CO and fine particulate matter.
The dust is removed from the gas stream by heated (300 °C)
electrostatic precipitators. The cleaned gases flow to
a condensing unit where the phosphorus is condensed with
water. The molten phosphorus is collected under water and
pumped to tanks where it is stored under water.
The molten phosphorus flows from the storage tank to a com-
bustion chamber where it is burned at 1500-1700 °C in air
to form PO^S* T*ie hot P20(i vaPor passes through a gas cooler
to a hydration tower where the P?0r is reacted with water
to form I^PO^j. The product acid normally contains 54% P2°5
but higher concentrations can be produced.
The offgas from the hydration tower contains some P2°5 and
phosphoric acid mist. This gas stream is treated by precip-
itators or scrubbers to react the P2Os ^md recover the acid
prior to discharge to the atmosphere. The acid recovered
is combined with the product from the hydrator.
When the slag is tapped from the furnace some P2®5 an<^ fluorine
(as SiF^) are evolved from the slag while it is cooling.
Air pollution regulations usually require that these evolu-
tions be controlled, so the gases are normally scrubbed
with water to remove the fluoride and PO^C*
68
-------
OFFGAS
ROCK
0\
AIR
PHOSPHORUS HYDRATOR
BURNER
PRODUCT SHIP
FIGURE 21 FLOWSHEET FOR PHOSPHORIC ACID PRODUCTION BY THE ELECTRIC FURMACE PROCESS
(See pp 107-108 for Waste Stream Volume and Composition)
-------
A portion of the fluoride in the phosphate rock leaves the
furnace with the CO-phosphorus off gas. A portion of this
fluoride, as particulate matter, is collected in the precip-
itator, while the volatile fluoride is collected with the
phosphorus in the water-sprayed condenser.
Condensation of the phosphorus with water and its subsequent
storage and handling under water leads to the formation
of (1) a waste water stream that is contaminated with phos-
phorus and fluoride (so called phossy water) , and (2) a
phosphorus sludge emulsion comprised of water, phosphorus
particles, and solids. Both waste streams present disposal
problems.
Normal Superphosphate Production
Normal superphosphate was once the predominant source of
phosphate fertilizer. In recent years triple superphosphate
and the ammonium phosphates have replaced normal superphos-
phate as the principal source of phosphate. However, because
of its ease of preparation, normal superphosphate is still
used in substantial quantities in the United States.
Basically, the preparation of normal superphosphate involves
the reaction of phosphate rock with sulfuric acid to form
mono-calcium phosphate. The overall reaction which occurs
can be written as follows:
+ H20 -> 2 CaSO4 + CaH4 (PO4) 2«H2O
The monocalcium phosphate thus formed is water soluble and
the phosphate is available for uptake by the soil solutions.
No attempt is made to separate the calcium sulfate from
the monocalcium phosphate.
The calcium fluoride content of the rock is also attacked
by the acid with the formation of hydrofluoric acid which
in turn reacts with silica in the rock to form SiF4 . The
SiF4 in turn reacts with water to form fluosilicic acid:
CaF-> + H0SO. -> CaSO. + 2 IIP , .
<* 2 4 4 (aq)
4 HF + Si02 ->
3 SiF, + 2 H00 -> 2 H^SiP^ + SiO
~
However, some SiF. will be volatilized before it can react
with the water. The amount of SiF4 volatilized depends
on the concentration of sulfuric acid used. The higher
the acid concentration the more SiF. is evolved.
70
-------
Both batch and continuous processes have been developed for
producing normal superphosphate and a number of modifications
of each process exist. Regardless of the process variation
used, the basic process operations are the same. Figure
22 shows a flowsheet which is typical of the continuous
normal superphosphate process. In the process phosphate
rock containing 30-35% PjOj. ^s 9round to 90%-100 mesh. The
rock is fed at a controlled rate to a mixer where it is .thor-
oughly mixed with sulfuric acid which has been diluted to
about 75%. The rock-acid slurry discharges from the mixer
to a pug mill where additional mixing occurs and the reaction
between the rock and acid starts. From the mill the slurry
discharges to a slow moving conveyor (called a continuous
den) where the reaction continues and the slurry hardens
to a plastic mass. Residence time on the conveyor is one
to two hours. As the plastic mass leaves the conveyor it
is cut into chunks. The sectioned product is then transferred
to a storage area for final curing. During storage, which
lasts from 2 to 6 weeks depending on process conditions,
the reaction between the rock and acid approaches completion.
The material hardens to a solid mass and cools. From the
storage area the normal superphosphate product is fed to
a pulverizer (usually a hammer mill) where it is crushed
and screened.
In some cases it is desired to produce a granular product.
The superphosphate can be granulated before or after curing
but granulation before curing is the preferred route.
As stated earlier, some fluorine, in the form of SiF4, is
evolved during the manufacture of normal superphosphate.
Air pollution control regulations require that this fluoride
evolution be adequately controlled. The fluorine evolution
occurs primarily in the mixing and dennirg operations, with
a much smaller amount being released during the curing step.
In practice, the mixer and den are enclosed and the fumes
from the enclosures are scrubbed with water to remove the
fluoride plus any acid mist and dust particles that may be
present.
The fluoride evolved in the curing area is much less but
it may still be necessary to control the fluoride evolution.
This is done by circulating air through the. storage (curing)
building and then scrubbing the air with water to remove
the fluoride.
Triple Superphosphate Production
Triple superphosphate has become one of the principal sources
of fertilizer phosphate in the United States. Monocalcium
phosphate is the principal fertilizer ingredient of triple
71
-------
CONE.
J '
SCALES
GROUND PHOSPHATE ROCK
SULFURIC ACID-
WATER'
/*\ / \
f~"**\ J^
PUG MILL
j CONTINUOUS DE
j i- CUTTER O^
H) CONVEYOR ()
. 1
EXHAUST
»!
1
V 1
v!
GAS
t
~*\
O CONVEYOR () ]
i
WATER
i
*• GAS D
i
h
1 *> WASTEWi
EXHAUST GAS
CORING , . ^ pm yrrTTro
BUILDING * PULVERIZER
PRODUCT
FIGURE 22 NORMAL SUPERPHOSPHATE FLOWSHEET
(See pp 110-111 for Waste Stream Composition)
-------
superphosphate as it is in normal superphosphate. However,
as the name implies, triple superphosphate contains about
three times as much P205 as normal superphosphate. This
is because there is no calcium sulfate in the triple super-
phosphate.
Triple superphosphate is made by the reaction of phosphate
rock with phosphoric acid to give soluble monocalcium phos-
phate. The overall reaction can be written as follows:
3Ca3 (PO4)2'CaF2 + 14 H3PO4 + 10 HjO
-> 10 CaH4(P04)2-H20 + 2 HF
The HF formed reacts with silica to form SiF4 which partially
hydrolyzes to fluosilicic acid and partially is evolved from
the reaction mass :
4 HF + Si02 -> SiF4 + 2 HjO
3 SiF, + 2 H00 •> 2 H-SiF, + SiO-
42 262
As was the case with normal superphosphate, triple superphos-
phate can be manufactured by either batch or continuous pro-
cesses but most production is by the continuous process.
Various modification of the continuous process exist but
the basic operations are the same in each case. Triple super-
phosphate is produced as either granular or run-of-pile material.
A flowsheet typical of the production of granular triple
superphosphate by the continuous process is presented in
Figure 23. In the process, phosphate rock is ground to 75%-
200 mesh. The ground rock is fed to a mixer where it is
mixed with phosphoric acid. The resultant slurry is fed
to a continuous belt where it solidifies. The discharge
from the belt is crushed and sent to a storage pile for curing.
Curing normally takes about 3 to 5 weeks to complete. The
cured product is then sent to a pulverizer where it is ground
and screened. The screened material is then sent to a granulator
where it is mixed with water and steam. The resultant wet
granules are discharged to an air dryer where the water is
evaporated to give a hard, dense, granular product. The
discharge from the dryer is screened and the product sent
to storage. Oversize particles are recycled to the pulverizer
and undersize to the granulator.
Evolution of fluoride, principally as SiF4, occurs primarily
at the mixer and conveyor belt, with only a limited amount
occurring in the product curing area. The mixer and conveyor
are enclosed; the off gases are collected and scrubbed with
water to remove the fluoride.
73
-------
PHOSPHORIC ACID
WATER
PHOSPHATE ROCK
MIXER A. /
CONTINUOUS DEN
o
CONVEYOR
CONVEYOR I
—^EXHAUST GAS
GAS SCRUBBER
WASTEWATER
WATER
CURING
BUILDING
EXHAUST GAS
GAS SCRUBBER
WASTEWATER
PRODUCT
FIGURE 23 FLOWSHEET FOR TRIPLE SUPERPHOSPHATE PRODUCTION
(See pp 111-112 for Waste Stream Composition)
-------
The offgas from the granulator dryer also contains contami-
nants which must be removed before the gas is discharged
to the atmosphere. This is normally done with water scrubbing.
Production of Ammonium Phosphates
The ammonium phosphates have become one of the two major.
sources of fertilizer phosphate. The principal advantages
which lead to their increasing popularity include low produc-
tion costs/ good physical and chemical properties/ and high
analysis.
The ammonium phosphates are normally prepared by the neutral-
ization of phosphoric acid with ammonia. The product obtained
depends on the ratio of acid and ammonia used/ and may be
monoammonium phosphate (NH^H^PO*)/ diammoniun phosphate,
or a mixture of the two. The diammonium phosphate product
is the preferred form in the United States. The ammonium
phosphates can be made from either furnace acid or wet process
acid/ but most ammonium phosphates intended for fertilizer
use are made from wet process acid.
A variety of processes can be used to produce ammonium phos-
phates. The ammoniator process developed by the TVA is typi-
cal of the processes used to prepare diammonium phosphate.
The flowsheet for the TVA process is shown in Figure 24.
In the process/ wet process phosphoric acid (about 40% £2^5)
is partially neutralized in a reactor with anhydrous ammonia.
Heat of reaction evaporates some water and a slurry forms.
The slurry is pumped to an ammoniator-granulator where it
is sprayed over a bed of recycled fines. More anhydrous
ammonia is added in the granulator until the product has
a NH-:PO4 ratio of 2:1. An excess of ammonia is required
to reach the desired ratio of 2:1. In the granulator, the
slurry hardens as the correct ratio is reached and a granular
product is formed. The solidified diammonium phosphate pro-
duct from the granulator flows to a dryer and then to screens
where the oversize and undersize are removed. The oversize
particles are crushed, combined with the undersize and recycled
to the granulator. The product is sent to storage.
The exhaust gas from the granulator and dryer contains water
and ammonia. The water which evaporates from the neutralizer
also contains a substantial amount of ammonia. The ammonia
in the two streams must be recovered before the gases are
discharged. This is done by scrubbing the streams with the
phosphoric acid which feeds the neutralizer. Scrubbing with
phosphoric acid effectively removes most of the ammonia from
the vapor. However, in the scrubbing operation fluoride
is stripped from the phosphoric acid and released to the
gas stream. A second scrubber system may be required to
75
-------
PHOSPHORIC ACID
GAS DISCHARGED
AMMONIA
OTHER MAT'LS
(OPTIONAL)
AMMOMATOR
GRANULATOR
FIGURE 24 FLOWSHEET FOR PRODUCTION OF DIAMMONIUM PHOSPHATE (DAP)
-------
reduce the fluoride discharging to the atmosphere. A water
scrubber is normally sufficient.
Ammoniated Superphosphate Production
Ammoniated superphosphates have been and still are popular
fertilizers in the United States. They are produced by re-
acting normal or triple superphosphate with ammonia (usually
in the form of an ammoniating solution) and the product com-
position can be varied by controlling the ratio of feed mater-
ials. In addition, it is possible to incorporate other fer-
tilizer materials such as urea, ammonium nitrate, potassium
chloride into the production scheme to make a wide range
of fertilizer materials.
The reactions involved in ammoniated superphosphate production
are quite complex. The overall effect is to convert the
monocalcium phosphate in the superphosphate to dicalcium
phosphate and the ammonia to monoammonium phosphate :
CaH4(PO4)2 + NH3 -* CaHPO4 + NII^PC^
Care must be taken to control the reaction since an excess of
ammonia can result in the conversion of the dicalcium phosphate
to insoluble tricalcium phosphate:
3 CalIPO4 + NH3 •> Ca3(PO4)2 + NH4K2P04
Two secondary reactions can occur if calcium sulfate is present
in the superphosphate. These are:
-I- CaS04 + NH3 -> CaHPO4
2 CaHPO4 + CaS04 + 2 NH3 -> Ca3
The first reaction is not harmful , but the second reaction
is to be avoided if possible because of the insoluble trical-
cium phosphate formed.
A great many variables affect the rate and extent of ammonia-
tion. Careful consideration must be given to such factors
as superphosphate particle size and moisture content, reaction
time and temperature, and method of ammonia addition. (3 ,19, 21)
Numerous processes have been developed for producing ammon-
iated superphosphates. They can be divided into two basic
types: batch ammoniation processes with continuous cooling, ,3
drying, and screening; and continuous ammoniator processes. '
The development of ammoniated superphosphate processes is
closely related to the development of granular fertilizer
processes. Almost all ammoniated superphosphate is produced
77
-------
in granular form, and the granulation techniques used are
generally applicable to other products such as granular N-
P-K mixtures and granular ammonium phosphates.
A typical flowsheet for production of granular ammoniated
superphosphates is given in Figure 25. The process uses
a rotary drum ammoniator-granulator. In the process, finely
ground superphosphate (either normal or triple) is screened
and fed to the ammoniator. Oversize particles are milled
and recycled to the screens. In the ammoniator, the super-
phosphate is reacted with ammoniating solution to give a
granular product. Retention time in the ammoniator is about
four minutes and the reaction temperature is between 150
and 250 °F. The reaction is exothermic but it may be necessary
to supply heat by adding steam directly to the ammoniator
to maintain the temperature. The granular product from the
ammoniator flows to a directly-heated rotary dryer where
any moisture present is removed.
The dry granules are then cooled, screened and sent to storage.
The undersize product is returned to the ammoniator while
oversize material is milled and returned to the product stream
to be rescreened.
It is necessary to control dust and fume evolution from several
operations in the process. Dust is generated by the dryer,
cooler, mills, and screens. Fumes evolve from the ammoniator
and dryer. The fumes consist primarily of water vapor and
a small amount of ammonia. The dust is collected by cyclones
and returned to the granulator. Any offgas containing ammonia
is scrubbed and the scrubber solution is returned to the
ammoniator.
Phosphate Rock Production
Phosphate rock is the principal raw material for all phosphate
fertilizer production. However, ground phosphate rock itself
is sometimes used as a fertilizer under special conditions.
The consumption of rock as a fertilizer is decreasing in
the United States but substantial tonnages are still used.
Very little processing is required to prepare the rock for
use. Normally it is simply ground to -200 mesh and dried
to remove the moisture. As long as the drying temperature
is held below about 250 °C no gaseous fluoride contaminants
are evolved. The exit gas from the dryer contains particulate
matter and this is recovered by cyclones. If for any reason
it is desired to calcine the rock then some fluoride will
be evolved. The fluoride will be in the form of hydrogen
Fluoride and silicon tetrafluoride with the higher calcination
cmperatures favoring hydrogen fluoride evolution. The offgas
from the calciner is first passed through cyclones to separate
73
-------
IO
-•—SOLID FERTILIZER HATERIALS
1 (OPTIONAL)
»• VENT
SCRUBBER
'ASTE WATER
FIGURE 25 AMMONIATED SUPERPHOSPHATE FLOWSHEET (Can be used for NPK Fertilizer Production)
-------
particulate matter and then scrubbed to remove the gaseous
fluoride. If the fluoride is not recovered as a by-product
then the scrubber solution must be disposed of.
POTASH PRODUCTION PROCESSES
The major sources of potash in the United States are natural
brines and sylvinite deposits, with the sylvinite supplying
approximately 90% of the potash. Flotation beneficiation
and solution-recyrstallization are the principal processes
used to recover KC1 from sylvinite ore. Flotation plus solar
evaporation is also used to recover KCl from the Bonneville,
Utah brines. A complex series of solution-crystallization
operations is used to recover KCl from Searles Lake brine,
Flotation-Belief iciation Process
In processing sylvinite by flotation, either the KCl or the
NaCl may be floated depending on the chemical reagents used.
Current practice in the United States favors flotation of
the KCl.~14) The exact features of the flotation process
vary somewhat among the various potash producers, depending
on such factors as ore composition and end product requirements.
A simplified flowsheet typical of the flotation beneficiation
process (with KCl floated) is presented in Figure 26. (14,22)
Typically, the coarsely-crushed slyvinite ore is sent to
a grinding circuit where it is crushed and screened to give
a -6 mesh fraction. Grinding to -6 mesh frees the individual
KCl and NaCl particles in the ore. The -C mesh feed is then
pulped with a brine saturated with KCl and NaCl. The result-
ing slurry is thoroughly agitated to disperse the clay slimes
and other insolubles in the ore. The sylvinite normally
contains about 1% clay. The slurry is then treated in a
series of screens and classifiers to remove the clay slimes
and insolubles. The clarified slurry next flows to a condi-
tioning tank where it is mixed with a depressant (starch,
guar, etc.). The depressant coats any residual clay particles
in the slurry so that they do not float with the KCl. Next,
the slurry flows to a second conditioning tank where selective
flotation agents are added. T7hen the KCl is to be floated
the flotation agents are usually aliphatic amine acetate
salts.^14) Aliphatic alcohols may also be added to serve
as frothing agents. From the conditioning tank the slurry
flows to a primary bank of flotation cells {rougher cells)
which are sparged with air. The treated KCl particles collect
in the froth and are skimmed off with paddles. The KCl in
the froth overflow from the rougher cells still contains
"ubstantial amounts of NaCl and other impurities. The KCl
Ich overflow is sent to a second set of flotation cells
•'cleaner cells) where the KCl is again floated off. The
overflow from the cleaner cells is then centrifuged to remove
80
-------
•BRINE H2°
BRINE
RECYCLE
*
SLIMES
REMOVAL
*
REAGENT
ADDITION
1
__ I
CLAY
SLIMES
IFUGE
1
CLAY SLIMES
WASTE
"*
ROUGHER
FLOTATION
*
SCREENS
(OPTIONAL)
i
CLEANER
FLOTATION
*
CENTRIFUGE
^ i FAr
HING
*
CENTR
BRINE ,
BRINE
BRINE BLEED
FOR IMPURITY CONTROL
PRODUCT
STORAGE
SHIP OR FURTHER PROCESSING
FIGURE 26 TYPICAL FLOWSHEET FOR RECOVERY OF POTASSIUM CHLORIDE FROM
SYLVINITE BY THE FLOTATION BENEFICIATION PROCESS (14)
81
-------
the KC1 from the brine and the brine is recycled. After
washing, the KC1 is dried, screened, and stored for shipment
or subsequent processing. The KC1 product is normally a
light pink in color due to iron and clay contamination, and
usually contains 96-97% KC1. Sodium chloride is the major
contaminant.
In some processes, screens are installed between the two
flotation cells and the larger KC1 particles are removed.
These particles are often of sufficient purity that they
do not require further processing to be acceptable as fertil-
izer grade KC1.
The clay slimes and insolubles which are removed from the
sylvinite slurry in the classifiers contains some KC1. The
slurries are normally leached with a minimum of water to
remove the KCl, centrifuged and discharged. The KCl-rich
leach solution is recycled to the process brine system.
The underflow from the primary flotation cells contains primar-
ily NaCl with some KCl. The underflow is heated to dissolve
the KCl, centrifuged, and the solid NaCl waste discarded.
The brine from the centrifuge is recycled to the process
brine system.
Careful control is maintained over the water balance of the
process. The major waste streams from the process are the
clay slimes tailings and the salt waste. The water content
of these wastes are held to a minimum by centrifuging, filter-
ing, or settling the solids prior to discharge. Some water
is lost by evaporation and is replenished by the water used
to leach the clay slimes for KCl removal. In addition, it
is necessary to bleed off some water from the process to
remove soluble impurities. This is held to a minimum to
reduce the KCl loss.
Solution-Recrystallization Process
The solution-recrystallization process for treating sylvinite
ore for KCl recovery is based on the different ratios of
solubility of KCl and NaCl in hot and cold water. Figure
27 shows a typical flowsheet for the recrystallization pro-
cess. (14,22) The ore is first crushed to -6 mesh to free
the individual NaCl and KCl particles. The crushed ore is
then contacted with a recycle brine at 100 °C in a counter-
current series of dissolvers. The hot recycle brine is
saturated with NaCl but unsaturated with respect to KCl.
The KCl in the ore dissolves together with a small amount of
the NaCl. The clay slimes in the ore are suspended in the
brine. The KCl-enriched brine passes to a clarifier where
the clay solids are removed. The clay solids are filtered
82
-------
I
DRYER
SYLVINITE
ORE
I
1
|
UVLHSlZh
BRINE RECYCLE
DISSOLVERS
J
•
CLARIFIER
J
.
CRYSTALLIZERS
I
•
FILTER
I
. fc LEACH — — ^ fFNTR TFIIRF , •»
t J
WATER NaCl WASTE
1
FILTER ^ WATER
CLAY SLIMES
WASTE
^ ' HDTNF
1 HEATER "
BRINE BLEED
FOR IMPURITY CONTROL
1
SCREENS
1
PRODUCT
STORAGE
-^•KCl SHIP OR FURTHER PROCESSING
FIGURE 27 TYPICAL SOLUTION-RECRYSTALLIZATION FLOWSHEET FOR RECOVERING
POTASSIUM CHLORIDE FROM SYLVINITE ORE (14)
83
-------
and discarded, while the filtrate brine is recirculated to
the clarifier. The NaCl solids from the dissolvers are
centrifuged and discarded and the brine recycled to the dis-
solver. The KCl-rich brine from the clarifier passes to
a series of vacuum crystallizers where it is cooled and KC1
crystallizes out. The KCl crystals are separated from the
brine by filtration or centrifugation, washed, dried, screened,
and stored. The NaCl-rich, KCl-depleted brine is reheated
and recycled back to the dissolvers. The wash solution is
combined with the brine and recycled.
The waste streams from process are the clay slines, the NaCl
tailings, and a bleed stream from the brine recycle system
which is used to control the soluble impurities in the brine.
The clay slimes and salt waste contain some KCl-rich brine.
Both wastes are normally washed with water to recover the
KCl and the wash solution fed to the brine recycle system.
To maintain the process water balance, the water used in
the wash steps is usually limited to the amount of water
removed in the brine bleed and that lost by evaporation.
The quality of the KCl product produced by the solution-
recrystallization process can vary depending on the NaCl
content of the brine. Normally the process is controlled
to give a soluble grade product (99% KCl).
Processes for Recovering KCl from Natural Brines
Recovery of KCl from the natural brine at Bonneville, Utah,
utilizes a combination of solar evaporation and flotation-
beneficiation.U4>223 The brine, which is a complex mixture
of metal chloride and sulfates (see page 38), is concentrated
by solar evaporation in a scries of ponds. In the first
pond the brine is concentrated with the crystallization of
NaCl. Evaporation is continued until the KCl begins to pre-
cipitate. The brine is then transferred to a second pond
v/here it is concentrated with the deposition of a mixture
of NaCl and KCl. Evaporation is continued until IlgCl, begins
to precipitate out with the KCl in the form of carnallite
(KCl-MgCl2*6I!20}. The brine is then pumped to a third pond
to precipitate NaCl, carnallite, and cainite (KCl-MgSO, • 311,0) .
The brine is next transferred to another pond to recover
the remaining magnesium.
The NaCl-KCl mixture from the second pond is collected mechani-
cally and sent to a refinery to separate the two materials.
A flotation beneficiation flowsheet, similar to Figure 26,
is used to recover the KCl. The process is somev:hat less
complex than that used to process sylvinite since the KC1-
NaCl mixture is relatively pure. The precipitates from the
-------
third pond containing NaCl, KCl-MgCl-«6II2O and KCl-MgSO^-3H?0
are washed with brackish water and the brine recycled back
to the various ponds depending on composition.
The processing of Bonneville brine presents no major problems
with regard to pollution since any aqueous waste stream
can be recycled back to the brine source.
Recovery of KC1 from the Searles Lake brine is part of a
very complex series of operations designed to recover a variety
of products from the brine. A number of excellent references
describing the Searles Lake operation are available in the
literature.^14' ''24' As with the Bonneville brines,
pollution control is not a major problem since any aqueous
waste stream can be recycled back to the brine source.
85
-------
AQUEOUS PROCESS EFFLUENTS
The fertilizer industry is quite complex and a variety of
types and grades of fertilizer materials are manufactured.
The bulk of the aqueous waste load from the fertilizer indus-
try, however, results from the preparation of a relatively
few major fertilizer products or intermediates. Production
of these materials has been analyzed based on the use of
standard or typical processes. The aqueous waste streams
from each process have been characterized as thoroughly as
possible. The specific operations in each process which
generate aqueous waste have been identified. Considerable
information is available on the principal contaminants found
in each waste stream and their pollutional effects.
Defining the volumes and compositions of the various waste
streams is more difficult. Water recycle, minor process
variations, and operating philosophies can result in a wide
range of waste stream compositions and volumes for a given
product between the different fertilizer plants. Wherever
possible, an attempt has been made to relate the waste stream
volumes to a unit quantity of product—usually on a gallons
per ton basis. Normally a range of flows must be specified
because of the wide variations between plants. The contam-
inant concentrations given for the various waste streams
show the same wide range of values. In many instances it
is extremely difficult to predict with any accuracy even
wide ranges of waste stream volumes and composition. For
example, in the cleanup of process spills aqueous waste streams
are generated. It is almost impossible to quantitatively
relate the volume or composition of these cleanup wastes
to the fertilizer product.
In analyzing the aqueous wastes generated by the fertilizer
industry one fact quickly becomes evident. The principal
pollutants found in the aqueous waste from most fertilizer
plants are inorganic materials. Except in a few instances,
organic contaminants are relatively unimportant. Therefore,
pollutional criteria such as BOD and COD are less important
in evaluating fertilizer plant effluents than is the case
with most industrial waste.
CLASSIFICATION OF AQUEOUS EFFLUENTS
There are in general four types of aqueous effluent streams
which can be generated in a fertilizer plant:
1. cooling water
2. steam condensate
87
-------
3. process effluents
4. sanitary waste.
If any of the incoming water to the plant is purified prior
to use then an additional waste stream from the purification
plant may also be generated. For example, when ion exchange
is used to demineralize water, regeneration of the resin
bed results in waste streams with high anion or cation con-
centrations which must be discharged. The handling and treat-
ment of these types of waste streams are well documented
in the literature; they have not been considered in this
study.
In almost any plant the sanitary waste stream is kept separate
from other aqueous wastes. The problems involved in handling
and treating sanitary waste from a fertilizer plant are no
different from those encountered in any industrial plant.
They are usually discharged to a municipal sewage system,
cesspool system or an on-site sewage plant. Technology for
processing sanitary wastes is well defined and has not been
considered in this study.
The major portion of the water used by the fertilizer industry
is for cooling purposes. This water is usually relatively
low in contaminants and is usually kept separate from the
more contaminated aqueous waste streams. One exception to
this is in wet process phosphoric acid manufacture where
cooling water is frequently combined with gypsum pond water.
Unless water is readily available, it is standard practice
to use cooling towers or ponds and recycle the cooling water.
The principal contaminants found in the cooling water comes
from three sources: leaks in process equipment and piping;
the buildup of dissolved solids from the feed water, and
chemicals added to the cooling water to control biological
growth, scale, and corrosion.
When cooling water is recycled using towers or ponds, natural
evaporation is used to cool the water. Since the makeup
water to the system always contains some dissolved solids,
there will be a buildup of dissolved solids in the recirculated
cooling water. Blowdown of the cooling water system is nec-
essary to control these solids. The blowdown stream will
be contaminated with not only the dissolved solids but also
with the materials leaked from the process and chemical addi-
tives. The materials which leak from the process can include
lubricants from pump and compressor bearings and any of the
materials present in the process as feeds, intermediates,
products, or by-products. The chemicals added to the cooling
88
-------
water can include both organics and inorganics. Chromates
and phosphates are frequently added for corrosion control
and organics for control of biological growth.
Another problem which must be considered in dealing with
cooling water is thermal pollution. Most states place restric-
tions on the temperature of aqueous streams which are discharged
to the environment. Means must usually be provided for cool-
ing any aqueous waste streams which exceed the prescribed
limits. When cooling water is used on a once-through basis
a cooling pond is usually provided for reducing the tempera-
ture of the stream prior to discharge into surface water.
The blowdown from recirculated cooling water systems is usually
relatively cool and does not need additional cooling prior
to discharge.
Steam condensate is another aqueous waste stream that is
normally quite low in contaminants. It is usually kept sep-
arate from highly contaminated waste streams. Process equip-
ment leaks are the principal means by which contaminants
are introduced into the condensate. The condensate may or
may not be recycled. When the condensate is recycled some
water is lost and dissolved solids can build up unless demin-
eralized makeup water is used. When demineralized water
is used, regeneration of the demineralizer will generate
a waste stream containing dissolved solids. If blowdown
of the steam condensate system is necessary, the blowdown
water will contain high levels of dissolved solids, materials
leaked from the process, and any chemicals added to control
scale formation and foaming.
On an overall industry basis, aqueous process effluents re-
present the major source of highly contaminated waste streams.
Process water accounts for only about 20-25% of the water
used by the fertilizer industry but contains the bulk of
the contaminants generated by the various processes. Aqueous
waste streams can normally be divided into five general classes.
1. By-product streams
2. Scrubber solutions (from gas scrubbing equipment)
3. Process spills
4. Wash solutions (from equipment cleanup)
5. Barometric condenser water.
All fertilizer processes will generate one or more of the
above streams.
89
-------
Despite the many different products manufactured by the fert-
ilizer industry the number of aqueous by-product waste streams
generated by the various processes are relatively few. How-
ever, the streams which are generated are usually large in
volume and are highly contaminated. A typical example is
the gypsum slurries resulting from wet process phosphoric
acid production.
Scrubber solutions represent a sizable and growing source
of contaminated water in the fertilizer industry. As air
pollution control regulations become more restrictive, scrubb-
ing of gaseous waste streams to remove pollutants becomes
more prevalent. In the fertilizer industry most scrubbing
operations are carried out using water or aqueous solutions.
This results in the formation of aqueous waste streams which
must be treated prior to discharge. The concentration of
pollutants in the aqueous waste can vary over wide ranges
depending on gas stream concentrations and on scrubber solu-
tion recycle. In some cases the water scrub is used on a
once-through basis and the concentration of pollutants in
the aqueous waste is relatively low. The stream may or may
not require treatment prior to discharge. In other cases
the scrub solution is recycled and the pollutant concentration
can build up to high levels, limited only by the equilibrium
between gas and aqueous phase concentrations. In normal
operation a side stream of scrubber solution is continuously
removed to control the concentration in the scrub recycle
below the equilibrium level. The side stream is usually
quite high in pollutants and must be treated prior to discharge.
A typical example of scrub solution wastes in the fertilizer
industry is the solution used to scrub fluoride-containing
gaseous effluents from superphosphate reactors and dens.
Recovery of fluoride from various gas streams by aqueous
scrubbing represents the major source of aqueous waste streams
in most phosphate fertilizer operations.
In any processing operation spills occur, regardless of the
care used in process control and maintenance. These spills
can be solid or liquid depending on the particular operation.
It is more or less standard practice in the fertilizer indus-
try to cleanup such spills using water or acid. The spilled
material is usually washed to a sump where it can be collected
for recycle to the process or subsequent discharge. The
pollutants associated with spill cleanup are the product
material itself, processing intermediates, or reactants.
In most large, well-run processing plants, spills account
for only a small percentage of the waste load (<1%) . In
small plants such as bulk blend or liquid mix plants, which
have small processing waste loads, spills can account for
a substantial portion of the total plant waste load.
90
-------
In certain processing operations it is sometimes necessary
to shut down various pieces of equipment for cleanup. The
cleanup operation can result in the generation of an aqueous
wash solution which may be contaminated with the various
materials used in or resulting from the processing operation.
In the bulk blend and liquid mix plants, where a variety
of fertilizer formulations are produced in the same equipment,
cleanup produces a major portion of the aqueous waste load
for the plant. Since the wash solution is normally not re-
cycled back to the process it must be discharged to the environ-
ment.
There are many operations in the fertilizer industry, such
as evaporation, drying, or process reactions, which generate
condensable gas streams—principally water vapor. These
condensable gas streams frequently contain volatile contami-
nants such as unreacted feed materials (e.g., ammonia). If
the contaminant concentrations in the condensables are high
then indirect cooling is normally used for condensation.
The condensate is returned to the process or treated to remove
the contaminants prior to discharge. In many cases the contam-
inants in the condensable stream are comparatively low and
a barometric condenser is used for condensation. This results
in a large aqueous waste stream containing low levels of
contaminants. By using a cooling tower or pond the barometric
condenser water can be recycled in which case the contaminant
can build up to substantial levels. If water is readily
available, the barometric condenser water is normally used
on a once-through basis.
AQUEOUS EFFLUENT VOLUMES AND COMPOSITIONS
In the production of the various major fertilizer materials,
each process will generate one or more of the several waste
streams described above. Table 4 lists the potential aqueous
waste streams for each product under consideration assuming
the typical processes described earlier are used. In every
process, spills will occur which can lead to an additional
waste stream. However, it is impossible to define the magni-
tude and concentrations of such streams. In a phosphate fer-
tilizer complex, waste streams generated by spills are usually
discharged to the acid plant gypsum pond where they become a
part of the overall waste water system for the complex. In
ammonia plant complexes the spill waste is usually combined
with the other highly contaminated waste streams. The magni-
tude and composition of the other waste streams generated by
the various processes are described in the following sections:
Unfortunately, in many plants specific information on individ-
ual waste streams is not available since the individual streams
are not monitored. Instead, most plants monitor only the total
waste load of the combined contaminated waste stream.
91
-------
TABLE 4
PROBABLE AQUEOUS WASTE STREAMS FROM FERTILIZER PROCESSING
Product
Ammonia
Ammonium nitrate
Ammonium sulfate
Potential Waste Streams
Process condensate
Cooling water
Steam condensate
Cooling water
Steam condensate
Off-gas scrubber solutions
Condenser water
Condenser water
Off-gas scrubber solutions
Urea Cooling water
Steam condensate
Off-gas scrubber solutions
Calcined phosphate rock Off-gas scrubber solution
Wet process phosphoric acid Cooling water
Gypsum slurry
Off-gas scrubber solution
Acid sludge
Steam condensate
Furnace phosphoric acid
Normal superphosphate
Triple superphosphate
Ammonium phosphate
Ammoniated superphosphate
Potassium chloride
(sylvinite)
Potassium chloride
(brine)
NPK fertilizers
Liquid blends
Bulk blends
Cooling water
Sludge
Phossy water
Off-gas scrubber solution
Off-gas scrubber solutions
Off-gas scrubber solutions
Off-gas scrubber solutions
Off-gas scrubber solutions
Recycle brine bleed
Salt-slimes tailings
Recycle brine bleed
Salt tailings
Off-gas scrubber solutions
Equipment washdown
Equipment washdown
*Waste streams generated by spill cleanup not included
92
-------
Ammonia Production
In the production of ammonia the principal contaminated waste
streams are process condensates, blowdown from cooling water
systems and wash solutions. Boiler water blowdown may also
contribute to the waste load.
When synthesis gas is prepared by steam reforming of natural
gas at high pressures/ various steps in the process require
cooling or quenching of the gas stream. When the gas is
cooled, water condenses and the condensate must be drained
from the system. Whenever possible the condensate is returned
to the process. However/ it is impossible to return all
of the condensate to the process and that which is not recycled
must be disposed of. For example, when the synthesis gas
is cooled and compressed before entering the synthesis cir-
cuit a knockout pot is used to collect condensed water and
oil from the gas stream, and this stream must be disposed
of. The condensate streams will contain C02 and some mono-
ethanolamine (an aqueous monoethanolamine (MEA) solution
is used to strip the CC^ from the synthesis gas and some
MEA enters the gas stream). Any process leaks and spills
and plant cleanup are normally washed to a sump system and
collected. They and the process condensates are usually
combined to give a single waste stream. Specific information
on the individual waste stream volumes and compositions is
limited. Most plants do not monitor the individual streams,
but only the combined stream. Typical composition and volume
ranges for the combined waste stream are presented in Table 5.
Ammonia production requires extremely large volumes of cool-
ing water. Volume requirements are reported to range from
20,000-185,000 gallons per ton of ammonia product. For a
600 ton per day ammonia plant this corresponds to a cooling
water flow of 8,000-74,000 gpm. When the cooling water is
used on a once-through basis contamination of the cooling
water effluent is normally not a problem, except for thermal
pollution. The water is normally discharged to the receiving
water without treatment except for temperature reduction.
The large volumes of cooling water required make once-through
operation impractical in most cases. Normally the cooling
water is recycled using cooling towers or ponds. The number
of cycles of concentration that can be used depends primarily
on the purity and availability of the fresh makeup water.
In most cases the number of cycles of concentration will
be in the range of 3-7. The volume of blowdown required
to control impurities is usually in the range of 1-2% of
the cooling water flow. Makeup water volume will usually
be several times the blowdown volume, because of water lost
by evaporation and windage in the cooling tower or pond.
93
-------
TABLE 5
TYPICAL AMMONIA PLANT WASTE WATER VOLUME AND COMPOSITION
Contaminant
Ammonia
CO-
MEA
mfc
20-100
150-750
50-100
Contaminant
BOD
COD
Oil
50-150
60-200
100-10,000
Waste Water Volume 100-1000 gal/ton NH3
*Does not include cooling water or boiler water blowdowns
94
-------
The volume of blowdown water from an ammonia plant cooling
water system can vary over a wide range and values from 200-
3500 gallons per ton of product have been reported. Normally,
however, the blowdown volume will run from 400-1000 gallons
per ton NH,. The composition of the blowdown water will
depend on the feed water composition, cycles of concentration,
process leaks, and materials added to the cooling water for
corrosion, scaling and plant growth control. Typical ranges
of concentration for various impurities are given in Table 6.
TABLE 6
COMPOSITION OF AMMONIA PLANT COOLING WATER BLOWDOWN
Contaminant
Chromate
Phosphate
Zinc
Heavy metals
Fluoride
Biocides
Misc. organics
mg/&
0-300
0-50
0-30
0-60
0-10
0-200
0-100
Contaminant
NH3
MEA
Sulfate
TDS
BOD
COD
Oil
mg/t
10-100
0-10
500-5000
500-10,000
10-300
15-400
10-1000
Blowdown volume 400-1000 gal/ton NH3
Ammonia production requires substantial quantities of steam
and process waste heat boilers are used to generate this
steam whenever possible. In some plants additional steam
generation may be required. Condensate recycle to the steam
boiler can lead to generation of a boiler water blowdown
waste which is high in dissolved solids and corrosion and
scale inhibitors.
Ammonium Nitrate Production
In the production of prilled ammonium nitrate there are var-
ious operations which can result in the generation of gas
streams which can contain ammonia or nitrate ion or both.
These operations include the neutralizer, where both ammonia
and nitric acid may be vaporized, the concentrator, and
95
-------
the prilling tower. Normally scrubbing is required to remove
the nitrogen species from the gas streams. Water scrubbing
is usually used and the scrub water may or may not be recycled.
The ammonia and nitrate recovered in the scrub system is
normally not returned to the process. The volume of nitrogen
containing waste water generated is difficult to define because
of the many operating variables involved. However, ammonia
losses can run as high as 1% or about 4 pounds per ton of
NH4NO3 product. Nitrate losses are usually much less.
Significant quantities of cooling water may be required in
ammonium nitrate production (up to 35,000 gallons per ton
of product). When large volumes are required cooling water
recycle is usually used. This means cooling water blowdown
is required. The volume of blowdown required will depend
primarily on makeup water purity. The composition of the
blowdown stream will be quite similar to that generated in
ammonia production, except that there will be a significant
nitrate concentration in the blowdown from ammonium nitrate
production.
Process spills and resultant cleanup can generate a signifi-
cant waste stream. Solid spills are usually recovered if
possible. Liquid spills, however, are usually washed to
a sump or drainage system. Normally the wash solutions are
not returned to the process because of the possibility of
oil or grease contamination and its resultant hazard.
Ammonium Sulfate Production
In the direct production of ammonium sulfate, ammonia and
sulfuric acid are reacted in a crystallizer to form the sul-
fate. Heat of reaction causes the evaporation of water from the
crystallizer solution and a loss of ammonia and sulfate.
The ammonia normally must be removed from the gas stream
prior to discharge. This can be done using a water scrubber,
barometric condenser, or noncontract condenser. Where a
water scrubber or barometric condenser is used, a water flow
of 1-5 gallons per minute per daily ton of product is required.
For a 300 ton per day plant this will correspond to a water
flow of 300-1500 gpm. If the water is used on a once-through
basis the ammonia concentration will vary between 10-100
mg/£ dpending on the crystallizer design. (The sulfate con-
centration will be less.) If the water is recycled the ammonia
concentration will be much higher. Where indirect condensa-
tion is used the water and ammonia condense to give a stream
which can be returned to the process.
The offgas from the ammonium sulfate dryer contains ammonium
sulfate particles and small amounts of ammonia. The parti-
culates are collected by dry means such as cyclones and
96
-------
returned to the process. Depending on local regulations,
recovery of the ammonia from the gas stream may or may not
be necessary. When recovery is required water scrubbing
is used. The ammonia concentration in the scrub solution
is quite low. If water is used on a once-through basis scrub
volumes vary from 60-600 gallons per minute for a 300 ton
per day plant. The ammonia concentration is usually less.
than 10 gm/i.
Process spills are usually washed to a retention sump with
a minimum of water. Frequently the liquor from the sump
is used as the wash solution. The sump liquor is normally
returned to the process.
Urea Production
There are several possible aqueous waste streams which may
be generated in a urea plant. They include cooling water
or cooling water blowdown, steam condensate or blowdown,
and waste water used to scrub ammonia from offgas streams.
The production of urea can require large volumes of water
for cooling. Values from 0-85,000 gallons per ton of urea
have been reported. Most plants, however, use 10,000-40,000
gallons per ton of product. The once-through urea processes
require the least amount of cooling, while total recycle
processes require the largest volumes. Normally the volumes
of cooling water required make once-through cooling water
use impractical. When once-through cooling is used, however,
the impurities in the effluent water are low. Treatment
prior to discharge is normally not required except for ponding
to reduce the temperature. The cooling water is normally
kept separated from other waste streams. When the cooling
water is recycled impurities build up and blowdown is required.
As in any cooling water recycle system the number of cycles
of concentration is normally determined by the purity and
availability of the feed water. The number of cycles of
concentration used in urea production usually varies from
3-7 as in most recycle cooling water systems. The concen-
trations of impurities in the blowdown stream can vary over
wide ranges but will be quite similar to those given for
the blowdown from ammonia plant cooling water systems. The
only significant difference is that urea plant blowdown streams
will contain small amounts of urea (up to 50 mg/O, and little
or no MEA. The volume of the blowdown stream can vary from
50-2000 gallons per ton of product with the usual range being
200-800 gallons per ton.
Urea production requires significant volumes of steam (up
to 8,000 pounds per ton of product). Steam condensate used
97
-------
on a once-through basis presents no disposal problem since
the condensate is very low in contaminants. Blowdown from
recycle boiler water systems can be a problem. Normally
10-20 cycles of concentration can be used because treated
makeup water is usually used. Boiler water blowdown will
normally amount to 5-10% of the boiler water. The blowdown
volume will range from 20-100 gallons per ton of urea. The
typical range of impurity concentrations are shown in Table 7.
TABLE 7
COMPOSITION OF UREA PLANT BOILER WATER BLOWDOWN
Contaminant mg/A
Phosphate 5-50
Zinc 0-10
Heavy metals 0-10
Sulfite 0-100
Suspended solids 50-300
Contaminant
EDTA
Misc. organics
TDS
Alkalinity
Hardness
0-50
0-200
500-3500
50-700
50-500
Several gas streams are generated in urea production which
can contain ammonia. Air pollution regulations may require
that these gas streams be scrubbed for ammonia removal prior
to discharge. In the total recycle process inert gases build
up in the recycle system and must be purged. The purge stream
will contain small amounts of ammonia and carbon dioxide.
The urea solution from the last decomposer stage will contain
small amounts of ammonia and possibly carbon dioxide. In
the concentrator this ammonia will be evolved with the water
vapor. In addition, any biuret formed will result in additional
ammonia release. Ammonia can also be released due to hydrolysis
of urea in the concentrator. As a result, the vapor stream
from the concentrator can contain significant quantities
of ammonia. Ammonia releases as large as 3-4 pounds per
ton of product are possible. Additional ammonia release
is possible in the prilling towers and the air stream exiting
the prilling tower may contain small amounts of ammonia.
When the ammonia loss is high, as from the concentrator,
ammonia recovery and recycle to the process is possible.
However, even when ammonia recovery is practiced, considerable
ammonia containing waste volumes may still be generated due
to gas scrubbing requirements.
98
-------
Process spills can generate significant volumes of waste water.
Wash solutions used to clean up spills are usually combined
with other contaminated waste streams and not recycled to
the process.
Little information is available on the individual contaminated
waste streams from urea production since most plants do not
analyze the individual streams. Compositions of the combined
waste streams (exclusive of cooling water and boiler water blow-
down) are available and typical compositions and volumes are
presented in Table 8.
TABLE 8
TYPICAL UREA PLANT WASTE WATER VOLUME AND COMPOSITION*
Contaminant
Ammonia
Urea
co2
mg/&
200-4000
50-1000
100-1000
Contaminant
Oil
BOD
COD
mg/&
10-100
30-300
50-500
Volume 50-2000 gal/ton urea
*Does not include cooling water and boiler water blowdown
Aqueous Wastes From Nitrogen Fertilizer Plant Complexes
The aqueous process waste streams generated by the produc-
tions of nitrogen fertilizers are quite similar in composi-
tion. In industrial complexes where two or more of the
nitrogen fertilizers are produced the aqueous process wastes
are usually combined and handled as a single stream for treat-
ment and disposal. Table 9 summarizes the typical range
of compositions and volumes of the individual waste streams
which could be encountered in a nitrogen fertilizer pro-
duction complex. Cooling water and boiler water blowdown
may also be combined with the process waste streams after
chromate removal/ but have not been included in the numbers
given in Table 9.
Phosphate Rock Calcination
When phosphate rock is calcined, fluoride is evolved. The
evolved species are silicon tetrafluoride and hydrogen fluoride,
The ratio of HP to SiF4 increases as the calcining temperature
is increased. Even at low calcination temperatures, however,
99
-------
TABLE 9
COMPARISON OF AQUEOUS PROCESS WASTE STREAMS FROM NIGROGEN
FERTILIZER PLANTS
Aqueous Process Waste Streams
Ammonia
Plant
Volume 100-1000
(gal/ton product)
Contaminants
(mg/jO
NH3
co2
NO5
SOj
MEA
Urea
Oil
BOD
COD
20-100
150-750
50-100
Ammonium
Nitrate
Plant
50-1200
200-2000
50-1000
Ammonium
Sulfate
Plant
Urea
Plant
**,
100-10,000 50-2000
10-1000
**
5-500
**
200-4000
100-1000
100-10,000
50-150
60-200
<20
<20
<20
<20
50-1000
10-100
30-300
50-500
* Does not include cooling water and boiler water blowdown
** Wide range due to possible recycle of aqueous scrubber
solutions
100
-------
hydrogen fluoride is the principal fluoride species. The
amount of fluoride evolved depends on the calcining temper-
ature , and Figure 28 shows typical fluoride evolution as
a function of temperature. Fluoride evolution does not begin
below about 250 °C and the rock must be fused to remove more
than 90% of the fluoride.
The fluoride evolved is scrubbed with water or dilute hydro-
fluoric acid. Since very little SiF4 is normally present,
the scrubber product is principally hydrofluoric acid:
HF(g) + H20 H, H
To effect the required fluoride removal a small flow of fresh
or neutralized scrubber water must be used in the final stage
of the scrubber. Recycle scrub solution can normally be
used in the first stage of scrubbing. If gypsum pond water
is available this can be used for scrubbing. If pond water
is not available (or recovery of the fluoride is desired)
the scrub water is recycled. The concentration of hydrofluoric
acid in the scrub solution builds up, and a bleed stream
is required to maintain the acid concentration at an accept-
able level. The maximum acceptable acid concentration will
depend on a number of factors; primarily gas stream fluoride
concentration and scrubber design.
The bleed stream can be used to recover salable fluoride
products. If not, then the bleed stream must be treated
for fluoride removal prior to discharge.
The total fluoride evolved will depend on the calcining con-
ditions. For a 3.5% rock the maximum fluoride evolution
would be 65-67 pounds per ton of rock treated. Fusion would
be required to release this amount of fluoride.
Wet Process Phosphoric Acid Production
In the production of wet process phorphoric acid by the di-
hydrate process the gypsum by-product (CaSO4-2H2O) represents
a major disposal problem. The quantity of gypsum formed
is about 1.5 tons per ton of rock processed which corresponds
to about 4.5-5 tons per ton of P2°5 product.
In a few cases the gypsum from the acid plant filters is
slurried with water and pumped into the ocean for disposal.
Usually, however, the slurried gypsum is pumped to a holding
pond (normally called a gypsum pond) where the gypsum settles
out to give a clear supernate. The gypsum pond usually is
a diked area which may be several hundred acres in area.
101
-------
100
80
O
o;
o:
o
60
40
a
i—i
o
20
500
1000
1500
CALCINING TEMPERATURE, °C
FIGURE 28 FLUORIDE EVOLUTION DURING CALCINATION OF PHOSPHATE ROCK
102
-------
Two general rules of thumb are that (1) the pond area should
be at least 0.5 acres per daily ton of P2°5 Production, and
(2) the deposited gypsum accumulates at the rate of about
one acre-ft per year per daily ton of P2°5« Normally all
the contaminated waste streams in the acid plant are fed
to the gypsum pond. In phosphate fertilizer complexes the
waste stream from other processes are usually discharged
to the gypsum pond as well. The clear supernate from the
gypsum pond is recycled back to supply most of the water
requirements for the acid plant and other processes. In
well-run acid plants recycled pond water supplies over 80%
of the gross water requirements of the plant. As the pond
water is recycled, dissolved impurities build up until an
equilibrium composition is reached. The typical range of
concentrations for the principal impurities in the pond water
(at equilibrium) are given in Table 10. During periods of
heavy rainfall the pond water composition may vary somewhat,
but not to any great degree.
TABLE 10
TYPICAL EQUILIBRIUM COMPOSITION OF GYPSUM POND WATER
Contaminant Concentration, mg/A
P205 6000-12,000
Fluoride 3000-5000
Sulfate 2000-4000
Calcium 350-1200
Ammonia 0-100
Nitrate 0-100
pH 1.0-1.5
Some fresh makeup water is required for certain process opera-
tions such as cooling compressors and cooling the tank where
the sulfuric acid is diluted prior to entering the digester.
The trend, however, is to reduce fresh water requirements to
a minimum.
In areas with little rainfall the evaporation of water from
the gypsum pond area makes the discharge of pond water to the
environment unnecessary. In most areas some pond water must
103
-------
be discharged. The discharge volume is normally dependent
on rainfall and will vary widely during different times of
the year for a given plant. In some localities, such as
Florida, the volume of water discharged from the pond is
very close to the volume of rainfall collected in the pond
during the rainy season. Discharge from the pond of a 600
ton P2C>5/day acid plant can run as high as 5000 gpm during
periods of high rainfall. Over a year period the discharge
should average 500-1000 gallons/ton of P2°5- In many acid
plants the gypsum pond discharge is the only plant waste
stream requiring treatment.
The soluble fluoride in the pond water, at equilibrium, will
have a significant partial pressure. Slack(20) reports the
following values:
pSiF^ = 0.00007 mm Hg
p IIP 3s 0.0008 mm Hg
This results in a considerable loss of fluoride to the atmos-
phere on windy days (up to 200 pounds per day from a typical
gypsum pond). However, the fluoride concentration in the
air is only about 1 ppb.
Large volumes of cooling water are required in wet process
phosphoric acid production. Fresh or salt water is sometimes
used for cooling on a once-through basis, and the discharge
water contains very few contaminants. The water is usually
discharged to the receiving water without treatment, except
for cooling and pH monitoring. In most cases the cooling
water is recycled, and in most plants the gypsum pond water
is used for most cooling purposes. Fresh water for cooling
is restricted to certain specific purposes as mentioned earlier.
In some plants the cooling water circuit is kept separate
from the pond water circuit. Cooling ponds or towers are
used, the cooling water is recycled, and only the cooling
water blowdown is sent to the gypsum pond.
Because of its pH and composition, the gypsum pond water
is quite corrosive and special materials of construction
are required when the pond water is used for cooling.
When phosphate rock is reacted with sulfuric acid to make
phosphoric acid the fluoride in the rock is also attacked
by the acid and freed. The distribution of the fluoride
between the various process streams can vary significantly
depending on the equipment design used. Table 11 shows the
fluoride distribution normally encountered in the dihydrate
process.
104
-------
TABLE 11
FLUORIDE DISTRIBUTION IN WET PROCESS PHOSPHORIC ACID PRODUCTION
% of Total Fluoride in Rock
Typical
Fluoride Disposition Range Typical Values ERGO
Evolved from Reactor
and Filters 4-7 5 5.5 5
Evolved from Concentrator 35-45 40 41.9 60-65
Remaining in Gypsum 25-30 30 27.8 20-25
Remaining in Acid Product 20-30 25 24.8 10
Reference (20) (26) (20) (27)
Air pollution control regulations require that the fluoride
evolved from the reactor, filters, and concentrator be re-
moved before the gas stream is discharged to the atmosphere.
The fluoride is usually scrubbed from the gas streams using
water or dilute fluosilicic acid. When the fluoride concentra-
tion in the gas stream is high, as in the water vapor-fluoride
stream from the acid concentrator, surface condensers are
sometimes used. The fluoride is evolved as silicon tetrafluo-
ride and hydrogen fluoride. Silicon tetrafluoride is the
predominant species in the vapors from the reactor and filters.
The vapor from the concentrator contains both HF and
When silicon tetrafluoride is scrubbed it reacts with the
water to form fluosilicic acid and silica:
3 SiF4 + 2 H20 -*• 2 H2SiFg + SiO2
When a mixture of SiF4 and HF is scrubbed the HF and SiF4
combine in the scrub medium to form H2SiFg but no silica
is formed if the HF-SiF2 ratio is 2:1. If the ratio is greater
than 2:1 free HF is present in the scrub solution. If the
ratio is less than 2:1 some silica is formed.
If the fluoride is not to be recovered for sale, gypsum pond
water is used to scrub the fluoride from the gas stream.
The scrubber solution recycles to the gypsum pond and accounts
for a major portion of the fluoride that builds up in the
pond water system. Fluoride recovery in the scrubbers usually
exceeds 95% and in some localities air pollution control
105
-------
regulations make greater than 98% recovery necessary. The
partial pressure of SiF4 and HF in the pond water makes it
difficult to reach these recovery levels and scrubber design
can be quite complex. In some instances a small fresh water
stream is used in the final stage of scrubbing to reduce
the fluoride in the exit gas to an acceptable level. This
stream is then used for earlier scrubbing stages, or is sent
directly to the gypsum pond.
A great deal of effort has gone into the development of suit-
able fluoride scrubbers and a number of references on scrubber
design are available in the literature.(28»2^/30/3D
When economics dictate, fluoride may be recovered as fluosilicic
acid for sale or for conversion to other fluoride products.
(See Chemical and Engineering News - April 19, 1971 - p. 18
for a discussion of U.S. fluoride market.) When recovery
is practiced the fluoride is scrubbed from the gas stream
using a dilute solution of fluosilicic acid. The acid solution
is recycled and the concentration can build up as high as
25% HoSiFg. A continuous side stream is removed from the
recycle system as product to maintain the acid concentration
at the desired level. Normally, a two stage scrubber system
is used since the recycle acid solution will not reduce the
fluoride in the exit gas to the desired level. The fluosil-
icic acid solution is used in the first stage scrubbing and
pond water or fresh water is used in the second stage. Fluo-
ride recovery for sale is usually practiced on the gas stream
from the concentrator. In most cases greater than 80% of
the fluoride is recovered for sale from the gas stream, and
most of the remaining fluoride enters the gypsum pond water
system.
The fluoride concentration in the gas stream from the reactor
and filters is quite low and fluoride recovery is less prac-
tical. When recovery is practiced the fluosilicic acid con-
centration in the scrubber solution rarely exceeds 10%. Over-
all fluoride recovery for sale is normally less than 50%
of that in the gas stream.
The product acid (52-55% P2°s) from the concentrator normally
contains 1-3% precipitated solids, principally calcium sulfate
and metal fluosilicates. With time, additional precipitates
form which consist primarily of complex metal phosphates.
They usually amount to about 2-5%. On storage these precipi-
tates settle out to form a sludge.
When the phosphoric acid is used on site the sludge does
not present a problem and is usually not removed from the
acid. When the acid is to be shipped, however, the sludge
can be a major problem since it can accumulate in shipping
containers or interferes with the customer's use of the acid.
106
-------
Washing of the sludge from the shipping containers can create
a significant waste stream. Some plants attempt recovery
of the sludge from the wash solutions, but normally it is
discharged to the gypsum pond if one is available.
The preferred approach when acid is to be shipped is to store
the acid in tanks for a substantial period of time. The
sludge settles out and is removed as an underflow from the
tanks by means of rakes. The recovered sludge, which contains
substantial amounts of phosphate, is usually disposed of
by blending into a fertilizer, usually triple superphosphate.
Significant quantities of steam are used in wet process phos-
phoric acid production. In many plants the steam is used
on a once-through basis. Uncontaminated steam condensate
is discharged to the receiving waters without treatment.
Steam condensate which is contaminated, such as that from
barometric condensers, and vacuum ejectors, is discharged
to the gypsum pond. When steam condensate is recycled only
the uncontaminated portion is returned to the steam plant.
In the phosphoric acid plant spills occur and they are usually
cleaned up by washing to a sump system from which they are
moved to the gypsum pond.
Furnace Phosphoric Acid Production
The principal waste stream generated in the production of
furnace grade phosphoric acid is a waste water stream con-
taminated with elemental phosphorus, called phossy water.
The phossy water can originate from three principal points
in the process: phosphorus condensers, the phosphorus storage
area, and phosphorus sludge handling.
The elemental phosphorus vaporized from the electric furnace
passes through dust collectors and is then condensed with
water in one or more condensers. The resulting mixture of
liquid phosphorus and water is collected in a sump. The
liquid phosphorus settles to the bottom of the sump, while
the water forms a cover layer over the phosphorus. Water
from the cover layer is recycled back to the condenser system.
Impurities will build up in the water and a bleed is necessary
to control the impurity level. The impurities consist princi-
pally of colloidal phosphorus particles and soluble fluoride
scrubbed out of the gas stream in the condenser.
When the phosphorus is condensed, most of the phosphorus
collects in a relatively pure form. However, some of the
phosphorus collects in a heterogeneous mixture called phos-
phorus sludge, which is composed of phosphorus, water and
inorganic solids. About 5-10% of the phosphorus produced
ends up in the sludge.
107
-------
As a general rule/ about 5% sludge is produced for every
1% of inorganic solids in the condensed phosphorus.'20)
solids in the sludge are composed of 40-60% SiC>2/ 5-15% CaO,
2-4% Fe203, 1-3% A12C>3, and 2-5% P2°5« They will normally
amount to between 1 and 5% in the crude phosphorus.(2^) The
sludge can be separated from the clean phosphorus by gravity
difference. When this is done the separated sludge is diluted
with water and sent to a centrifuge to separate the contained
phosphorus. The water discharged from the centrifuge is
a second source of phossy water.
The liquid phosphorus is stored under water to prevent oxida-
tion. When the phosphorus is moved, water must be added
to or removed from the storage tanks. This water is contami-
nated with phosphorus and is the third potential source of
phossy water.
Usually, the various sources of phossy water are combined
and handled as a single waste stream. The composition and
volume of the phossy water can vary significantly. Typical
range of compositions and volume are shown in Table 12.
TABLE 12
TYPICAL PHQSSY WATER COMPOSITION AND VOLUME
Contaminant mg/ft Contaminant mg/&
Phosphorus 1000-2000 Fluoride 100-1000
Phosphate 50-500 Suspended 2000-4000
Solids
pH 3-6
Volume 1000-2000 gal/ton P2O5
When slag is tapped from the furnace, some fluoride, P2°s/
and S02 may be evolved from the molten slag. If the slag
is air cooled the evolution of gas is quite low, amounting
to about one pound of total fluoride, SOp and P2°5 Per ton
of P2°5- When the slag is water quenched fluoride evolution
increases and amounts to about 2.5 pounds of fluoride dis-
charge per ton of P2°5 product. If the slag is treated to
expand it, the fluoride evolution will increase to about
8 pounds per ton of P2°5 product.(32) When the slag is air
cooled scrubbing of the fumes may not be necessary. In other
cases, scrubbing will usually be required in order to meet
air pollution control regulations. Fluorides are usually
108
-------
scrubbed from the fumes with water; and the water discharging
the scrubber can contain up to several thousand ppm fluoride
and P2° ^ t^ie scrubber water is recycled.
Preparation of the phosphate rock for charging to the furnace
may require agglomeration by calcination. This can lead
to the generation of a fluoride-containing gas stream which
must be scrubbed for fluoride removal. The scrubbing oper-
ation and waste stream generated are similar to those dis-
cussed in the previous section on phosphate rock calcination.
The preparation of phosphoric acid via elemental phosphorus
requires large volumes of cooling water (25,000-45,000 gallons
per ton P2C>5 produced) . If water is readily available it
can be used on a once-through basis. The once-through cool-
ing water effluent is normally low in impurities and can
be discharged to the receiving waters without treatment except
for reduction of the temperature to an acceptable level.
When water is in limited supply the cooling water must be
recycled. The accumulation of impurities requires a periodic
blowdown of the cooling water. The cycles of concentration
used normally range from three to seven depending on the
purity of the makeup water. The composition of the blowdown
water can vary over wide ranges. Typically , the range of
impurities in the water are as shown in Table 13.
TABLE 13
COMPOSITION OF FURNACE GRADE PHOSPHORIC ACID PLANT COOLING
WATER BLOVfDOWN
Impurity mg/& Impurity mg/A
Fluoride 10-500 SO2 10-100
Phosphate 10-1000 TDS 500-10,000
Sulfate 500-5000 Oil 0-10
Chromate 0-100 Other Organics 0-100
Heavy metals 0-20 Suspended solids 50-200
NH3 0-5
PH 5.5-6.0
Nitrates 0-5
Cooling water blowdown volume 250-1000 gal/ton
109
-------
Normal Superphosphate Production
In the production of normal superphosphate gaseous fluorides
(principally silicon tetrafluoride) are evolved from various
steps in the process. The principal points of fluoride evolu-
tion are the mixer, den, and storage area. Overall fluoride
evolution will vary between 10-50% of the fluoride in the rock.
Assuming a 3.5% fluoride concentration in the rock, the fluo-
ride evolved will be between 4-20 pounds per ton of product.
The amount of fluoride evolved will depend primarily on the
concentration of the sulfuric acid used. The higher the
acid concentration the greater the fluoride evolution. The
average fluoride evolution is about 35% or approximately
14 pounds per ton of product for a rock containing 3.5% fluo-
ride. Typically, the fluoride initially in the rock would
be distributed as follows:
% of Fluoride in
Fluoride Disposition Original Rock
Evolved in mixer 2
Evolved in den & conveyors 30
Evolved in storage (curing) 3
Retained in product 65
The fluoride concentration in the offgas streams from the
mixer den and conveyors is relatively high, and they are
usually combined and scrubbed in a single scrubber system.
The gases evolved in the storage building are usually re-
moved from the building by air circulation. The fluoride
concentration in the air stream is normally quite low making
fluoride removal by scrubbing impractical. Recent improve-
ments in building design reduces the air flow required and
has made scrubbing for fluoride removal practical.(20)
The fluoride is scrubbed from the gas stream with water or
dilute fluosilicic acid. Some particulate matter is pre-
sent in the offgas and it plus the silica formed by the re-
action of SiF4 and water complicate scrubber operation.
It is difficult to define the composition of the solution
exiting the scrubber. Scrubber design, water source, and
ultimate fluoride disposal all affect the type and volume
of scrub solution used. When the normal superphosphate plant
is operated in conjunction with a wet process phosphoric
acid plant gypsum pond water can be used as the scrub solution
which is then recycled back to the pond after use. In some
110
-------
instances a small fresh water scrub, low in fluoride, is
used in the final scrub stage to reduce the fluoride in the
exit gas to an acceptable level. This stream also recycles
to the gypsum pond or is used in first stage scrubbing.
When pond water is not available, then a holding system .is
necessary so the scrub liquor can be recycled (normally the
scrub water cannot be used on a once-through basis because
of the potential for fluoride contamination of water supplies).
The concentration of fluosilicic acid in the recycle water
is allowed to build to substantial levels. Makeup fresh
water is used in the last stage of scrubbing. The acid solu-
tion can be recovered for sale as 20-25% fluosilicic acid.
When the acid is not recovered for sale a fraction of the
recycle liquor must be bled off to maintain the fluosilicic
acid concentration. The makeup fresh water replaces the
acid removed as a bleed or for sale. The bleed stream repre-
sents the principal aqueous effluent from the normal super-
phosphate plant which operates independent of an acid plant.
This stream will carry from 4-20 pounds of fluoride per ton
of product. The concentration of fluoride in the stream
can vary over a widerange depending on the scrubber design
but normally will be at least several percent. The volume
can vary from 10-200 gal/ton of product. By treating the
bleed stream for fluoride removal it can be recycled as makeup
water or released to the environment.
Ammonium fluoride solution can also be used for fluoride
scrubbing. (33) The silicon tetrafluoride reacts with the
ammonium fluoride to form ammonium fluosilicate:
SiF4 + 2 NH4F •»• (NH4)2SiF6
The process has the advantage that no silica is formed. How-
ever, the process has not been used to any extent commercially.
In general, the problems of fluoride scrubbing in a normal
superphosphate plant are quite similar to those encountered
in a wet process phosphoric acid plant. Scrubbing liquors
represent the only aqueous waste stream from most normal
superphosphate plants/ except for spill cleanup. Solid spills
are usually recovered if possible. Liquid spills are washed
to a sump for collection and sent to the gypsum pond if a
pond is available.
Triple Superphosphate Production
Gaseous fluorides are evolved in the production of triple
superphosphate. Fluoride evolution is more constant than
111
-------
in normal superphosphate production and varies between 32-40%
of the fluoride in the rock. For a 3.5% fluoride rock this
corresponds to a fluoride evolution of 10-12 pounds per ton
of product. In run-of-pile triple superphosphate production
fluoride is evolved principally in the mixer and conveyor
belt (or den} with a minor amount evolved in the storage
area. Slack (20) reports the fluoride distribution for HOP
triple superphosphate production to be as follows:
% Total Fluorine
Fluoride Disposition Initially in Rock
Evolved at mixer, belt and 31.4-33.3
transfer conveyors
Evolved in storage 2,0-2.8
Retained in product 63.9-66.6
In addition, when the triple superphosphate is granulated
after curing, scrubbing of the fumes from the dryer may be
required as small amounts of fluoride may be evolved.
When granular triple superphosphate is made directly from
rock and phosphoric acid, fluoride evolution occurs princip-
ally in the reactor, mixer, dryer, cooler, and screens; the
total offgas must be scrubbed for fluoride removal.
Almost all triple superphosphate plants are operated in con-
junction with a wet process phosphoric acid plant. There-
fore, unless the fluoride is to be recovered for sale, the
scrubber solution is principally pond water plus some fresh
water, both of which are returned to the acid plant gypsum
ponds after use. Scrubber operation is similar to that en-
countered in normal superphosphate production, and the waste
stream compositions are similar. Fluoride added to the gypsum
pond will amount to about 10-12 pounds per ton of product.
Fluoride can be recovered for sale by using a dilute fluosilicic
acid scrub solution, as in normal superphosphate production,
if economics justify the recovery.
Ammonium Phosphate Production
In the preparation of ammonium phosphate substantial quantities
of ammonia are volatilized from the neutralizer and granula-
tor. Process economics require that the ammonia be recovered.
This is normally done by scrubbing the gas stream from the
neutralizer-granulator with phosphoric acid. The phosphoric
acid scrub combines with the incoming process acid which
feeds the neutralizer. A portion of the mixed acid stream
recycles back to the scrubber. Scrubbing removes most of
112
-------
the ammonia from the gas stream. However, the phosphoric
acid scrub solution contains 1-3% fluoride, and optimum scrubber
operation for ammonia recovery results in the stripping of
some of the fluoride from the acid. The result is a gas
stream containing substantial amounts of fluoride and a small
amount of ammonia. The fluoride volatilized ranges from
0.1 to 0.3 pounds per ton of product.
Air pollution control regulations usually require that the
fluoride be scrubbed from the gas stream. Scrubbing is usually
accomplished with water. The scrubber flow required varies
from 500-3000 gallons per ton of product; the fluoride con-
centration in the exit scrub water ranges from 25-250 mg/1.
The ammonia concentration in the scrub water varies from
5-50 mg/1. Scrub solution can be recycled with a resulting
increase in contaminant concentrations.
When the ammonium phosphate plant is operated in conjunction
with a wet process phosphoric acid plant the waste scrub
solution can be discharged to the gypsum pond. Pond water
can be used as the scrub solution.
Some ammonia is also evolved from the dryer. The gas from
the dryer also contains substantial amounts of particulate
matter. The solids are removed in a cyclone separator and
the gas stream is scrubbed with phosphoric acid to recover
the ammonia. The scrubbing can occur in the same unit used
to scrub the gas from the neutralizer and granulator or in
a separate scrubber. Where a separate scrubber is used,
a separate tail gas scrubber for fluoride removal may also
be required. The scrub solution from this tail gas scrubber
would then be combined with that from the other fluoride
scrubber.
Process spills are washed to a sump system with water and
returned to the process if possible. Otherwise they are
normally transferred to a holding pond and then discharged.
If a gypsum pond is available the wash solution may discharge
to it.
Ammoniated Superphosphate Production
In the production of granular ammoniated superphosphate ammonia
and dust particles are evolved from the ammoniator-granulator
and dryer The dust is removed from the gas stream with
cyclones.' The ammonia is then removed by scrubbing the gas
with recycled water. To control the ammonia concentration
in the recycled scrub water at the required level a portion
of the scrub is bled off and returned to the process. The
bleed is usually mixed with the ammoniating solution or
113
-------
returned directly to the ammoniator. When the scrub is not
recycled to the process the scrub solution may be used on
a once-through basis. The ammonia concentration is quite
low and the solution is discharged to a holding pond or re-
ceiving waters. Gypsum pond water can be used as the scrubb-
ing medium when available.
When spills occur they are usually washed to a sump system
with a minimum of water. The preferred practice is to recycle
the material from the sump to the process. However, in some
plants the material is discharged to a pond or receiving
water. The composition of the wash solution is difficult
to predict but usually contains substantial amounts of ammonia
and phosphate.
Granular NPK fertilizers can be produced in equipment similar
to the equipment used for ammoniated superphosphate production.
Waste stream problems are similar to those encountered in
ammoniated superphosphate production. Waste streams generated
from process spills are more complex since they may contain
potassium, chloride, sulfate, nitrate, and urea in addition
to ammonia and phosphate.
Potash Production
The production of potassium chloride by flotation beneficia-
tion generates three waste streams: a clay slimes waste,
a solid sodium chloride waste, and a brine bleed for impurity
control. The sylvinite ore may contain up to 1.5% insoluble
clay materials. The clay is separated fm the NaCl-KCl mix-
tures as a slime. The slime is washed to remove KC1, dewateredf
and discharged to a tailing dump. On a dry weight basis,
the clay slimes accumulate at the rate of 30-80 pounds per
ton of KC1 product depending on the clay content of the ore.
In some cases the sodium chloride from the process may be
sold. Normally, however, the sodium chloride is dewatered
to recover the KCl-containing brine and then discharged to
the tailings dump. Transfer of the NaCl to the dump is usually
as a saturated water slurry. The NaCl discharged can amount
to as much as 1.5 tons per ton of KC1 product depending on
the NaCl-KCl ratio in the sylvinite ore. in some plants
the clay slimes and sodium chloride waste are combined and
sent to a single tailings dump.
In order to control the level of impurities in the recycle
brine solution it is necessary to bleed off a portion of
the brine from the system on a continuous basis. The dis-
carded brine is saturated with sodium chloride and potassium
chloride. It will also contain lesser amounts of magnesium,
114
-------
calcium, and sulfate, as well as small amounts of organic
materials which are added to the brine in the flotation circuit.
It is the presence of the magnesium, calcium, and sulfate
which makes the bleed necessary. The bleed volume is held
to a minimum to reduce the KC1 loss. The volume of bleed
required depends on the concentration of the critical impurities
in the sylvinite ore, and can range as high as 100 gallons
per ton of product.
In the production of potassium chloride from sylvinite ore
by solution-recrystallization the same three waste streams
are produced. The only difference is that the brine bleed
for the solution-recrystallization process will not contain
the organics used in the flotation process.
In the production of potassium chloride from brines at Bonne-
ville and Searles Lake any aqueous waste streams generated
are not a problem. They are simply returned to the brine
source.
Liquid and Bulk Blend Fertilizers
In the normal operation of liquid fertilizer and bulk blend
fertilizer plants no aqueous waste streams are generated
by the processes themselves. The wastes from these plants
are generated either by process spills or by washdown of
equipment. Where possible, solid spills are collected and
returned to the mixing equipment. Liquid spills are washed,
usually with water, to a sump system. Contaminated water
from equipment washdown is also sent to a sump system. it
is impossible to define the composition of the aqueous wastes
collected in the sump. However, it will normally contain
ammonia, potassium, nitrate, chloride and phosphate, and
possibly sulfate and urea. The volume of waste generated
is equally difficult to define since it will depend on the
frequency of equipment washdown required and on the care
used in controlling process spills.
115
-------
ENVIRONMENTAL EFFECTS OF AQUEOUS EFFLUENTS
EFFECTS OF AQUEOUS EFFLUENTS ON RECEIVING WATERS
Aqueous effluents from fertilizer plants which discharge
into receiving waters can have three basic effects on the
water environment. They can affect:
• aesthetic or recreational value of the receiving water
• renovation or reuse of the receiving water
• biological and physiological changes in organisms in
the receiving water or those using the water.
In most cases these effects are interrelated. Wastes which
reduce the aesthetic or recreational value of water can also
affect water organisms and water reuse.
It is impossible to define quantitatively the aesthetic value
of a particular water environment. However, it is easy to
recognize the effect of uncontrolled waste discharge on such
an environment. Floating material, oil slicks, scum, turbidity,
color, odor, excessive plant growth, all serve to detract
from the beauty of the water source. The ultimate aesthetic
value of a water environment will depend on the value which
mankind places on maintaining the environment in a clean
and natural state.
There is an increasing awareness in the United States of
the need to maintain an adequate supply of water for recrea-
tional purposes. This has resulted in initiation of long-
range planning to supply these recreational needs. The National
Technical Advisory Committee on Water Quality Criteria^34'
has made the following recommendation with regard to recrea-
tion water used in secondary contact recreation activities:
"Surface waters should be suitable for use in
secondary contact recreation—activities not
involving significant risks of ingestion—
without reference to official designation of
recreation as a water use. For this purpose,
in addition to aesthetic criteria, surface
waters should be maintained in a condition
to minimize potential health hazards by
utilizing fecal coliform bacteria.
Surface waters, with specific and limited excep-
tions, should be of such quality as to provide
for the enjoyment of recreational activities
based upon the utilization of fishes, xvaterfowl,
and other forms of life, without reference to
official designation of use.
117
-------
Species available for harvest by recreation users
should be fit for human consumption. In areas
where taking of mollusks is a recreational
activity, the criteria shall be guided by the U.S.
Public Health Services manual "Sanitation of Shell-
fish Growing Areas," 1965 revision."
For recreational waters used in primary contact activities (swim-
ming, diving, skiing, etc.) the following recommendations were
made:
"Fecal coliforms should be used as the in-
dicator organism for evaluating the microbio-
logical suitability of recreational waters.
As determined by multiple tube fermentation
or membrane filter procedures and based on
a minimum of not less than five samples
taken over not more than a 30 day period,
the fecal coliform content of primary contact
recreational waters shall not exceed a log
mean of 200/100 ml nor shall more than 10%
of total samples during any 30 day period
exceed 400/100 ml.
In primary contact recreation waters the
pH should be within the range of 6.5-8.3
except when due to natural causes and in
no case shall be less than 5.0 nor more
than 9.0. When the pH is less than 6.5 or
more than 8.3, discharge of substances
which increase the buffering capacity of
the water should be limited.
For primary contact recreation waters,
clarity should be such that a Secchi disc
is visible at a minimum depth of four
feet. In 'learn to swim1 areas the
clarity should be such that a Secchi disc
on the bottom is visible. In diving
areas the clarity shall equal the min-
imum required by safety standards, de-
pending on the height of the diving
platform or board.
In primary contact recreation waters,
except where caused by natural condi-
tions , maximum water temperature
should not exceed 30 °C."
118
-------
The overall objectives of the committee's recommendations
are to insure that water quality management programs for
recreation will include criteria to:
1. Provide for and enhance general recreational use
of surface waters.
2. Enhance recreation value of waters designated for
recreation use.
3. Provide special protection for the recreation user
where significant body contact with water is involved.
If these criteria are to be met careful control must be main-
tained over waste waters discharged into recreational xvaters.
Waste streams discharged into receiving waters can have a
significant affect on the subsequent use or renovation of
the water. Dissolved organics, dissolved inorganics, and
toxic materials can seriously interfere with subsequent util-
ization of receiving waters for public water supplies, animal
life, and irrigation. Dissolved materials can cause corrosion
problems, scaling and undesired side reactions in waters
used for industrial purposes, and require expensive water
purification prior to use.
Industrial water quality criteria vary over wide ranges.
Some industries can get by with very poor quality water,
while others, such as the brewing industry, require high
quality water. In general, any waste water dumped into re-
ceiving waters will impair the usefulness of the water to
subsequent industrial users.
The same general statement can be made with regard to the
effects of waste water on aquatic organisms. Any pollutants
added to the receiving water can change the natural environ-
mental conditions of the water and affect aquatic organisms.
If these changes are excessive, serious damage to the organisms
may result. To minimize these effects careful control must
be maintained over waste waters discharged to the environment.
PARAMETERS FOR EVALUATING AQUEOUS EFFLUENTS
There are a number of factors which must be considered in
evaluating the effects of aqueous effluents from fertilizer
plants which discharge to the environment. These include
the normal or conventional pollution parameters such as pH,
BOD, temperature, etc. In addition, there are a number of
other contaminants, principally inorganic ions, which must
also be considered on an individual basis. These include
plant nutrients such as ammonia and phosphate, and potentially
119
-------
toxic materials such as fluoride and elemental phosphorus.
A list of the major factors to be considered in evaluating
fertilizer plant aqueous effluents is given below. Each
of these factors and its effect on the receiving waters is
discussed in the following sections.
Conventional Pollution Parameters
pH - Alkalinity - Acidity
Biochemical Oxygen Demand (BOD)
Chemical Oxygen Demand (COD)
Total Oxygen Demand (TOD)
Total Organic Carbon (TOC)
Dissolved Oxygen (DO)
Dissolved Solids
Turbidity
Color
Temperature
Coliform
Major Inorganic Contaminants
Phosphate
Ammonia
Nitrate
Fluoride
Potassium
Sulfate
Chloride
Phosphorus
Suspended Solids
120
-------
Minor Inorganic Contaminants
Chromates
Heavy Metals
Principal Organic Contaminants
Urea
Monoethanolamine
Lubricating Oils
The limits set by the regulatory agencies on various contami-
nants and pollution indicators in receiving waters vary
significantly depending on the anticipated use of the re-
Reiving waters. They also vary from state to state. It is
impossible, therefore to define a standard limit for the
various parameters which is applicable to all areas of the
United States. Typical of the water quality standards used
are those of the State of Florida which are presented in
Part in Table 14.
Conventional Pollution Parameters
The pii of an aqueous solution is defined as the negative
logarithm of the hydrogen ion concentration. The pH scale
ranges from 0 to 14 and a pH of seven represents a neutral
solution. When the pH value is less than seven a solution
is acidic, while a pH value greater than seven indicate an
alkaline solution. Most natural fresh waters have a pH close
to eight and aquatic life/ both animal and plant, exists
^ost effectively in water that is near this value. Most
aquatic life can tolerate a pH range from 5-9 without serious
consequences. If the pH value of a water environment is
outside this range for more than brief periods, the mortality
rate of most aquatic life will be high. Therefore, care
ttust be taken to control the pH of waste streams discharged
to the environment. Most state water quality standards set
the pH range for fresh water systems between pH 6.5 and pH
8«5.
The pH is an indication of, but not a measure of, the alkaline
°r acidic materials present in a water environment. Acidity
and alkalinity are reciprocal terms. Total acidity is defined
as the amount of standard base required to bring a water
sample to pii 8.3. Total alkalinity is the amount of acid
121
-------
TABLE 14
TYPICAL LIMITS FOR CONVENTIONAL POLLUTION PARAMETERS
The following general water quality standards were set b.y
the Florida Air and Water Pollution Control Commission.
28-5.05 Water quality standards; specifies. -
(1) The criteria of water quality herein-
after provided will be applied only after rea-
sonable opportunity for mixture of wastes with
receiving waters has been afforded; the rea-
sonableness of the opportunity for mixture of
wastes and receiving waters shall be determined
on the basis of the physical characteristics
of the receiving waters and the methods in which
the discharge is physically made shall be ap-
proved by the regulatory agency.
(2) The following water quality standards
shall be the criteria for pollution when con-
centrations exceed following limitations:
pH - of receiving waters shall not be
caused to vary more than one (1.0) unit
above or below normal pH of the waters;
and lower value shall be not less than
six (6.0), and upper value not more
than eight and one-half (8.5).
Dissolved Oxygen - shall not be artifi-
cially depressed below the values of
four (4.0) ppm.
BOD - shall not be altered to exceed
values which would cause dissolved
oxygen to be depressed below the limit
listed above and, in no case, shall it
be great enough to produce nuisance
conditions.
Dissolved Solids - not to exceed five
hundred (500) mg per liter as a monthly
average or exceed one thousand (1000)
mg per liter at any time.
Turbidity - shall not exceed fifty (50)
Jackson units as related to standard
candle turbidimeter above background.
122
-------
TABLE 14 (Continued)
Temperature - shall not be increased
so as to cause any damage or harm to
the aquatic life or vegetation of the
receiving waters or interfere with
any beneficial use assigned to such
waters.
Oils and Greases - shall not exceed
fifteen (15) mg/1, or that no visible
oil, defined as iridescence, be pre-
sent to cause taste and odors, or in-
terfere with other beneficial uses.
Detergents - shall not exceed one-half
(0.5) mg/1.
Fluorides - for waters not used for
public water supplies, shall not exceed
10.0 mg/1 as fluoride ion or will not
interfere with other beneficial uses.
Chlorides - chlorides shall not ex-
ceed two hundred fifty (250) mg/1 in
streams considered to be fresh water
streams; in other waters of brackish or
saline nature the chloride content
shall not be increased more than ten
per cent (10%) above normal background
chloride content.
Chromium - shall not exceed 0.50 mg/1
hexavalent or 1.0 mg/1 total chromium
in effluent discharge and shall not
exceed 0.05 mg/1 after reasonable mix-
ing in the receiving stream.
*Abstracted from the rules of the Florida Air and Water
Pollution Commission Chapter 28-5, Supplement #62.
123
-------
required to bring a water sample to pH 4.5. Thus, a water
sample can be both alkaline and acidic at the same time.
Strong acids are toxic if they drop the pH below 6.0 but
do not appear to be toxic above a pH of 6.0. Strong bases
do not appear to be toxic below pK 9.0. Weak acids can be
toxic at a pH value considerably above pH 6.0 because of
the anion or the compound itself; weak bases can be toxic
at pH values below 9.0 due to the cation or unionized mole-
cule. The additions of any materials which lowers the pH
below 6.0 or above 9.0 can be harmful to aquatic life.
Biochemical oxygen demand (BOD) is defined as the oxygen
required for oxidation of soluble organic matter by bacterial
action in the presence of oxygen. As such it is a measure
of the biodegradable organics present in a waste stream.
If the biochemical oxygen demand of a waste stream raises
the BOD of the receiving water excessively it can result
in the reduction of the dissolved oxygen in receiving water
to a point where aquatic life suffers deleterious effects.
A high BOD can be an indication of the presence of toxic
organic materials in a waste stream. However, it is gener-
ally necessary to know the particular organics present to
determine the toxic effects of a given waste stream.
Chemical oxygen demand (COD) is defined as the oxygen required
to chemically oxidize the organic compounds present in a
waste water stream. The COD is a measure of most, but not
all, of the organic compounds which may be dissolved in a
waste- stream. It is an important parameter since it measures
the nonbiodegradable organics in the waste water as well
as those which are biodegradable. A high COD for a waste
stream can result in a reduction in the dissolved oxygen
in the receiving water and a resultant deleterious effect
on aquatic life. As is the case with BOD, COD is no measure
of the toxic affects of specific organic materials.
The total oxygen demand (TOD) of a waste stream is defined
as the oxygen required for the oxidation of all carbon, hy-
drogen, and nitrogen in the waste and partial oxidation of
the sulfur compounds in the waste. TOD is a recently developed
pollution parameter and has not been widely used. However,
it is a valuable tool for evaluating the oxygen requirements
for a given waste stream. As is the case with BOD and COD,
a high TOD in a waste stream indicates the strong possibility
of a decrease in dissolved oxygen in the receiving water
and subsequent damage to aquatic life.
124
-------
T°tal organic carbon (TOG) is a measure of all the organic
m^terials present in a waste water stream. In combination
with COD and BOD it is the best measure of the nonbiodegrad-
fkle organic materials in the waste. The TOG measurement
ls a relatively new one and quantitative TOG data for most
waste streams is limited. A high TOG indicates high potential
*°r low dissolved oxygen.
°xygen is required to sustain aquatic life. The dissolved
°Xygen content of a waste stream or receiving waters is,
therefore, an important parameter in determining the pollu-
tional effects of the particular water source. The solu-
bility of oxygen in water is temperature dependent and ranges
j-tom 11.3 mg/1 at 10 °C to 5.6 mg/1 at 50 °C. In warm waters,
ftesh water organisms require a continuous dissolved oxygen
c°ntent of at least 5 mg/1 to sustain good growth. I-34' in
c°ld water, fresh water organisms require a higher dissolved
Ojfygen (>6 mg/1) for good growth. Salt water organisms
aPparently can exist for long periods of time on a lower dis-
solved oxygen than is possible with fresh water life. As a
general rule, any time the dissolved oxygen drops below 4 mg/1
"°r any period of time all aquatic life will be adversely
affected.
dissolved solids, as the name implies, is a measure of
solid material dissolved in a water sample. It does not
, however, soluble materials such as vegetable oils,
oils, and petroleum oils which are volatile or extract-
with organic solvents. It is, therefore, essentially a
re of the dissolved inorganic materials in a water sample,
Principally mineral salts. Kigh dissolved solids in a waste
5tream can have several effects on the receiving waters . They
impart a mineral taste to drinking water and possibly pro-
physiological effects in humans and animals. High inin-
salt concentrations in drinking water have resulted in
?astrointestinal symptoms, wasting disease, and sometimes death
** livestock. The United States Public Health Service set a
of 500 mg/1 dissolved solids in public water supplies.
dissolved solids may contain materials which are toxic at
low concentrations. These toxic materials must be deter-
by additional tests, and the limits for public water
Applies are set for specific materials. The 500 mg/1 dissolved
solids limit assumes the individual dissolved solids are rela-
tively nontoxic and are harmful only at high concentrations
where their osmotic effect in organisms becomes excessive.
dissolved solids can cause corrosion of metals and materials
of construction. They can cause scaling problems in cooling
^ater systems, and can make more frequent blowdowns necessary
£& cooling water and steam generation systems which recycle
their water.
125
-------
Turbidity is a measure of the suspended material in water, and
is determined by the opacity or light scattering caused by the
suspended matter. Turbidity in a waste stream can have a
number of harmful effects on the receiving waters. It can have
an adverse effect on the aesthetic quality of the water. High
turbidity can reduce the light reaching aquatic plant growth
thus reducing photosynthesis. Settling of suspended matter
can cause damage to stream, river, and lake bottoms which
interfere with fish spawning and adversely affect other animal
life. Long term exposures to high levels of suspended solids
(>100 mg/1) can have a very serious effect on plant and animal
life. At 270 mg/1, suspended solids were reported to cause
a high percentage of deaths in rainbow trout,f35) and concen-
trations above 80 mg/1 are unlikely to support good fish
growth.<36) Suspended solids in irrigation water can cause
excessive plugging and wear in irrigation equipment and re-
strict water flow. Deposits of suspended solids on irrigated
land can cause crusting which interferes with seed emergence.
Suspended solids in water used for industrial purposes may
make filtration or coagulation necessary prior to use.
Color in waste water can have three principal effects on
the receiving waters. Color can adversely affect the aesthetic
value of the water; reduce the light reaching plant life;
and is objectionable in drinking water. Excessive color
in water to be used in public water supplies must be such
that it can be removed by the water treatment plant. The
limit set by the United States Public Health Service is 75
color units. Excessive color can reduce the aesthetic and
recreational value of water not only from the color itself
but by reducing the water clarity. Photosynthesis by plant
life depends on the light reaching the plants. A minimum
of 5% of full sunlight is required to maintain the oxygen
balance in water by photosynthesis and it is estimated that
many plants require at least 25% of full sunlight for maximum
photosynthesis.<34) Color in the water can serve to reduce
the light reaching the plants at lower depths and thus limit
photosynthesis.
Thermal pollution of water sources must be considered a serious
problem. High temperature waste waters can raise the temp-
erature of receiving waters with several deleterious effects
resulting. Most aquatic organisms require oxygen to maintain
life. This requirement is normally supplied by oxygen dis-
solved in the water. The solubility of oxygen in water de-
creases with increasing temperature. Therefore, if the water
temperature becomes too high the dissolved oxygen may be
reduced to the point where growth of aquatic organisms is
adversely affected. Chemical and biochemical reactions which
occur in water increase in rate when the temperature increases.
126
-------
Therefore, the oxidizable materials will be reacted in a
shorter stream length at higher temperatures. This can lead
to an oxygen depletion in the water which could adversely
affect the growth of aquatic organisms. All aquatic organisms
grow best within specific temperature ranges. Any changes
in temperature which exceed the optimum range can adversely
affect organism growth. The optimum temperatures can vary
widely for the many different aquatic species of plant and
animal life. Table 15 shoxvs the maximum and minimum temp-
eratures for several varieties of fish acclimated to specific
temperatures. 137,38)
The optimum temperature ranges for aquatic plants vary depend-
ing on the species. For instance, different algae species
grow best at different temperatures. Blue-green algae grow
best at 35-40 °C, green algae at 30-35 °C/ and diatoms at
18-20 °C. (39) Long term changes in water temperature can
lead to changes in the predominant algae species in a water
environment.
TABLE 15
(37 18 i
TEMPERATURE RANGE FOR FRESHWATER FISHES* y<
Species Maximum* Acclimated to Minimum
,$£ Hr¥ PC"»£giFi"
Large mouth bass 32 72 20 5 24
Bluegill 31 60 15 3 24
Channel catfish 30 24 15 0 24
34 24 25 6 24
Brook trout 23 133 3
25 133 20
^Values are LD5Q temperature tolerance limit. 50% of fish
survived water temperature.
Coliform bacteria are the recommended indicator of the pre-
sence of fecal matter in water. Fecal coliform organisms
in water are indications of recent fecal pollution. Nonfecal
coliform organisms may be due to less recent fecal pollution,
soil runoff water, or fecal pollution containing only those
organisms. Water containing human or animal waste is dang-
erous because of the possible presence of pathogenic micro-
organisms. These include members of the salmonella and
127
-------
shigella genera, vibrio coruma, endamoeba histolytica, and
enteric viruses.* ' Coliforms, although they come from
the same source as the pathogens, serve as harmless indicators.
In general, the presence of coliforms is an indication of
possibly dangerous pollution. As long as sanitary waste
is kept separate from other fertilizer plant waste streams,
it does not represent a major problem to the fertilizer in-
dustry. Such streams are treated in the same manner as any
other sanitary waste.
Inorganic Contaminants
Inorganic materials are the principal contaminants found
in fertilizer plant waste waters, and represent the major
water pollution problems facing the fertilizer industry today.
The rapid increase in production of ammonia and ammonia-
based fertilizers has resulted in the generation of large
volumes of aqueous wastes contaminated with ammonia. The
technology required for reducing the ammonia in the waste
to the desired levels is not well developed and the processes
that are available are costly.
Ammonia in waste water can have a number of deleterious effects
when discharged to the receiving water:(41)
1. Ammonia can contribute to rapid algae growth which
can ultimately lead to eutrophication.
2. Ammonia can be toxic to fish.
3. Ammonia can restrict waste water renovation and reuse.
4. Ammonia can have a detrimental effect on the disin-
fection of water supplies.
5. Ammonia can be corrosive to some metals and materials
of construction.
6. Ammonia can have a detrimental effect on water odor
and taste.
7. Ammonia can exert an oxygen demand.
Ammonia is a nutrient for plant growth. Considerable experi-
mental evidence shows that nitrogen, as ammonia, can contri-
bute greatly to the rapid growth of algae. In general, ammoni-
acal nitrogen is assimilated more.rapidly by plant life than
any other form of nitrogen.t4!/42' Very rapid algae growth
due to an excess of nutrients can lead to eutrophication.
Considerable controversy exists among reputable scientists
as to the role of nitrogen (and ammonia) in the problem of
eutrophication. Until the controversy over the cause(s)
of eutrophication is resolved, nitrogen in waste water in
any form, and particularly as ammonia, should be considered
as a likely cause of eutrophication and treated accordingly.
128
-------
Ammonia in water can be toxic to fish. Lethal ammonia expo-
sures for various species are reported to vary from 0.2-
5.0 mg/1. (43' The toxicity is dependent primarily on nonionic
ammonia and unassociated ammonium hydroxide. Therefore,
PH has an important effect on ammonia toxicity.^4' Field
tests show that when the pH is 8 or above the ammonia content
as nitrogen should not exceed 1.5 mg/1. (34> At levels above
1 mg/1 ammonia inhibits the ability of hemoglobin to combine
with oxygen and fish may suffocate. (34 ,44)
When water is disinfected with chlorine any ammonia present
in the water will react with the chlorine to form chloroamines .
The chloroamines are bactericidal but are slower acting and
less effective than chlorine. <45) Chloroamine concentrations
as low as 0.4 mg/1 are reported to be toxic to fish. I46)
Ammonia water can be quite corrosive to certain metals. Copper
and zinc alloys are rapidly attacked even at low ammonia
concentrations. ^7, 48) Ammonium salts also attack concrete
made from Portland cement. (43)
Most states do not at present have a specific water quality
standard for ammonia. For those states which do have ammonia
standards, they vary from 0.02 to 3.5 mg/1 as nitrogen. I41'
In states without ammonia standards ammonia can be included,
implication, in those standards for toxic substances.
Phosphorus in the oxidized state is a plant nutrient and
essential for all forms of plant growth. When other condi-
tions are favorable, additions of low phosphate concentrations
to an aqueous system can lead to rapid growth of algae and
other vegetation. Many experts consider phosphorus and nitro-
gen to be the principal causes of excessive aquatic plant
growth which leads to eutrophication. The available phosphorus
Required for algae growth is reported to be quite low. Con-
centrations ranging from 0.01-0.05 mg/1 are reported to be
sufficient for growth. ^ ' Available phosphorus is dependent
°n the total phosphorus concentration.
Phosphorus in the oxidized state is not normally considered
toxic to animal life. As long as the phosphate concentration
is less than 100 mg/1, its presence in public water supplies
is not considered a problem. Phosphates can cause scaling
Problems in water used for steam generation and cooling pur-
Poses. At concentration greater than 0.5 mg/1 they interfere
With normal coagulation and settling processes in water pur-
ification plants. (34)
Elemental phosphorus must be considered quite toxic to animal
life. Elemental phosphorus dissolves in water only to the
129
-------
extent of 3 mg/1. However, when water contacts phosphorus,
colloidal phosphorus particles may be formed which raise
the phosphorus concentration drastically. The lethal dose
for elemental phosphorus in humans is reported to be 100
mg and the chronic dose is 1 nig/day. (49' In-take of elemental
phosphorus causes bone and liver damage and a chronic poison-
ing resulting in progressive weakness and anemia. (50,51)
Elemental phosphorus concentrations as low as 0.05 mg/1 were
reported to be toxic to fish when the exposure time was 163
hours. <52)
Elemental phosphorus in waste waters represents a potential
fire hazard. Evaporation of waste water containing colloidal
phosphorus particles can allow the phosphorus particles to
become dry. Oxidation of the dry particles can result in
a fire if additional combustible materials are present.
Nitrate ion is a nutrient for aquatic plant growth, and as
such can be considered a potential cause of eutrophication.
Assimilation of nitrate by plants is, in general, less rapid
than the uptake of ammonia. Nitrate ion can be toxic to
humans and ruminants. It is reported to cause methomoglobinemia
(nitrate cyanosis) in babies if the nitrate concentration
in water exceed 70 mg/1. (51,53) Tne united States Public
Health Service sets the maximum nitrate content for drinking
water at 45 mg/1. Nitrates are also reported to cause meth-
emoglobinemia in cattle and pigs. However, fairly high con-
centrations (>1000 mg/1) are required. (34^
Soluble fluorides present in waste water must be considered
to be toxic to animal and plant life. Fluorides in drinking
water can cause serious problems. Continuous ingestion of
water containing 2-3 mg/1 of fluoride is reported to cause
mottled teeth, and greater concentration can cause severe
fluorosis and even death. (51) Deaths have been reported ,~.\
from the continuous use of water containing 13 mg/1 fluoride.
The United States Public Health Service has set the maximum
limit for fluoride in drinking water at 1.7 mg/1. Fluoride
concentration greater than 1.5 mg/1 are reported to be toxic
to fish, ' although other studies indicate concentrations
of 50 mg/1 only stunt growth and reduce reproduction. '20'
There is some evidence that fluorides can accumulate in cer-
tain organisms which makes potential fluoride poisoning of
animal life more likely. Aquatic plant life do not appear
to be adversely affected by concentrations of soluble fluorides
of less than 50 ppm. Control authorities of the State of
Florida report that at pH values above 6.0 a total fluoride
content up to 20 mg/1 is not harmful for water not used for
human consumption, and Florida water quality regulations
set a limit of 10 mg/1 fluoride in water not used to supply
drinking water requirements. '2^)
130
-------
Potassium is not considered to be especially toxic to animal
life. As long as the concentration in drinking water does
not exceed the limit for total dissolved solids (500 mg/1)
potassium should not be a hazard to human life. The United
States Public Health Service has not set a limit for potas-
sium in drinking water. Potassium is a nutrient required
for plants. As such, potassium in waste waters can serve
to accelerate plant growth. If potassium is limited in a
given aquatic environment, then potassium added to the system
through waste water could possibly result in excessive and
objectionable plant growth.
Sulfate ion in water in limited amounts (<500 mg/1) is not
considered hazardous to animal life. High concentrations
in drinking water can have a laxative effect on humans , I")
and high concentrations (2104 mg/1) are reported to cause
a progressive weakening and death of cattle.^5' The United
States Public Health Service has set a limit for sulfate
in drinking water of 250 mg/1. Sulfur is a secondary nutrient
Required for plant growth. Therefore, sulfate ion in waste
water serves as a nutrient for aquatic plant growth. Sulfate
ion in water used for cooling purposes or steam generation
can be a potential cause of scale. Normally, however, the
sulfate concentration must be quite high to be a problem.
High sulfate in boiler or cooling water which is recycled
causes more frequent blowdown and a larger blowdown waste
stream.
Chloride in waste water is not a particular problem if the
concentration does not exceed 500 mg/1. The United States
Public Health Service has set the chloride limit for drinking
water at 250 mg/1. The principal objection to chloride in
drinking water is the effect on taste. Chloride concentra-
tions of greater than 5000 mg/1 have been reported to be
toxic to certain types of bird life. (34) The effect of chloride
c-n plant life is difficult to evaluate. Many aquatic plants
can adjust to wide ranges in chloride concentration. Some
Plants are, however, susceptible to chloride damage and con-
centrations as low as 100 mg/1 in irrigation water can be
harmful to certain citrus and fruit. '5') Chloride in boiler
and cooling water contributes to the dissolved solids. When
the water is recycled, high chloride concentrations lead
to more frequent blowdown and a larger blowdown waste stream.
High chloride concentrations can also lead to increased corro-
sion of metal surfaces.
Chromium in the hexavalent state (CrO^, Cr^Oy) must be consid-
ered extremely toxic to plant and animal life. The United
States Public Health Service sets the limit for hexavalent
chromium in public water supplies at 0.05 mg/1. Continuous
131
-------
intake of chromates has been reported to cause intestinal
inflammation, kidney damage, and lung cancer. '5^ Chromate
concentrations greater than 0.2 mg/1 are toxic to algae and
other aquatic plant life, and higher concentrations (>1 mg/1)
are toxic to many species of fish.(34) Marine life, except
for mollusks, shows a somewhat greater tolerance to hexavalent
chromium than most fresh water life. Mollusks, such as oysters,
have a very low tolerance to chromates, and concentrations
of 0.01 mg/1 are reported to be toxic to oysters.^34'
The heavy metals, as a group, must be considered extremely
toxic to plant and animal life. The United States Public
Health Service sets the limit for most heavy metals in drink-
ing water at 0.05 mg/1. Some heavy metals are permissible
at higher levels (i.e., Zinc - 5.0 mg/1) but still must be
considered very toxic. The effect of many heavy metals on
plant and animal life is complicated by the fact that their
toxicity depends somewhat on the concentration of calcium
and magnesium in the water.
Organic Contaminants
Organic contaminants are not a major problem in the aqueous
effluents from fertilizer plants. The principal problems
are urea from urea plant and blending plant wastes; monoeth-
anolamine in ammonia plant wastes, and oil leaked from pumps
and compressors into various waste streams. Some organic com-
pounds are frequently added to cooling water systems for biocidal
purposes, but the concentrations are usually quite low and
do not represent a major contamination problem.
Urea does not represent a serious pollution problem at the
present time, and no specific standards for urea in water suppli©3
have been set by any of the states. Urea is a source of nut-
rient nitrogen for aquatic plants and can contribute to increased
plant growth. However, urea assimilation by plants is rela-
tively slow, and it is doubtful that urea can be considered
as contributing significantly to eutrophication. Urea has
been reported to be toxic to certain species of fish but only
at concentrations of 30,000 mg/1.(58> urea is used as an^
animal food supplement so should not be considered as toxic
to animal life.
Monoethanolamine (MEA) is a minor pollution problem in the
aqueous effluents from ammonia plants. The principal effect
of MEA on the waste stream and receiving water is that it
adds to the BOD load of the stream. MEA concentrations greater
than 1500 mg/1 are reported to be toxic to certain species
of fish upon long term exposure.
132
-------
°ils in waste streams can be a serious pollution problem.
The deleterious effects of oil in the receiving waters are
several. The oil can create an odor and taste problem in
Public water supplies. It can coat the water with a thin
film, thus affecting the aesthetic value of the water. The
°il film can interfere with the dissolution of oxygen in the
wfter and with photosynthesis. Direct contact of fish with
°il can result in the oil interfering with the respiratory
a°tion of the gills. The oil may also taint the flesh of
the fish if ingested. Oil can coat plant life, such as algae,
an<3 kill it. Suspended solids in the water may be covered
with oil and settle to the bottom where they can interfere
with plant growth and fish spawning. The oil can coat the
Plummage of water f owl , thereby destroying their natural buoy-
ar*cy and insulation. Oil can also affect egg laying in water-
fowl if ingested, and oil coating of eggs can reduce hatching.
Components of the oil may be toxic to plant and animal life,
ar*d it has been reported that concentrations as low as 0.3
can be toxic to certain fish species. l-^,:>yj
OF AQUEOUS EFFLUENTS
A<*equate control of fertilizer plant effluents requires careful
Monitoring and anaylsis of waste streams. When possible,
c°ntinuous monitoring of waste streams is the preferred approach.
However, reliable continuous analytical procedures are not
ae c
(or have not been adapted to industrial use) for
rnq most of the contaminants found in fertilizer plant
streams The only parameters that are analyzed on a
uouT S«is in a significant number of plants are PH
nd conductivity.
Th* frequency of sampling and extent of analysis of most waste
st*eams independent primarily on plant location and existing
**£ ^lityPregulatLns Most Plant sample^aste streams
?UHn rdaily'Ssis?" "as wLre water quality regula-
t4°ns are less stringent samples are usually taken less fre-
^ntly and some Plants may sample once a week or less. The
**thod of Sasle disposal can also influence the monitoring
of waste streams. Companies using deep well
0^
eamS I ^ waters wSIte streams which discharge
arg
-------
The analytical methods for measuring the conventional pollu-
tion parameters (i.e., pH, turbidity, BOD, etc.) used by the
fertilizer industry are well developed, and are described
in several references.(60,61,62)
A variety of analytical procedures is available for each of
the various inorganic ions found in fertilizer plant waste
streams. Most of the methods suitable for analyzing waste
water streams to the required minimum concentration levels
are described in the "Manual on Industrial Water and Industrial
Waste Water" published by the ASTM.(62) Several other good
references are also available which describe the required
analytical procedures. (60,62,63,64,65)
134
-------
CONTROL AND TREATMENT OF AQUEOUS EFFLUENTS
This chapter describes the general methods used by the fertil-
izer industry to control and treat aqueous waste streams.
Use of these methods by the various segments of the fertil-
izer industry is discussed.
CLASSIFICATION OF TREATMENT METHODS
Despite the multitude of fertilizer materials produced by
the fertilizer industry, the methods used by the industry
to control and treat contaminated aqueous waste streams are
relatively limited. These methods can be divided into five
general categories:
1. Disposal methods
2. Process improvements and modification
3. Water recycle and reuse
4. Physical treatment methods
5. Chemical treatment methods.
Biological treatment methods, which are widely used in treat-
ing other industrial and municipal wastes, are almost unknown
in the fertilizer industry. This reflects the fact that the
Principal contaminants found in fertilizer industry waste
Caters are primarily inorganic materials.
°ne other approach used by the fertilizer industry for control
and treatment of waste water must be considered. It involves
the segregation, or lack thereof, of the various waste water
streams within a plant. Normally some attempt is made to
effect some segregation. Uncontaminated waste waters are
Usually kept separate from contaminated screams. Once-through
°ooling water and rainwater runoff from uncontaminated plant
a*eas are typical examples of uncontaminated effluents which
are usually kept separate from other waste streams. Separa-
tion of various contaminated waste streams may be practiced
Depending on the treatment each stream requires prior to dis-
charge. As a typical example, cooling water blowdown, which
is high in chromate, may be kept separate from other contam-
inated waste until the chromate is destroyed.
Sigposal Methods
Disposal methods are available which currently permit discharge
°f untreated or partially treated waste waters. These include
Dilution, ocean disposal, deep well disposal, and discharge
to municipal sewer systems. Under the current water quality
standards of most states these disposal methods are widely
used. As water quality standards become more restrictive,
135
-------
however, these methods will probably become unacceptable.
If effluent standards are developed which base the quantities
of contaminants which can be released on a unit of produc-
tion (i.e., pounds of contaminant discharged per ton of pro-
duct) then it is a certainty that the disposal methods will
be unacceptable.
Disposal of waste water by dilution is based on the concept
that dilution of the waste water by the receiving water reduces
the contaminants present in the waste water to an acceptable
level. The restrictions on this form of disposal are that
sufficient receiving water must be available for dilution,
and that the intended usage of the receiving water must not
be adversely affected. Dilution disposal is often combined
with some form of physical treatment such as settling or hold-
ing ponds.
Disposal by dilution is widely practiced in many areas of
the fertilizer industry. However, as water quality regula-
tions become more restrictive and effluent standards are
adopted, particularly on nutrient discharge, this form of
disposal will become unacceptable. The many plants which
now rely on dilution for disposal of untreated or partially
treated waste waters will be forced to develop and install
treatment processes.
Ocean disposal finds limited use in the fertilizer industry.
It is essentially a variation of disposal by dilution. As
such it is becoming less acceptable as control criteria be-
come more stringent.
Deep well disposal is widely used by many industries for dis-
posing of contaminated waste waters. Several plants in the
fertilizer industry have reported using deep well disposal.
In the process the waste stream is injected under pressure
into geological formations far below surface and sub-surface
water supplies. The disposal depths can range up to 12,000
feet depending on water tables and underlying strata.(6°)
In certain cases some form of preliminary treatment may be
required before final disposal. This usually involves settl-
ing for solids removal and/or oil separation . Care must be
taken that deep well disposal does not contaminate surface
or sub-surface water supplies.
In a few cases, fertilizer plant wastes are discharged into
municipal sewage systems. This practice is usually limited
to small plants with small waste loads. Typical examples
are small liquid mix plants which may discharge equipment
washdown and spill solutions into the local sewage systems.
This form of disposal has been only partially successful,
and will undoubtedly be less acceptable in the future.
136
-------
Process Improvement and Modification
The most satisfactory method of controlling aqueous waste
streams is to prevent or reduce their formation. This approach
can result in improved process economics through better yields,
as well as reduced waste treatment costs. There are numerous
Ways in which such improvements can be effected. Improved
operational control and preventive maintenance can reduce
leaks and spills which create waste streams as well as loss
of valuable materials. Process changes can be made so that
spill cleanup solutions can be recycled back to the process.
Waste streams created by scrubbing feed or product materials
from gaseous effluents can be recycled back to the process.
Continuous or extensive monitoring of waste streams can be
initiated to help in identifying process malfunctions and
ieaks. Wherever possible, recovery of potentially valuable
by-products from waste streams can substantially reduce waste
treatment requirements.
Hater Recycle and Reuse
Recycling and reuse of waste water can be a most effective
and inexpensive method for controlling industrial waste loads.
It is widely practiced in the fertilizer industry. Effective
^euse of water reduces the volumes of waste water to be treated,
Deduces the need for large volumes of fresh water/ and allows
ftore freedom from upstream water users. Currently, about
75-80% of the gross water requirements of the fertilizer in-
dustry are supplied through recycling and reuse of waste water.
The fertilizer industry requires large volumes of cooling
water. Recycling of the cooling water is widely practiced
in all segments of the industry. Recycling decreases the
Duality of the water and requires the use of additives and
special materials of construction to control scaling and cor-
posion. The advantages of cooling water recycle, however,
offset these added costs. Once-through cooling water operation
is only used when water is readily available and when the
cooling water effluent can be discharged without treatment
e*cept for temperature reduction. Contaminated water from
other plant operations can also be used for cooling purposes
to reduce water consumption. As an example, gypsum pond water
can be used for cooling purposes in many operations in a phos-
phoric acid-fertilizer plant complex. Blowdown from the cooling
water recycle system represents a small-volume contaminated
^aste stream. Blowdown can be treated separately from other
Contaminated wastes, or combined with them to be treated as
a single v/aste stream.
137
-------
Recycle of steam condensate represents another effective method
for water recycle. Normally, boiler operation requires that
only condensate which is not contaminated with process materials
be recycled. This frequently eliminates the recycling of
condensate from such operations as steam stripping. However,
condensate not returned to the boiler water can frequently
be used to supply other plant water requirements. Boiler
water blowdown represents another small-volume source of contam-
inated waste water. It is often combined with cooling water
blowdown for treatment.
Other major consumers of water in the fertilizer industry
are the scrubber operations required for treatment of gaseous
waste streams. Once-through use of scrubber solution would
result in the generation of large volumes of low level waste
which would be extremely difficult to treat. By recycling
the scrubber solution the waste volume is substantially reduced,
and the small-volume, highly contaminated waste stream can
be more effectively treated.
Physical Treatment Methods
The physical treatment methods used by the fertilizer industry
are relatively few. They are limited primarily to gravity
separation techniques and to stripping operations.
Gravity separation methods are used primarily to: remove sus-
pended solids from the untreated waste water; separate solids
formed by chemical treatment of the waste water; and to separate
any oils present from the waste water. Many of the waste
streams generated by the fertilizer industry will contain
significant amounts of suspended solids. An obvious example
is the gypsum slurry discharged from wet process phosphoric
acid plants. Holding or settling ponds are normally used
to allow the solids present to agglomerate and settle out.
Flocculents are sometimes added to the waste stream to assist
settling.
Chemical treatment of waste waters to remove dissolved solids
usually generates precipitate which must be removed before
the waste water is discharged. Clarifiers, thickness and
settling ponds are all used to obtain the required separation.
Filtration and centrifugation are rarely used because of the
volumes of waste water to be treated and the cost involved.
In some cases the waste water streams may contain significant
amounts of oil. Typical examples are process condensate from
ammonia plants and cooling water used on pumps and compressors.
Oil is a troublesome contaminant in waste waters and should
be removed whenever possible. Oils are most easily removed
when their concentrations are high; low oil concentrations
138
-------
can be difficult to treat. If the oil is emulsified with
the water, its removal is even more difficult since gravity
separators are less effective on emulsions. Most plants use
^Pi-design oil separators to obtain the required oil removal .
properly operated separators can reduce the oil level in most
waste water streams to 25 mg/1 or less. (66, 67}
Air flotation can also be used for oil removal. In this pro-
cess, air is dissolved in the waste water under pressure.
The air is then released at atmospheric pressure in a tank
°r basin. Air bubbles, formed by the release of pressure,
rise to the surface, gathering oil and suspended solids which
can be removed by skimming. Air flotation can reduce oil
Bevels to less than 20 mg/1 in most streams, as well as removing
substantial amounts of suspended solids. (67>
Stripping can be used to remove volatile species from waste
Caters. Steam, air, and inert gases are the most commonly
Vsed stripping agents. Steam is the preferred stripping agent
ift most cases since it, plus most volatile contaminants, can
ke readily condensed to give a concentrated aqueous stream.
Air or inert gas stripping of waste waters may simply substi-
tute a gaseous waste problem for a waste water problem. In
the fertilizer industry, stripping is limited almost entirely
to the removal of ammonia from waste water.
c°oling must be provided to prevent the discharge of warm
waters which exceed specified temperature limits. This is
usually accomplished through holding or settling ponds which
Provide sufficient capacity to allow the required cooling
to occur before the waste water is discharged. Once-through
c°oling water is normally treated in this manner. Slowdown
from a recycle cooling water system is usually removed from
the system at the outlet of the cooling tower or pond and
Inquires no further cooling before discharge. Boiler water
Slowdown, on the other hand, must normally be cooled before
Discharge. Cooling is frequently obtained as a result of
°ther treatment processes (i.e., settling ponds for solids
°fte additional physical treatment method finds some use in
the fertilizer industry. In areas of low rainfall and warm
°^imate, solar evaporation can be used to dispose of contami-
nated waste water. The waste stream is pumped to a holding
Pond or dump where the v/ater is evaporated and the nonvolatile
c°ntaminants collect as a solid residue. This method is
^specially attractive for fertilizer plant waste which are
Composed principally of nonvolatile dissolved inorganic materials.
iow to the pond cannot exceed solar evaporation (plus some
Sub-Surface drainage) for extended periods of time without
some aqueous discharge from the pond resulting. Normally,
139
-------
water levels in the ponds are maintained substantially below
pond capacity to allow for extended periods of high rainfall
or low evaporation.
Chemical Treatment Methods
Chemical treatment methods play an important part in the con-
trol of fertilizer plant effluents. They are used principally
for neutralization of aqueous wastes, precipitation of dissolved
inorganics, coagulation of suspended solids and destruction
of toxic materials.
Neutralization of strongly basic or acidic waste waters is
normally required prior to discharge. In the fertilizer indus-
try, the bulk of the waste waters are acidic or neutral. Where
neutralization of acidic waste is required, lime or limestone
are the preferred neutralizing agents.
Chemical precipitation can be used to remove certain contami-
nants from waste waters. Frequently, neutralization and pre-
cipitation are combined to effect removal of contaminants.
Typical of the combined treatment is the neutralization of
acidic gypsum pond water with slaked lime which precipitates
both the fluoride and phosphate in the pond water.
Chemical coagulation can be used to assist in the separation
of suspended solids and precipitates from waste waters. Coagu-
lation is always followed by some physical treatment to effect
separation of the solids from the waste water. Normally,
some form of gravity separation is used such as clarifiers
or settling ponds.
Waste waters will frequently contain toxic materials which
must be destroyed by chemical means. The treatment used will
depend on the contaminant under consideration. Typical of
such materials is chromate ion which is used for corrosion
control in cooling water systems. Blowdown from cooling water
systems can contain substantial levels of chromate. The chro-
mate is quite toxic and must be removed from the blowdown
stream prior to discharge. Chemical treatment is usually
used to destroy the chromate ion since it cannot normally
be precipitated as such.
TREATMENT OF NITROGEN FERTILIZER EFFLUENTS
In the production of ammonia, most plants rely on disposal
techniques for control of contaminated waste waters. Disposal
by dilution or deep well injection are the principal means
of handling all contaminated waste waters. Treatment of pro-
cess waste streams prior to final disposal is usually limited
140
-------
to oil separation and settling. Contaminated cooling water
(blowdown) is normally treated for oil separation and chroiaate
destruction before discharge.
"typically, in an ammonia plant the contaminated process waste
waters (principally process condensate) flow to oil separators
where the oil level is reduced below 30 mg/1. From the separ-
ator the waste water flows to a holding pond for settling.
Residence time in the holding pond is usually in the range
°f 10-20 days. On entry to the pond, the waste water is sampled
for analysis on a routine basis. From the holding pond the
waste water discharges to the receiving waters, and dilution
^s relied upon to reduce the contaminants to an acceptable
level in the receiving water. Some plants may adjust the pl-l
°f the waste water prior to discharge.
Cooling v/ater blowdown may be combined with the process waste
water or treated separately. Typically, oil is removed from
the blowdown using oil separators, and then the chroma te is
Destroyed chemically. Chroma te removal can be accomplished
ky treating the blowdown with sulfur dioxide and lime. The
treated blowdown waste stream, containing <0.1 nvg/1 chroraate,
then discharges to the holding pond.
water from ammonia plants can be treated for ammonia
temoval by stripping. However, stripping is practiced in
°nly a few plants. When stripping is used ammonia removal
Barely exceeds 60-70%. With high gas flows ammonia removal
can be increased to better than 90%. However, such recoveries
can only be obtained when the initial ammonia concentrations
a^e high. Reduction of ammonia levels below 30 mg/1 using
stripping is difficult to achieve. Use of air as the stripp-
irig agent can make condensation and recovery of the stripped
ammonia a problem. The petrochemical industry makes great
Use of stripping to remove ammonia from sour water streams t67'
vhere ammonia concentrations are quite high (1000-5000 mg/l> .
Stripping is of less value in treating ammonia plant waste
streams in which the ammonia concentrations are much less.
current water quality standards for ammonia (or lack thereof)
^ many localities has resulted in the use of dilution as
the principal means of controlling ammonia plant waste waters.
^s water quality criteria for ammonia becomes more restrictive,
Dilution will no longer be acceptable as a disposal method
(a number of plants are already encountering this situation) .
^ost plants recognize this problem and are centering their
Development efforts on ways of reducing waste water loads
increasing the recycle of contaminated waste streams.
ultimate objective will be to achieve complete closed
water reuse. Typical of the methods devised to reduce
waste loads is recycling of certain contaminated process con-
back to the process, as now practiced in some plants.
141
-------
In the production of nitrogen fertilizers such as NI! .NO3,
(NH^-SO^, and urea, the waste water disposal problems are
essentially the same as those encountered in ammonia production.
The methods of controlling and treating waste v/ater discharges
are the same as well. At the present time, most waste streams
are disposed of by dilution or deep well injection with little
or no treatment prior to final disposal. Any treatment of
process effluents prior to discharge usually consists of pll
adjustment and settling in holding ponds. Oil separation
and chromate destruction is usually provided for cooling water
blowdown as well.
Producers of nitrogen fertilizers recognize that disposal
by dilution is becoming less aceptable. Possible methods
for reducing waste loads, increasing the recycle of contaminated
waste water and obtaining completely closed loop water reuse
are being extensively studied. Methods for removing contaminants
from the waste water to facilitate discharge are receiving
less attention.
TREATMENT OF PHOSPHATE INDUSTRY EFFLUENTS
In the phosphate fertilizer industry the treatment and control
of aqueous waste waters is greatly affected by the availability
of a gypsum pond. In wet process phosphoric acid plants and
phosphate fertilizer complexes the gypsum pond serves as a
reservoir for contaminated waste. It is standard practice
to discharge all contaminated waste waters to the pond. The
settled supernate from the pond serves as the source for most
plant v/ater requirements including cooling. In some plants,
recycled pond water accounts for up to 90% of the gross v/ater
requirements. Overflow from the gypsum pond is normally the
only contaminated v/aste stream to be treated prior to discharge.
Uncontaminated waste waters such as once-through cooling water
and rainfall runoff from uncontaminated plant areas are kept
segregated from the pond water system and discharged without
treatment. Plants using recycle cooling v/ater systems separate
from the pond water system frequently discharge the blowdown
stream to the gypsum pond after chemical treatment for chromate
removal. When a gypsum pond is not available the phosphate
fertilizer plant must provide other treatment facilities for
processing the various contaminated waste water streams.
As described earlier, waste water discharged from a phosphoric
acid plant gypsum pond contains substantial amounts of fluoride
and phosphate. The discharge must be treated to reduce these
contaminants to an acceptable level before it is released to
the ground v/aters. In almost every case the treatment procedure
used consists of lime neutralization and settling. In most
cases double liming and possibly triple liming is required to
142
-------
obtain the desired impurity removal. A typical flowsheet
for the double liming of pond water discharge is shown in
Figure 29. In this process, slaked lime or a slaked lime
slurry is mixed with pond water in a mixing tank. The result-
ing slurry passes to an agitated settling tank or clarifier.
Clear supernate from the clarifier passes to a second mixing
tank where additional slaked lime is added. Underflow from
the first clarifier is returned to the gypsum pond. The slurry
from the second mixing tank flows to a second clarifier. Clear
supernate from the second clarifier is discharged to the ground
water or is returned to the plant processes as clean water.
Underflow returns to the gypsum pond.
Limestone can be used in the first stage of neutralization,
but slaked lime is much preferred. Normally, fresh water is
used to slake the lime.
With good agitation in the mixers, slurry residence time in
each is less than one hour, and in most plants a residence
time of 10-15 minutes is normally used. Settling times in
the clarifiers are much longer and can run to several hours.
In the first stage of neutralization using slaked lime, the
pH of the solution is raised to about 3.6-3.8. Most of the
fluoride present in the pond water precipitates, probably as
calcium fluoride. The soluble fluoride content of the super-
nate leaving the first clarifier is in the range of 40-60
tttg/1.(69) The phosphate content of the supernate is only
slightly reduced (5-10%) in the first neutralization stage.
In the second stage of neutralization the pH is raised to
6-7. Almost all of the phosphate and most of the remaining
fluoride precipitates. The fluoride probably precipitates
as fluorapatite. The supernate from the second clarifier
contains 5-20 mg/1 fluoride and 10-30 mg/1 phosphate. Calcium
and sulfate concentrations are reduced somewhat from their
original pond water values, but not to the same level as the
fluoride and phosphate.
Under water quality standards currently existing in most states,
the water from the second stage of liming can be discharged
to receiving water assuming some dilution is possible. If
either fluoride or phosphate criteria become more restrictive,
it is doubtful if lime neutralization can provide the neces-
sary fluoride and/or phosphate removal. Elimination of pond
water discharge through increased water recycle, or an improved
treatment method for fluoride and phosphate removal, v/ill
then be required.
In some localities, water evaporation and sub-surface drainage
from the gypsum pond eliminate the need for controlled dis-
charge of a waste water stream from the pond. In such cases
143
-------
QUICK LIME
FRESH WATER
POND WATER
WASTE RECYCLE
TO POND
CLEAR EFFLUENT
TO DISCHARGE
WASTE RECYCLE
TO POND
FIGURE 29 FLOWSHEET FOR LIME NEUTRALIZATION OF GYPSUM POND WATER (20)
144
-------
pond serves as a reservoir where plant waste waters are
Disposed of without additional treatment. However, as air
and water pollution control regulations become more restric-
tive this technique may not be acceptable and new treating
methods will have to be developed.
AS described earlier, fluoride-containing scrubber ^ solutions
can be a major source of contaminated waste water in any phos-
phate fertilizer or phosphoric acid plant. Treatment of some
form is almost always required for fluoride removal before the
waste water can be discharged to the ground water. The ideal
s°lution is to operate the scrubber system in such a way that
the fluoride can be recovered for sale as hydrofluoric and
fluosilicic acid, metal f luosilicates or metal fluoride. A
Dumber of scrubber systems have been designed to permit fluoride
recovery. (30,31)
When fluoride recovery is not economical or practical, some
form of waste water treatment is required. When a gypsum
Pond is available, the fluoride scrub solutions can be dis-
charged to the pond where they become a part of the overall
Pond water system. When a gypsum pond is not available another
^ethod of disposal is required. In most cases, this involves
Urne neutralization of the scrubber solution followed by
collection of the precipitated fluoride. To reduce the waste
volume to be treated, the scrub solution is recycled and only
the bleedstream used to control the fluoride concentration is
Seated. By neutralizing to pH 7 with slaked lime, the fluoride
concentration of the clear effluent will be reduced to 20-30
after settling.
e streams generated by spill cleanup and equipment washdown
be combined with the scrubber solution and treated as
single waste stream. When this is done lime neutralization
will remove any phosphate in the xvaste stream as well as fluoride.
^he TVA has developed a process for treating scrubber solutions
f°r fluoride removal using granulated furnace slag. In this
Process the scrub solution, containing 700 mg/1 of fluoride,
flows to a settling pond at the slag pile at the rate of 400
Spm. After settling, the solution percolates through the slag
Pile. The fluoride reacts with the slag to precipitate cal-
cium fluoride, which collects in the pile. The clear effluent
from the pile, containing about 25-30 mg/1 fluoride, discharges
^o the receiving ground water.
water from elemental phosphorus production contains sub-
stantial amounts of fluoride and elemental phosphorus which must
removed before the water is discharged to receiving waters
recycled to the process. Several processes have been used
145
-------
for treating the phossy water. The TVA has developed a treat-
ment process which allows complete recycle of the treated
phossy water (Figure 30). The phossy water is collected in
a large storage tank where it is continuously agitated. A
stream of the phossy water is fed to a clarifier where a floccu-
lating agent is added. The suspended solids in the phossy
water (including most of the phosphorus) are coagulated and
removed as underflow from the clarifier. The supernate is
pumped to a clarified water storage tank for recycle to the
process. The underflow of solids and elemental phosphorus
is centrifuged and the filtrate is returned to the phossy
water system. The solids are burned to produce impure phos-
phoric acid. The treatment as described is quite effective
in removing suspended solids and elemental phosphorus as the
following data indicate:
Stream Concentration - mg/1
ElementalSuspended
Phosphorus Solids
Phossy water 1700 2200
Clarified water 120 170
Dissolved solids build up in the clarified water as it recycles/
and a bleed stream must be removed continuously to control
the impurity levels. Typically the composition of the clari-
fied recycle water will be:
Contaminant mg/1
Phosphate 17/000
Sulfate 2,000
Fluoride 1,000
Ammonia 900
Phosphorus 120
The bleed stream amounts to about 6% of the clarified water
flow. Most of the bleed stream is used to slurry the solids
collected in the dust precipitators located in front of the
phosphorus condenser (the slurried solids are recycled back
to furnace feed preparation system). The remainder of the
bleed stream is treated for dissolved solids removal by pass-
ing it over a bed of granulated furnace slag. The fluoride
and phosphate react with the slag and precipitate. The solu-
tion percolates through the bed, solids are removed, and a
clear effluent discharges from the slag pile. The effluent
containing 10-30 mg/1 fluoride/ and <0.5 mg/1 phosphorus is
discharged to the receiving waters.
146
-------
PHOSSY WATER
PHOSSY WATER
STORAGE
TANK
FLOCCULANT
CIRCULATING PUMP
CLARIFIER
I
OVERFLOW
TANK
LIQUOR RECYCLED TO
PHOSSY WATER SYSTEM
J
CENTRIFUGE
SOLIDS BURNED
TO GIVE IMPURE
PHOSPHORIC ACID
CLARIFIED WATER
STORAGE
TANK
CLEAN WATER
RECYCLE
BLEED STREAM TO CONTROL
SOLUBLE IMPURITIES
TO SLURRY TANK
GRANULATED
SLAG PILE
CLEAR EFFLUENT
TO DISCHARGE
FIGURE 30 TVA FLOWSHEET FOR TREATING PHOSSY WATER (50)
147
-------
Other processes used for treating phossy water usually involve
settling to allow the elemental phosphorus to separate out,
followed by lime neutralization to precipitate the fluoride
and phosphate. The solids are collected in a thickener and
the clear supernate is discharged or recycled to the process.
With good settling, the phosphorus content of the treated
effluent is usually low enough to allow direct discharge to
the ground waters. Sometimes, however, additional treatment
is required to reduce the phosphorus to an acceptable level.
This can be accomplished by oxidizing the elemental phosphorus
with chlorine. Dissolved chlorine is quite toxic to aquatic
life, however, so care must be used to control the chlorine
addition. Oxidation with chlorine reduces the elemental
phosphorus concentration below 0.5 mg/1.
POTASH INDUSTRY EFFLUENTS
Disposal of waste streams from potassium chloride production
is not a problem. When the KC1 is produced from brines, as
happens at Bonneville and Searles Lake, any waste water streams
produced are simply recycled back to a discharge area at the
brine source. No treatment prior to discharge is required.
Waste streams generated in the production of KC1 from sylvinite
are discharged to dumps where solar evaporation is used to remove
the water. The solid waste remaining after evaporation accum-
ulates at the dump. Normally, the waste streams are discharged
to the dumps without treatment.
148
-------
ECONOMICS OF EFFLUENT CONTROL
"The economics of providing adequate treatment of fertilizer
plant waste waters can have an important effect on fertilizer
Production. The cost required to meet water quality standards
can ultimately affect the production processes used and the
Market value of the fertilizer products.
*n evaluating the economic effects of effluent control on the
fertilizer industry there are two factors which must be con-
sidered:
1. The capital and operating costs of treatment facilities
required to meet water quality standards.
2. The cost of in-plant modifications designed to improve
effluent control through reductions in waste loads
and waste water volumes and increased waste water
recycle.
Unfortunately, only a limited amount of information is avail-
able in the literature on the cost of treating fertilizer
Plant waste waters. Even less information is available on
in-plant modifications. Overall, the fertilizer industry
has done a good job of meeting current water quality standards.
In most cases, however, they have failed to publicize their
ef forts to the extent that other industries have.
cost data available on treatment of fertilizer plant waste
Caters are generally reported as total costs for the complete
waste treatment system. These costs can vary over wide ranges
Depending on the type and degree of treatment used. Plant
data obtained in the current study show the following typical
cost ranges for treatment of ammonia and nitrogen fertilizer
Plant waste waters.
Capital Cost Operating Cost
($/ton daily capacity) ($/ton product)
200-1000 0.08-0.78
150-600 0.06-0.50
0-250 0-0.14
200-1000 0.08-0.68
The costs are due principally to oil separation and settling
Pond treatment. The low operating costs reflect the minimum
treatment provided these waste streams.
Ammonium Nitrate
149
-------
In the production of wet process phosphoric acid the cost
of treating gypsum pond water discharge by lime neutralization
is reported to be $0.80~$1.80 per thousand gallons.t20' The
cost per ton of P2°5 product is reported to be about $1.50
for a 600 ton P205-per-day plant.(^0) Data obtained in the
current study show the operating costs varying from $0.90-
$3.00 per ton of P2°5 product and average about $1.90 per
ton of 1?2^5 product. When no discharge from the gypsum pond
is required the only operating costs involved are for pond
maintenance and pumping.
Capital costs for gypsum ponds can vary over wide ranges be-
cause of the many variables involved in construction. A typical
range of construction costs for ponds and related equipment
as a function of plant size is shown in Figure 31.
Costs for treating phosphate fertilizer plant waste waters are
largely dependent on how the scrubber solutions are handled.
When the fluoride is recovered for sale, the value of the pro-
duct approximates the cost of recovery. When the waste waters
discharge to a gypsum pond the fertilizer plant product bears
a proportionate share of the cost of treating the pond water
dishcarge.
In phosphate fertilizer complexes approximately 11-15% of capi-
tal expenditure is for pollution control equipment and facilities.
Of this amount 35-40% is for water pollution control.
A certain amount of data is available in the literature on
unit costs for specific processes and operations within a waste
treatment system. Those process data available which are rela-
vent to fertilizer plant waste water treatment include data
on oil separation, ammonia stripping, holding and settling ponds,
and treatment of cooling water blowdown. Each of these opera-
tions is discussed in the following sections.
Oil removal from waste water is a significant problem in the
fertilizer industry. Industry data on the cost of oil separa-
tion are lacking. However, there are considerable data in the
literature on the cost of oil removal from petroleum refinery
and other industrial waste waters. These data should apply
to the treatment of fertilizer plant waste waters as well.
Figure 32 shows typical capital costs and annual operating
plus maintenance costs for API separators as a function of
daily waste water flow. The data presented are for petroleum
refinery waste water treatment.t10' ' The costs should not
be significantly different for fertilizer plant waste water
treatment.
150
-------
800
600
CO
o
o
o
I
CO
co
o
400
co
z
o
200
I
200 400 600 800 1000
PLANT CAPACITY - TONS PO PER DAY
1200
FIGURE 31 TYPICAL GYPSUM POND CONSTRUCTION COSTS
151
-------
10'
VJ-
oo
o
C_5
Q-
e£
O
10'
10
1 I I I I
\ I 1 1 I I I
I I I I I 1
J I I I I I I I
10'
101
co
I—
CO
o
o
to
o.
o
10
1 10
WASTE FLOW - MILLIONS OF GALLONS PER DAY
FIGURE 32 CAPITAL AND OPERATING COSTS FOR OIL SEPARATORS (10,70)
152
-------
flotation removal of oil is somewhat more expensive.
Capital costs are approximately twice those of API separa-
tors and operating and maintenance costs are about 1.5 times
as high. (7D)
Many fertilizer plants make use of holding or settling ponds
f°r storing waste waters prior to discharge. The capital
cost of settling or holding ponds is dependent on a number
°f factors such as plant location, land availability, etc.
Figure 33 shows a typical range of construction costs for
E>°nds as a function of pond area. Pond maintenance and oper-
ating costs are normally very low.
little cost data are available on the stripping of ammonia
fertilizer plant waste waters. The capital cost of an
a stripper and condenser for a 1200 ton per day ammonia
is reported to be $60,000. <66) Ammonia strippers used
the petroleum industry and capable of handling approximately
same waste flows cost in the range of $40 ,000-$60,000.
Derating costs for ammonia strippers are difficult to predict
since costs depend primarily on the degree of ammonia removal
tequired. Data on ammonia removal from secondary sewage efflu-
by stripoing show the operating costs to be about $0.05/1000
ens treated to achieve about 80% ammonia removal. < '-^
of cooling tower blowdown is normally required to
^emove chromate corrosion inhibitor. Chromate destruction
is usually achieved by treating the blowdown waste with SO2
^d lime. Table 16 summarizes typical cost data for removing
chromate from blowdown waste water using S02 and lime.*'-*'
TABLE 16
TYPICAL COST DATA FOR CHROMATE REMOVAL
FROM COOLING WATER BLOWDOWN('*I
trihibitor
Astern
Only
chr ornate
jjjjromate
^Qsphate
Initial
Concentration
(mq/1)
200-500
8-35
17-65
8-35
10-15
30-45
Concentration In
Treated Waste
(mg/1)
CrO4-0.05
Cr -3.0
Cr04-0.05
Cr -3.0
Zn -5
CrO^-O.OS
Cr -3.0
Zn -5
Disposal Cost
($/1000 gallons)
0.70
0.16
0.13
153
-------
10'
CO
CO
o
£ 10"
o
o
I I I
i I i i i i
I til r i i
i i i
I ,
1 10
POND AREA - ACRES
\ r
FIGURE 33 TYPICAL CONSTRUCTION COSTS FOR HOLDING AND SETTLING PONDS
154
-------
estimated capital equipment costs for a plant to treat
gpm of blowdov/n waste are presented in the following table. '
Inhibitor System Capital Cost
Chromate $25,000
Zinc-chromate 30,000
Zinc-chromate-phosphate 31,000
organics are present in the blowdown streams (i.e., organic
Phosphate corrosion inhibitors and biocides) other treatment
tosthods are required and treatment costs may be substantially
higher.
155
-------
RESEARCH AND DEVELOPMENT REQUIREMENTS
FOR REQUIRED CONTROL OF AQUEOUS EFFLUENTS
technology is currently available for controlling many of
pollutants found in fertilizer plant waste streams. There
however, certain problem areas where additional development
effort is needed to provide feasible waste treatment methods.
^n addition, it is almost a certainty that water quality stan-
dards will become more restrictive in the future. Waste treat-
^ent methods which are sufficient for current water quality
standards may not be satisfactory for meeting future standards.
^his means new waste treatment technology will have to be developed
f°r those situations where current technology may prove inadequate.
Bellowing is a list of current and potential problem areas
WUch need investigation to assist the fertilizer industry
in maintaining adequate control over aqueous effluents. (No
significance should be placed on the order in which the prob-
lem areas are discussed.)
1. Realistic Water Quality Standards - The basic problem
facing the fertilizer industry in regard to waste water
pollution control is the assurance that future water
quality standards will be both realistic and reason-
able, particularly with regard to the nutrient content
of waste waters. Unrealistic standards for nitrogen,
phosphate and fluoride could have a disastrous effect
on future fertilizer production. A program is needed
which will permit definition of realistic standards
for those pollutants of primary interest to the fer-
tilizer industry.
2. Ammonia Removal - The use of dilution as a means of
controlling ammonia bearing waste waters is unaccept-
able in some areas today, and will be less acceptable
in the future. Economic treatment methods for ammonia
removal from fertilizer plant waste waters are needed.
A number of processes have been studied for ammonia
removal from waste water. In most cases, however,
the processes have been applied to the removal of
ammonia from sewage. With the exception of stripping,
none are currently being applied to fertilizer plant
waste waters. Selective ion exchange for ammonia
removal appears to be the most attractive method
currently available for treating fertilizer plant waste
waters.^'1^ A much needed program (supported by the
EPA under Grant No. 12020 EGM) is currently under-
way at the Farmers Chemical Association, Inc. plant
at Tyner, Tennessee to evaluate the use of ion exchange
157
-------
for removing ammonia and nitrate from fertilizer
plant waste waters. However, the magnitude of the
ammonia problem justifies additional ion exchange
studies. Application of other ammonia removal methods
to fertilizer plant waste waters should also be eval-
uated .
3. Nitrogen Removal - Methods are also needed for the
removal of other forms of nitrogen from fertilizer
plant waste waters, particularly nitrate and urea.
As is the case with ammonia, dilution is still the
basic means of control for nitrate and urea. If dilu-
tion becomes unacceptable, as appears likely, there
are no demonstrated treatment methods available for
removing nitrate and urea from fertilizer plant waste
waters. The Farmers Chemical Association study on
ammonia and nitrate removal by ion exchange may prove
satisfactory for nitrate removal. However, studies
should be undertaken to develop a suitable method
for urea removal.
4. Gypsum Utilization - Gypsum produced in phosphoric
acid manufacture is not usually considered a major dis-
posal problem. It is simply allowed to accumulate
in the gypsum pond. However, considerable land area
is required to accommodate the large tonnages of by-
product gypsum turned out by the phosphate industry
each year. In areas where phosphoric acid manufacture
is heavily concentrated, such as central Florida, the
land required for gypsum disposal can become very
large. Various methods for utilizing the by-product
gypsum have been studied in the past and in most cases
have been found to be uneconomical. It would appear,
however, that additional efforts to develop ways of
utilizing the gypsum are needed and are economically
justified.
5. Fluoride Removal - Fluoride is a principal contaminant
in all waste waters discharge from phosphoric acid and
phosphate fertilizer plants. Lime neutralization, when
combined with dilution, provides sufficient fluoride
removal to meet current fluoride water quality standards,
It appears quite likely, however, that fluoride stan-
dards will become more restrictive. It is very doubt-
ful that line neutralization would provide the fluoride
removal needed if the fluoride standards are reduced
much below their current levels. Therefore, a pro-
gram should be undertaken to develop alternate methods
of fluoride removal and possible recovery.
158
-------
6. Phosphate Removal - Lime neutralization is also used
for phosphate removal from fertilizer plant waste
waters. The treatment, combined with dilution,
is sufficient to meet current phosphate standards,
but will not be satisfactory if phosphate standards
are reduced significantly. A suitable alternative
to lime neutralization should be developed.
A variety of processes has been developed for phos-
phate removal from sewage.''3'74/'5''6/ Host of these
processes will reduce the phosphate to quite low levels,
Application of these processes to the treatment of
fertilizer plant waste waters should be investigated
for both technical and economic feasibility.
7. Lime Neutralization Mechanism - When phosphate is pre-
cipitated by lime neutralization the precipitated
species at pH 6 and above is reported to be hydroxya-
patite [Ca50H(P04)3]. The reported solubility of the
hydroxyapatite is extremely low. However, the con-
centration of phosphate in neutralized gypsum pond
water is relatively high (>10 mg/1). An investiga-
tion of the precipitation mechanism and phosphate
species present could lead to improvement in the lime
neutralization process and extend its usefulness.
8. Continuous Monitoring Procedures - With the exception
of pll measurement, reliable continuous analytical
methods for monitoring the principal contaminants
found in fertilizer plant waste waters are not avail-
able. Development of continuous methods for fluoride,
phosphate, ammonia and nitrate analysis would greatly
assist the fertilizer industry in controlling waste
water discharge. Process leaks and malfunctions
could be more easily detected, and process losses
could be reduced.
159
-------
REFERENCES
1. N. S. Petersen, "The Phosphate Rock Industry of the Pacific
Northwest-Pacific Northwest Economic Base Study for Power
Markets/1 Volume II, Part 13E, United States Department of
the Interior (1964).
2- Anon. "Chemical Economics Handbook," Stanford Research
Institute (1967).
3- V. Sauchelli, Editor, Fertilizer Nitrogen, Reinhold Pub-
lishing Corp., New York, New York(1964).
4« W. H. Waggaman, Phosphoric Acid, Phosphates, and Phosphatic
Fertilizers, Reinhold Publishing Corp./ New York, New York
(1952) .
5- Anon. "Chem. and Engr. News," May 25, pp. 22-23 (1970).
6- C. J. Pratt, "Chem. Eng. Prog.," £3, No. 10, pp. 37-46 (1967)
7» Anon., "Farm Chemicals," 35,, No. 1, pp. 29-32 (1968).
®- J. VI. Sweeney and R. N. Kasslacher, "The Phosphate Industry
in the Southeastern United States and Its Relationship to
World Mineral Fertilizer Demands," 1C 8459, Bureau of Mines,
United States Department of Interior (1970).
9« A. V. Slack, Chemistry and Technology of Fertilizers, Inter-
science Publishers, New York, New York (1967) .
10. C. W. Rice and Company, "Projected Waste Water Treatment
Cost in the Organic Chemical Industry," Cost of Clean
Water Series, Volume IV, Federal Water Pollution Control
Administration (1969) .
^•1. United States Department of the Interior, "The Economics
of Clean Water, Volume III, Inorganic Chemicals Industry
Profile," (1970).
12. A. Payne and J. A. Conner, "Chem. Engr. Prog.," 65, No. 5,
pp. 81-88 (1969).
•'•3. United States Bureau of Mines, "The United States Position
and Outlook in Potash," Information Circular 8387 (1970).
^4. E. C. Kapusta, Role of Potassium in Agriculture, ASA,
Madison, Wisconsin (1968).
*5. United States Bureau of the Census, "Census of Manufacturers,
1963," MC63(1)-10, United States Government Printing Office,
Washington, D.C., (1966).
161
-------
16. R. N. Shreve, Chemical Process Industries, 3rd Ed., McGraw-
Hill Book Co., New York, Nev; York (1967).
17. J. B. Allen, "Chem. Engr. Prog.," 61., No. 9, pp. 473-483 (1965).
18. R. Noyes, Ammonia and Synthesis Gas,Noyes Development Corp.,
Park Ridge, New Jersey (1967).
19. C. J. Pratt and R. Noyes, Nitrogen Fertilizer Chemical Pro-
cesses, Noyes Development Corporation, Park Ridge, New Jersey
(1965).
20. A. V. Slack, Editor, Phosphoric Acid, Parts 1 and 2, Marcel
Dekker, Inc., New York, New York (1968).
21. C. E. Franklin, "Proc. Pert. Ind. Round Table," 21_ (1960).
22. R. E. Kirk and D. P. Othner, Encyclopedia of Chemical Tech-:
nology, Second Edition, Volume16,John Wiley and Sons,
Inc., New York, New York (1966).
23. D. E. Garrett, Industrial Mineral and Rocks, 3rd Ed., AIME,
New York, New York (1960).
24. D. E. Garrett, "Chem. Engr. Prog.," 54, No. 12, pp. 65-69
(1958).
25. T. P. Hignett and T. N. Hubbuch, "Ind. Engr. Chem.," 38,
pp. 1208-16 (1966).
26. E. J. Fox, J. M. Stinson, and G. Tarbutton, "Superphosphate,
Its History, Chemistry and Manufacture," United States
Government Printing Office, Washington, D.C. (1964).
27. J. H. Forster, "16th Ontario Industrial Waste Conference
of the Ontario Water Resources Commission," Niagra Falls,
Ontario, June (1969).
28. A. J. Teller, "Chem. Engr. Prog.," £3, pp. 75-79 (1967).
29. D. A. Linou, "Chem. and Proc. Engr.," 45, pp. 604-611 (1964).
30. M. English, "Chem. and Proc. Engr.," 48, No. 11, pp. 43-47
(1964).
31. F. E. Gartrell and J. C. Barber, "Chem. Engr. Prog.," 62,
No. 10, pp. 44-47 (1966).
32. J. C. Barber, "Chem. Engr. Prog.," 64, No. 9, pp. 78-82 (1968)-
33. J. C. Barber and T. D. Farr, "Chem. Engr. Prog.," 66, No. 11
pp. 56-62 (1970).
162
-------
34. National Technical Advisory Committee, Water Quality Criteria,
Federal Water Pollution Control Administration, Washington,
D.C. (1968).
35. i. E. Wallen, "Bull, Oklahoma Agric. Mech. College, Biol.
Series 48:27" (1951).
36. European Inland Fisheries Advisory Committee; Working Party
on Water Criteria for European Fresh Water Fish, Tech. Paper
No. 1 (1964).
37. Pennsylvania Department of Health, "Heated Discharges -
Their Effect on Streams," Report No. 3 (1962).
38. Technical Advisory and Investigations Branch, FWPCA,
"Temperature and Aquatic Life - Laboratory Investigations,"
Ho. 6, Cincinnati, Ohio (1967).
39. Pacific Northwest Water Laboratory, "Industrial Waste
Guide on Thermal Pollution," FWPCA, Corvallis, Oregon(1968) .
40. N. c. Brady, Editor, Agriculture and the Quality of pur
Environment, AAAS Publication 85, Washington, D.C. (1967).
4L Pacific Northwest Laboratory, "Ammonia Removal from Agri-
cultural Runoff and Secondary Effluents by Selected Ion
Exchange," FWPCA Report No. TWRC-5. Cincinnati, Ohio (1969).
42. C. N. Sawyer, "Jour. Hew England Water Works Assn.," 61,
No. 2, pp. 109-127.
43. J. E. McKee and H. W. Wolfe, "Water Quality Criteria,"
Second Edition, Publication No. 3A, State Water Quality
Control Board, State of California (1963).
44. p. ooudoroff and M. Katz, "Sew. Ind. Wastes," 2£, pp. 432-58
(1960).
45, A. S. Behrman, Water is Everybody's Business, The Chemistry
of water Purification, Anchor Booas, Garden City, New
York
46. j. R. E. Jones, Fish and River Pollution, Butterworths,
London (1964).
4?. Betz Laboratories, Beta Handbook of Industrial Water Con-
ditioning, Philadelphia, Pennsylvania (1962).
48. T. R. Camp, Water and Its Impurities, Reinhold Publishing
Corporation, New York, New Jfork (1963) .
163
-------
49. Pesticide Chemicals Official Compendium, Association of
American Pesticide Control Officials, Inc., College Park,
Maryland (1958).
50. J. C. Barber, "Chem. Engr. Prog.," 65, Ho. 6, pp. 70-73
(1969) .
51. N. I. Sax, Handbook of Dangerous Materials, Reinhold
Publishing Corp^., New York, New York (1951) .
52. B. G. Isom, "J. Wat. Poll. Cont. Fed.," 32, pp. 1312-16
(1960).
53. C. J. Marienfield, "Nitrates in Human Health," Missouri
Agric. Expt. Station Special Report No. 55, pp. 37-38 (1965)'
54. R. Y. Eagers, Toxic Properties of Inorganic Fluorine Com-
pounds , Elsevier Publishing Co., Ltd., New York, New York
(1969) .
55. N. L. Peterson, "Sulfates in Drinking Water," Official
Bulletin, North Datota Water and Sewage Works Conference,
18_, pp. 6-11 (1951) .
56. I. S. Allison, "Science," 7^, pp. 559 (1930).
57. L. Bernstein, ASTM Special Technical Publication No. 416,
Philadelphia, Pennsylvania (1967) .
58. W. S. Spector, Handbook of Toxicology, Volume 1, W. B.
Saunders Company, Philadelphia, Pennsylvania (1956).
59. W. A. Chipman and P. S. Galtsoff, "Effects of Oil Mixed
with Carbonized Sand on Aquatic Animals," U.S. Fish and
Wildlife Service Special Report Fish No. 1 (1949).
60. Anon. Standard Methods for the Examination of Water and
Waste Water, American Public Health Association, New York,
New York (1965) .
61. D. A. Clifford, "Total Oxygen Demand - A New Instrumental
Method," Proc. 24th Annual Scientific Meeting, Midland
Section, Am. Chem. Soc., Midland, Michigan (1967).
62. Anon., Manual on Industrial Water and Industrial Waste
Water, 2nd Ed., ASTM. Special Technical Publication No.
148-1, Philadelphia, Pennsylvania (1966).
63. F. J. Welcher, Editor,Standard Methods of Chemical Analysis/
6th Ed., D. Van Nostrand Co., Inc., New York, New York (1969)'
164
-------
64. p. D. Snell and C. T. Snell, Colorimetric Methods of Analysis.
D. Van llostrand Co., Inc., New York, New York (1959).
65 * M. G. Mellon, "Detection and Analysis of Chemicals in Water-I.
Inorganic Constituents," Proc. Conf. on Phys. Aspects of
Water Quality, United States Public Health Service, Vol. 27
(I960) .
6g. E. P. Gloyna and D. L. Ford, "Petrochemical Effluents
Treatment Practices," FWPCA Report No. 12020, United States
Department of the Interior, Robert S. Kerr Water Research
Center, Ada, Oklahoma (1970).
67.
M. R. Beychok, Aqueous Wastes from Petroleum and Petrochem-
ical Plants, John Wiley and Sons, New York, New York (1967).
68• Anon., Manual on Disposal of Refinery Wastes, American
Petroleum Institute, New York, New York.
69. R. C. Specht, "J. Wat. Poll. Cent. Fed.," 3J2, pp. 964-974
{I960} .
7°- Anon., "Industrial Waste Profiles No. 5 - Petroleum Refining,"
Cost of Clean Water Series - Volume III, Federal Water
Pollution Control Administration, Washington, D.C, (1967).
7l- Anon., "Summary Report - Advanced Waste Treatment," FWPCA
Publication No. WP-20-AWTR-19, Cincinnati, Ohio (1968).
72 • G. E. Glover, "Industrial Process Design for Water Pollution
Control," Volume 2, AIChE Workshop Proceedings, pp. 74-81
(1969).
73* E. P. Earth, et.al., "J. Wat. Poll. Cent. Fed.," £0, p. 1240
(1968) .
?4- M. G. Dunseth, et.al., "Ultimate Disposal of Phosphate from
Waste Water by Recovery as Fertilizer," Federal Water
Quality Administration Report Ho. 17070ESJ, Cincinnati,
Ohio (1970) ,
7s- G R Bell et.al., "Phosphorus Removal Using Chemical
Coagulation and A Continuous Counter-current Filtration
Process," Federal Water Quality Administration Report Mo.
17010EDC (1970) .
76• J J Convery, "The Use of Physical-Chemical Treatment
Techniaues for the Removal of Phosphorus from Municipal
Waste Waters," Presented at the Hew York WPCA Meeting,
January 29, 1970.
165
-------
PART II PHOSPHATE ROCK MINING AND BENEFICIATION
167
-------
PHOSPHATE ROCK MINING AND BENEFICIATIQN
INTRODUCTION
phosphorus, one of the major plant food elements, is obtained
^ the United States from sedimentary deposits of phosphate
?°ck. Current production capacity of phosphate rock is approx-
imately 50 million short tons of material mined. In 1970,
fche actual production of phosphate rock was about 39 million
short tons. Total world production of phosphate rock in 1970
v*s about 89 million short tons.*1) The United States is
bV far the major current world producer of phosphate rock.
The United States exported about 11.3 million short tons of
Phosphate rock; thus, the major amount of that mined in the
a*Uted States is used domestically for production of phosphoric
acid, either by wet acid processing or by production of elemental
Phosphorus in an electric furnace.
phoSphate mining operations are conducted .in but eight of
fche states In order of production capacity, these are Florida,
Tennessee, " North Carolina, Idaho, Montana, Utah, Wyoming, and Cali-
f°rnia. Almost 80 percent of the production capacity is in
the state of Florida. For that reason, more emphasis has
been given in this study to the operating practices in the
s°utheast and particularly in Florida.
Auction of phosphate rock for acid manufacture involves
^ning and physical benef iciation of the mined rock to con-
^ntrate the phosphate values. These operations are generally
a°ne by hydraulic mining and washing methods involving usage
Of tremendous amounts of circulating water. For the most
P*rtfthe water used in the mining, and benef iciation of phosphate
tQck is recirculated with evaporation losses made up with
f*esh water However, because vast amounts of water are retained
*it the Solids rejected from the washing and beneficiation
ations? there are inherent problems in the handling of
rejects that relate to the water ecology of the region
to potential sources of water pollution.
-,.,. t, = =0 nf t-he studv was to review current
purpose of this phase of the s u y iation Qf ho hate
^on^SosSiblJ wate? pollution aspects and to
rog?esst0aSSpia;: for alfeviatlng problems being
by this industry.
169
-------
SUMMARY
A review was made of the phosphate mining and beneficiation
Practice in the United States in relation to the effluents
Generated by these operations and the potential water pollution
Problems invoIved.
?he major effluent from phosphate mining and beneficiation
is the discharge of water suspensions of sand tailings and
Phosphate containing slime solids. These discharges are wholly
contained within the mined out areas and in carefully con-
structed slime ponds; therefore, these wastes from the mining
^nd beneficiation operations are not discharged to the environ-
toent. However, clarified water from the ponds is discharged
Psriodically during periods of heavy rainfall to maintain
a safe level of water in the slime ponds. There is no reported
evidence that any components in this discharge, either dissolved
°r suspended, have an adverse effect on the ecology of the
surrounding water shed, with the possible exception of trace
nutrients which promote growth of algae.
Although the effluents from the mining and beneficiation opera-
tions were not found harmful in the concentrations characteristic
°f these operations, the problem of disposal, particularly
°£ slimes disposal, is a severe problem to the industry. The
Potential hazard of dam breakage is always present and damages
through sudden spillage of slimes into surrounding waterways
are documented by a history of such occurrences. Moreover,
slimes disposal by present methods means a constant demand
for large amounts of fresh water. This is a drain on the
Aquifer or water resource of the area.
Although the phosphate slimes disposal problem is one that
^•s well known by the industry and has received attention by
the operators and by various government research and development
Agencies, the problem is largely a regional problem. For
that reason the magnitude of this disposal problem, particularly
from the standpoint of water and land usage, has not generally
keen recognized by the public. Despite extensive research
through the years by both industry and government, the problem
°f improving the means to dewater or otherwise dispose of
these effluents remains, which attests to the limitations
°f present technology and the economic factors invovled. For
these reasons, it is recommended that a coordinated research
Activity be established to continue to investigate various
Possible methods that prior work has indicated can provide
Potential solutions to the slimes problem. The specific areas
this research are recommended in this report.
171
-------
TliE PHOSPHATE ROCK INDUSTRY
Phosphorus, one of the major plant food elements, is mined
in the United States almost entirely in the form of phosphate
rock from sedimentary deposits. Three-fourths of the phos-
phate rock mined in the United States is used in agriculture
in fertilizer form. The remainder is used to make phosphoric
acid-based chemicals . Phosphate chemicals are used in bone
meal for animal feed, detergents, water softeners, rust proof-
ing, flame-resisting compounds, and for several other minor
uses.
Phosphate rock production decreased by 3.2 percent in 1969 C 3
and remained at the 1969 level during 1970. W The decrease
in the production of phosphate in the United States was the
first decline since 1957. *2' However, expansion is expected
to resume in the next few years. TVA anticipates a rise of
some 18 percent in capacity betx-jeen 1970 and 1975. (2) A bal-
ance between supply and demand in the United States is ex-
pected by 1972 and an annual growth rate of some 4 to 5 percent
is anticipated thereafter . (2) Table 17 presents the estimated
production of phosphate rock in the United States. t3' Based
on these estimates, the resources of phosphate rock in the
United States would last some 45 to 64 years. (3) if the low-
grade rock could be used, the duration of availability of
Phosphate rock would be about 7 times greater. (3)
TABLE 17
ESTIMATED PRODUCTION OF MARKETABLE PHOSPHATE
- ROCK IN THE UHITED STATES^3)
(Thousands of Short Tons)
Year Projection Range Phosphate Rock P^Og Content
(a)
(a)
39,436(a) 9,132(a)
39,044
39,770(al 12,464Ca)
Low 68,000 21,000
1985 Medium 87,000 27,000
High 102,000 32,000
Low 96,000 30,000
2000 Medium 160,000 50,000
High 208,000 65,000
(a) Actual
173
-------
THE NATURE OF PHOSPHATE MINERALS AND DEPOSITS
Phosphate rock does not have a definite chemical composition
and phosphate deposits may contain one or more phosphate min-
erals. The principal minerals have the apatite structure
and have a generalized formula of 3Ca3P2Og-CaF2. This formula
shows a theoretical content of 92,25 percent tricalcium phos-
phate, Ca3P208/ or bone phosphate of lime (BPL) as it is commonly
called. The beneficiated rock in marketable form usually
contains 68 to 77 percent BPL. Table 18 illustrates the extremes
in composition of the phosphate rock(4) showing substitutions
of such elements as Mg and Na.
TABLE 18
VARIATIONS IN PHOSPHATE ROCK COMPOSITIONS
Component Composition Range, percent by weight
CaO 55.6 55.1
MgO 0.0 0.7
Na20 0.0 1.4
P205 42.2 33.95
C02 0-0 6.3
F 3.77 5.04
The apparent hardness of phosphate rock varies from 2 to 5
(Mohs scale). Structures include hard rock, granular, and
a loosely consolidated earthy mass.
In Florida the phosphate deposit occurs principally as a loose
consolidated conglomerate mixed with sand and clayf within
a region about 30 miles in diameter in Polk and Eillsborough
Counties. The North Carolina deposit occurs as interbedded
phosphatic clays, limestones, and sands covering approximately
a 700 square mile area in eastern Beaufort County. The Tennessee.
brown rock deposits are a weathered phosphatic limestone,
that occurs in a north-south trending belt across the state
passing through Nashville and Columbia. The western phosphates
occur as a folded and faulted shale member of the Permian
Phosphoria formation covering an area 350 miles north and
south by about 100 miles wide in portions of Montana, Idaho,
Wyoming, and Utah.
174
-------
MINING PHOSPHATE ROCK
Florida, Tennessee, and North Carolina account for 86 percent
of the United States phosphate rock mine production.^) Figure
34 shows the location of the commercial phosphate rock deposits.
Table 19 gives the actual location and tonnage capacity of
phosphate rock mining operations in the United States.v°>
Most phosphate rock is mined by open-pit methods. The general
practice is to strip the overburden with electric-powered
draglines and then remove the phosphate rock. The mined rock,
referred to as matrix, is then placed into a previously prepared
sluice pit. Hydraulic monitors developing 200 psi pressure
and using as much as 10,000 to 12,000 gpm of water break up
the matrix and slurry it at approximately 40 percent solids
into the suction end of large centrifugal dredge pumps. The
matrix is then pumped through movable steel pipelines for
distances of up to 6 miles to the beneficiation plant.
Although phosphate rock is mined in some locations in the
West by underground techniques rather than by open-pit methods,
such operations account for only a minor part of the total
production.
The common practice in Florida and North Carolina is to strip
the overburden and place it in windrows along side the cut.
The overburden is then bulldozed to build the dams for the
slime ponds. The cuts are usually about 125 to 200 feet in
width and several hundred feet long, depending upon the size
of the dragline.
After the overburden is removed, the ore is placed into sluice
pits where hydraulic monitors break up the ore and slurry
it into the suction of large dredge pumps. The matrix is
pumped through large steel lines to the beneficiation plant.
The beneficiation plant is usually located within 6 miles
of the mining area.
The open-pit mining in Tennessee and the Western States is
somewhat different in that it is common practice to remove
the overburden well in advance of the actual mining opera-
tions. Also, the matrix, when mined, is loaded into trucks
or railroad cars and hauled to washer plants instead of being
pumped.
The underground deposits of the Western phosphates generally
are mined by the room-and-pillar method. Figure 35 illustrates
a typical room-and-pillar mining method. Other underground
mining methods that have been utilized to a limited extent
include shrinkage stopping, top slice, cut-and-fi.il, and long-
wall stopping.
175
-------
I-1
«j
CPl
FIGURE 34 GEOGRAPHICAL LOCATION OF PHOSPHATE ROCK OPERATIONS
-------
TABLE 3
PHOSPHATE ROCK MINES IN THE UNITED STATES
(6)
Capacity, 1000 Short Tons
Year Ending
Company
American Cyanamid Company
Borden Chemical Company
Cominco-American, Inc.
Continental Oil Company
Cuyama phosphate Corporation
El Paso Natural Gas Co.
W.R. Grace & Company
International Minerals &
Chem. Corp.
Mobil Chemical Company
Monsanto Company
Mountain Fuel Supply Co.
Occidental Chenical Co.
Occidental Chemical Co. -
Hooker
Presnell Phosphate Co.
George Relyea
J.R. Simp lot Company
Stauffer Chemical Company
Swift & Company
Tennessee Corporation-USPP
Tennessee Valley Authority
Texas Gulf Sulphur Corp.
U.S.S. Agri. Chemicals-
Armour
Location
Brewster, Florida
Teneroc, Florida
Douglas, Montana
Garrison, Montana
Pierce, Florida
Cuyama, California
Soda Springs, Idaho
Bonny Lake, Florida
Bonnie, Florida
Kingsford, Florida
Fort Meade, Florida
Mt. Pleasant, Tenn.
Ballard, Idaho
Columbia, Tenn.
Conda , Idaho
White Springs, FL
Columbia, Tennessee
Columbia, Tennessee
Garrison, Montana
Fort Hall, Idaho
Soda Springs, Idaho
Melrose, Montana
Hot Springs, Idaho
Montpelier, Idaho
Vernal, Utah
Cherokee, Utah
Leefe, Wyoming
Mt. Pleasant, Tenn.
Watson, Florida
Silver City, FL
Fort Meade, FL
Knob Creek, Tenn.
Franklin, Tenn.
Lee Creek, NC
Armour, Florida
Lake Hancock, FL
Fort Meade, Florida
Columbia, Tenn.
1968
3
1
6
1
2
5
1
1
3
1
3
2
3
1
,650
,500
750
,500
300
400
,550
,000
,000
,700
200
,000
,000
250
,000
750
700
100
,000
600
200
200
400
500
600
,000
,000
200
,000
,500
90
1969
3,
1,
6,
1,
6,
2,
5,
1,
1,
3,
1,
3,
2,
3,
1,
2,
650
500
750
500
300
400
550
000
000
700
200
000
000
250
000
750
700
100
000
600
200
200
400
500
600
000
000
200
000
500
000
90
1970
3
1
6
1
6
2
5
1
1
3
1
3
2
3
1
2
,650
,500
750
,500
300
400
,550
,000
,000
,700
200
,000
,000
250
,000
750
700
100
,000
600
200
200
400
500
600
,000
,000
200
,000
,500
,000
90
TOTAL UNITED STATES
48,640 50,640 50,640
177
-------
MUCK
00
UNDERGROUND SHOP
AND MAINTENANCE AREA
MUCK
DRILLED ----
OUT -
DRILLED ;"-
OUT --
MUCK
INCLINE TO HAULAGE
LEVEL FOR EQUIPMENT
ACCESS
°RE BED
PLAN VIEW
STOfE PILLAR
--/-/•—r~
ORE BED
SECTION AA
ORE PASS
HAULAGE LEVEL
FIGURE 35 GENERAL PLAN FOR CONVENTIONAL ROOM-AND-PILLAR MINING
-------
BENEF 1C IATING PHOSPHATE ROCK
The methods of beneficiation of phosphate rock differ to some
extent with each operating company. Important factors related
to the methods used include: the size and analysis of the
matrix; the ratio of phosphate, clay, and sand in the matrix;
and preference by the operator for certain equipment. In
general, the first treatment step is the separation of coarse
phosphate rock from the clay, sand, and fine phosphate; this
is accomplished in a washer plant. The washer plant separates
the matrix into three components: slimes, fine phosphate and
sand, and coarse phosphate. A generalized flowsheet for the
benef iciation operation is shown in Figure 36. A diagram show-
ing the approximate distribution of products from the beneficia-
tion operation is provided in Figure 37.
It can be noted from the general beneficiation plan shown
in Figure 3 6 that the coarse phosphate removed at the washer
is stockpiled as a marketable product. The fine material
consisting of sands, clay, and fine phosphate xs generally
deslimed to remove clays then sent to the flotation plant
where the fine phosphate is removed by flotation. The reminder
of the material is then discarded and referred to as sand
tails. The slimes, containing only 4 to 6 percent solids,
are generally pumped to slime ponds that have been constructed
in the mined out areas.
It is important to note that the slimes account for one-third
of the total tonnage of matrix mined Thus, about one-third
of the nhosnhate values is currently lost as phosphate slimes.
It is also important to note that the volume of these slimes
is from 25 to 100 percent greater than the matrix that is
*ined. Therefore, the disposal of the slimes cannot be done
within the mined out areas except by construction of dams
to contain the increased volume. Dams are constructed to
slimes to depths of up to 40 feet. The water leaving
ni^Jh these slimes amounts to about one-half the
oflhe mining and beneficiation operation.
As nnt-Pd nreviouslv, the beneficiation operations vary with
location ^operator. Generalized flowsheets which illustrate
'"<• « ««
of these flowsheets that, in general,
•"•s then split into two bo-^^w __ rt4.u«y -*q hv i ^n meeti f^r
for gravity concentration and the other 35 by 150 mesh for
179
-------
00
o
COARSE PHOSPHATE
ROCK
ORE
CLEAR WATER
OVERBURDEN
MINE ft.
I
FINE PHOSPHATE
CONCENTRATE
SLIMES
FINE PHOSPHATE
AND SAND
FLOTATION
SLIME
PONDS
SAND TAIL
FIGURE 36 GENERALIZED FLOWSHEET OF A PHOSPHATE ROCK MINING AND BENEFICIATION PLANT
-------
OTHER
CONSTITUENTS
BONE
PHOSPHATE
OF LIME
EQUIVALENT
36.0%
FLORIDA
PEBBLE PHOSPHATE
MATRIX
SLIMES
32.4% B.P.L.
WASTE
TAILINGS
9.0« B.P.L.
FLOTATION
CONCENTRATE
75.8* B.P.L.
PEBBLE
PHOSPHATE
72.2% B.P.L.
PRODUCTS
DERIVED FROM
MATRIX
WASTE
PRODUCTS
70.35%
RECOVERED
RO
29
SLIMES
WASTE TAILINGS
FLOTATION
CLUIHIIUN
PRODUCTS
PEBBLE
PHOSPHATE
. WASTE
39.02
RECOVERED
61 .0%
PROPORTIONS OF
TOTAL B.P.L. IN
PRODUCTS
FIGURE 37 DISTRIBUTION OF PRODUCTS DERIVED FROM LAND-PEBBLE
PHOSPHATE ROCK, percent
181
-------
MATRIX FROM MINE
(15 TO 20 PCT P205)
30 TO 40 PCT SOLIDS
_J MUD BALL SCALPER
ROTARY TROMMEL
VIBRATING SCREENS
MUD BALL DISINTEGRATOR
HAMMER MILL
DISINTEGRATOR
PRIMARY PEBBLE SEPARATION
FLAT SCREENS
VIBRATING SCREENS
HYDROSIZER
FINAL PEBBLE SEPARATION 1 ^\
VIBRATING SCREEN J^Sf
HYDROSIZER
WASHER DEBRIS
PRIMARY DESLIMING
HYDROSEPARATOR
CYCLONES
SLIME WASTE
STORAGE OF FLOTATION FEED
MINUS 14 PLUS 150-MESH
FIGURE 38 GENERALIZED FLOWSHEET OF A WASHER PLANT
FOR RECOVERING FLORIDA PEBBLE PHOSPHATE
182
-------
FEED STORAGE
I MINUS 14 PLJS 150-MESH
PRIMARY CLASSIFICATION
RAKE CLASSIFIER
SCREW CLASSIFIER
V-BOX
HYDROSIZER
MINUS 14 PLUS 35-MESH
FINES MINUS 35 PLUS 150ZMESH
SECONDARY CLASSIFICATION
HYDROSIZER
~1 HYDROSCILLATOR
1 SCREENS, DSM TYPE
REAGENT CONDITIONER
PADDLE
ROUGHER FLOTATION
10-AIRFLOW
7-SUB-A-DENVER
1-TURBO
1-AIR CELL
1-FAGERGREN
REAGENT CONDITIONER
ROTARY
PADDLE
FROTH
ACID
AGITATOR
TAILS-
•WASTE-
•TAILS—I
DEOILING
SCREW CLASSIFIER
COARSE CONCENTRATION
5-BELT SEPARATORS
3-SHAKING TABLES
2-FLOTATION MACHINE
2-UNDERWATER SCREENS
1-SPIRAL
SILICA FLOTATION
COARSE CONCENTRATE
FROTH
I
WASTE
CONCENTRATE
FIGURE 39 GENERALIZED FLOWSHEET OF A FLOTATION
PLANT FOR RECOVERING FLORIDA PHOSPHATE
183
-------
ORE FROM MINE
(15 PCT P205)
25 PCT SOLIDS
ORE DISTRIBUTOR
ORE SPLIT INTO THREE
PARALLEL CIRCUITS
SCREEN
MINUS 2-INCH
PLUS 2-INCH
DEWATERING SCREENS
1 mm INCLINED
WASTE
VIBRATING SCREENS
14-MESH
SECONDARY
SCALPING SCREEN
. . _ . . , .
14-MESH \ /
14-MESH \ L
MINUS 200-MESH
SLIMES
WASHER PRODUCT
3 TO 5 PCT BY WEIGHT OF TOTAL
HO PEBBLE PRODUCT
REJECT TO WASTE
FIGURE 40 GENERALIZED FLOWSHEET OF A WASHER PLANT FOR
RECOVERING NORTH CAROLINA PHOSPHATE
184
-------
FLOTATION FEED
MINUS 14-HESH SLURRY
r
MINUS 200-HESH—
SLIMES
FIRST STAGE FLOTATION
DESLIMING CYCLONES
SCRUBBERS
SCREW CLASSIFIERS
1
I REAGENT CONDITIONER
=ATTY ACID CONDITIONING
TANKS 1
FLOTATION DISTRIBUTOR
FEED SPLIT INTO 5
FLOTATION BANKS
SECOND STAGE FLOTATION
REPULPING TANK
SECOND STAGE AMINE
SCREW CLASSIFIER
I
REAGENT CONDITIONER 1
AMINE REAGENTS ADDED
^ IN LAUNDER )
AMINE FLOAT DISTRIBUTOR
(SPLIT INTO 3 BANKS
DEWATERED
1 BANK
UNDERFLOW
00
01
r
TAILINGS
QUARTZ SAND UNDERFLOW
t
5-CELL
OPEN FLOTATION BANK
UPGRADES FROM 15 TO 28
PCT P205
FROTH CONCENTRATE
t
DECKING
AGITATED TANKS
(SULFURIC ACID)
AMINE CYCLONES
SCREW CLASSIFIER
1 BANK
*_
6-CELL
OPEN FLOTATION BANKS
(AIR ADDED)
QUARTZ SAND (FLOAT)
— 8 TO 10 PCT OF AMINE FEED
ABOUT 6 PCT P0
- 1
J
1
UNDERFLOW
PHOSPHATE CONCENTRATE, 30.7 PCT
WASTE
DEWATERING
CYCLONES
SCREW CLASSIFIER
r
SINGLE FLOAT CONCENTRATE
62 PCT BPL (28.3 PCT P205)
I _ DOUBLE FLOAT PRODUCT
— 67 PCT BPL (30.7 PCT PjO
~1
STORAGE
STORAGE
FIGURE 41 GENERALIZED FLOWSHEET OF A FLOTATION PLANT
FOR RECOVERING NORTH CAROLINA PHOSPHATE
-------
HIGH-GRADE CIRCUIT
LOW-GRADE CIRCUIT
00
I
DUMP TRUCK
r
SOAKING BIN .— PL
r
WET MATRIX
r
CYLINDRICAL B
SCRUBBER
1 1
SCREEN | '
MINUS 3/16-INCH
*
PUMP •
i »
( SCREEN
PRODUCT MINUS 24-MESH ,
PLUS 24-MESH « 1
HYDROSEPARATORS | '
UNDERFLOW
f
PUMP
I
i
US 3/16-INCH— 1 r
ACKWASH |
1 LOU-GRADE PRODUCT
4 1
1
UNDERFLOW
4
S
1
OVERFLOW • HYDROSEPARATOR
1
UNDERFLOW
t
SCREEN
DEWATERER
*
PRODUCT
- 8-POCKET
•" SIZER
1
f
DRAG
DEWATERER
_^.
*
OVERFLOW TO WASTE
MINUS 225-MESH
BACKV
i
DUMP TRUCK
1
SOAKING BIN
WET MATRIX
f
CYLINDRICAL
SCRUBBER
1
1 1 SCREEN
1
MINUS 24 PLUS 224-MESH
f
HYDROSEPARATORS
• OVERFLOW J
UNDERFLOW
f
PUMP
i
CYCLONE | 1
f
1 1 , , OVERFLOW
1 -* *• TO WASTE
1 DRAG MINUS 225-MESH
ASH DEWATERER
UNDERFLOW
OVERFLOW TO WASTE
HIGH-GRADE PRODUCT
LOW-GRADE PRODUCT
FIGURE 42 FLOWSHEET OF A WASHING PLANT FOR TENNESSEE
BROWN ROCK PHOSPHATE
-------
iov
SLIMES f
THICKENER CONDITION
, >
OVERFLOW CYf ONF
i \
DENS FIE
i i
CONDITIO
ROUGHER PLOT
i T
CONDITION
i
MI
CRU
J
GRIN
CLASS
1
CLASS
ERFLOW
ER
ELLS
R
NER
AILS
ER
OAT
NE
SHER
DING
IFIER
OVERFLOW
IFIER
1 UNDERFLOW
BALL KILL
1
CLEANER FLOTATION
FROTH 1 CONCENTRATE
FROTH
FIGURE 43 TYPICAL FLOWSHEET OF A WESTERN PHOSPHATE ROCK OPERATION
187
-------
treatment by flotation. The flotation process yields a silica
tail, a coarse concentrate, a fine concentrate, and a lev/-
grade phosphate waste product.
In the washer, the main concern is to break up the lumps of
clay to free the phosphate nodules. This is usually accomplished
by a hamiuermill or similar apparatus, except in the Western
operations where grinding mills are used. The nodules themselves
are attrition-scrubbed in the log washers to remove adhering
clay particles from the phosphate nodules. The washer rock
is generally a little lower in phosphate content than the
flotation concentrates. Figure 44 shows views of a typical
Florida washer plant.
In the flotation plant feed preparation section, the size
split for flotation plant feed is made by cyclones, rake classifiers/
spirals, or hydroseparators. In general, this size split
varies between 20 to 35 mesh with the fine fraction being
sent to flotation.
The size fraction 14 by 35 mesh is then concentrated by several
methods such as spirals, tables, belt separators, and coarse
flotation. The minus 35 plus 150 mesh fraction is dewatered
to approximately 60 percent solids, then placed in vertical
agitated conditioning tanks. Reagents of the fatty acid type
are added to the conditioning tanks to allow the flotation
reagent to coat the mineral particles. The conditioned pulp
is then floated in the rougher flotation section with the
tailings going directly to the disposal area. The concentrate
(froth product) is then put into a conditioning tank with
sulfuric acid to remove the reagents added in the rougher
flotation step. The concentrate is then washed. The cleaned
and washed concentrate is then reagentized with an amine
to float the silica away from the phosphate. The silica pro-
duct thus floated is then discarded, while the fraction that
did not float is dewatered and sent to storage as a phosphate
rock concentrate.
The generalized reagent consumptions are given in Table 20.
188
-------
Washer plant showing
cyclones and hydro-
classifier
Washer plant showing the
cyclones and rake classifier
FICUBE 44 VIEWS OF A TYPICAL FLORIDA WASHER PLABT
189
-------
TABLE 20
GENERALIZED REAGENT CONSUMPTION III CONCEHTRATIHG
PHOSPHATE DY FLOTATION (V)
(Pounds per ton of ore)
Fatty Acid Flotation Step
NaOH - 0.3 to 1.5
Fatty Acids - 0.75 to 1.25
Fuel Oil - 1 to 3
Cleaning Step
K2CO4 - 0.5 to 3.0
Amine Flotation Step
IlaOH - 0.1 to 0.3
Amine - 0.1 to 0.3
Kerosene - 0.0 to 0.50
In general/ reagent consumption is determined by how well
the flotation feed has been deslined and the extent of condi-
tioning of the mineral surfaces.
The plant output to wet storage is analyzed for size, BPL,
Ma, K, F, and As. The products are stored by BPL content
and size. Come of the product is ground as a customer service
and some lower-grade material is blended to meet shipping
specifications. The common shipping grades in percent BPL
are 68, 72, 75, and 77. A small percentage of the phosphate
concentrate is calcined to remove organic material for use
in the production of furnace acid.
190
-------
EFFLUENTS FROM MIMING AND DENEFICIATION
The effluents produced in the mining and benef iciation of
phosphate rock are contained in the water suspensions leaving
the washer plant. These suspensions are the phosphate slimes
and the sands tailings. The amounts of these effluents rela-
tive to the water balance and water flov; rates for a typical
Florida plant operation processing 36,000 tons per calendar
day of phosphate matrix (approximately 6000 tons P2°5 Per
is given in Table 21. The flows presented correspond to a
water usage of about 13 gallons per minute per daily ton of
P2C>5 production assuming a phosphate matrix containing about
16.5% P°-
TABLE 21
WATER DISTRIBUTION FOR A TYPICAL FLORIDA
PLANT OPERATION ( S '
Gallons Per Percent of Total
Minute Water Used
Source for Water Used
Reclaimed from all ponds 57,200 74
Water from deep wells 20,000 ££
TOTAL 77,200 100
Estimated Water Usage
Mining (matrix) 10,000 13
Processing 22,000 28
Slimes disposal 39,000 51
Sand tailings 6/200 _ 8
TOTAL 77,200 100
It can be noted that the major effluent is that of the slimes
which contain a suspension of clays and very fine solids amount-
ing to about 4 to 6 percent by weight of the slimes stream.
A typical analysis of these slime solids is provided in Table
22.
As nreviously noted, these slimes are impounded in slimes
ponds to allow settling and clarification to occur. Clear
Water is returned from the ponds to the bencf iciation plant.
Figures 45 and 46 show views of slime ponds and water drainage
for recycle to the benef iciation plant.
191
-------
TABLE 22
APPROXIMATE MINERALOGIC AND CHEMICAL COMPOSITION
OP PHOSPHATE SLIME SOLIDS
(8)
Approximate Mineralogical ^Weight Composition
phosphate Slimes Solids
Mineral Percent
Carbonate fluorapatite 20-25
Quartz 30 - 35
Montmorillonite 20-25
Attapulgite 5-10
Wave1lite 4-6
Feldspar 2-3
Heavy minerals 2-3
Dolomite 1-2
Miscellaneous 0-1
Chemical Composition of Phosphate Slimes Solids
Chemical Typical Analyses, Range,
percent percent
P205 9.06 9 - 17
Si02 45.68 31 - 46
Fe203 3.98 3-7
A1203 8.51 6-18
CaO 14.00 14 - 23
MgO 1.13 1-2
CO2 0.80 0-1
F 0.87 0-1
LOI (1,000 C) 10.60 9-16
BPL 19.88 19 - 37
192
-------
FIGURE 45 VIEWS OF FLORIDA SLIME PONDS AND DIKES
193
-------
FIGURE 46 VIEW OF WATER DRAINAGE FROM SLIME PONDS
FOR RECIRCULATION TO PLANT
194
-------
ENVIRONMENTAL AND POLLUTION EFFECTS OF EFFLUENTS
FROM MINING AND BENEFICIATION
The environmental effects of the aqueous effluents from phos-
phate mining and beneficiation were investigated by R. C.
Specht in 1947 in relation to the waste disposal practices
of the Florida phosphate companies. The investigation was
conducted along the river basins of the Peace and Alafia Rivers
Extensive sampling was conducted and numerous tests were made
to determine the amount or effect of:
1. Turbidity
2. Dissolved oxygen and biochemical oxygen demand
3. Alkalinity
4. Toxicity of flotation chemicals.
The turbidity of the stream waters was found to increase by
about 2.5 percent as a result of process effluents; however,
tests with fish of several varieties showed no effect on fish
life. The effect of turbidity on plankton growth was not
investigated.
The dissolved oxygen (DO) and biochemical oxygen demand (BOD)
analyses indicated that the dissolved oxygen content of the
stream was increased and thereby improved, and the biochemical
oxygen demand was not changed significantly by the phosphate
operation effluents.
No change in alkalinity of the receiving streams was detected
except in one case where the effluents were beneficial in
Neutralizing trace acidity in the stream.
The only chemical used in the operation that could be toxic
at existing concentrations was tall oil; however, no tall
oil was found in the effluent waters. The tall oil is used
to coat the phosphate mineral to promote flotation, and tends
to go with the phosphate concentrate. However, any residual
tall oil will tend to be precipitated by the calcium in the
streams. The kerosene, acids, amines, and caustic did not
appear in sufficient amount to be toxic to the fish tested.
The conclusion of the Specht study was that aside from an
Apparent increase in algae growth, the effluents from the
Phosphate operations are not harmful to the environment.
^igure 47 shows a view of plant growth in an old slime pond.
°ne fact that should be noted here is that because of the
large water consumption necessary in phosphate operations
this water usage might lower the water table, thereby causing
contamination of the aquifer. One such case has been recorded;
however, available details are insufficient to draw any con-
fusions. *« 'W
195
-------
FIGURE 47 VIEW OF A SLIME POND SHOWING
GROWTH OF WATER PLANTS
196
-------
CONTROL AND TREATMENT OF AQUEOUS EFFLUENTS
The aqueous effluents from mining and benef iciation of phos-
phate rock are presently controlled by the use of large settl-
ing ponds. The slimes are allowed to settle and the clear
water is drawn off and returned to the plant for reuse. The
slime ponds are built to rigid specifications and surveil-
lance of the pond areas is maintained.*8' The restrictions con-
cerning pond construction in Florida are enforced by both
the state and phosphate mining companies. The effectiveness
of present effluent control practices is good. There has
been no major slime pond failure for the last 3 years.
Means to significantly improve effluent control are not apparent
from the result of work to date. Considerable research has
been conducted by the Bureau of Mines/ TVA, and the individual
companies on the benef iciation or dewatering of the slimes
and possible uses for the slime solids. This work has not
led to practical results. At present, only a minor amount
of basic research is being conducted on means to alleviate
the slimes disposal problem. Most of the research being con-
ducted is on variations in the physical handling of the slimes
to reduce costs and water losses.
An example of work currently being performed in the industry nn.
is the program at American Cyanamid Company's Brewster Plant. li0'
American Cyanamid conducted tests for several months, trying
to determine a simple economical way of reducing the water
content of the slimes and the volume they occupied. A trench
dug that was 5,000 feet in length, 136 feet wide, and
I
feet deep. I' The slimes were poured into the trench
allowed to settle to a density of approximately 25 percent
solids. Sand was then poured into the trench, and within
a few hours water was noted to be forced to the surface. This
water was drained off, leaving the resulting mixture approxi-
^ately 80 to 90 percent solids. American Cyanamid found that
the volume occupied by the sand and slime mixture was the
same or less than the volume occupied by the slimes. This
Method of slime disposal reduces the slime handling costs
«y approximately one-third and returns the land to useful
Purposes. Photographs showing two views of this operation
shown in Figure 48. This procedure appears to provide
definite improvement in handling a part of the slimes; how-
this approach is only a partial answer because the amount
°f slime that is generated is greater than the amount of sand
tails normally produced in the mining operation.
*he general practice of the Florida phosphate companies is
to monitor the streams below their plants for water flow,
8Uspended solids, phosphate, and fluorine. These streams
197
-------
P^ V
?*•
FIGURE 48 VIEWS OF AMERICAN CYANAMID'S NEW STAGE-FILLING
TECHNIQUE FOR SLIME DISPOSAL
198
-------
are checked on a daily basis so that any rise above normal
in any one of the measured variables can be quickly detected
and the source traced down. New methods to continuously
monitor these variables are now being studied, including
specific ion electrodes, direct reading spectrographs, chem-
iluminescence and conductivity.
ECONOMIC ASPECTS OF EFFLUENT CONTROL
Control of the primary effluent from the phosphate rock mining
and beneficiation is accomplished by building and maintaining
the slime ponds. A significant portion of the total operating
costs is related to this part of the operation. Table 23 gives
a breakdown of the estimated total capital and operating costs
for a typical slime disposal system.(°J It is evident from
Table 23 that, to be of significant near-term interest to pre-
sent operators, any new procedure for slimes and tailings
disposal must be accomplished at costs not greatly exceeding
$0.25 per ton of phosphate rock product or about $0.50 per
ton of dry slime solids.
TECHNICAL ASPECTS OF IMPROVED EFFLUENT CONTROL
It is significant that approximately one-third of all phosphate
value mined is now lost in the slimes. Currently, the slimes
produced annually from the Florida operations contain more
phosphate than was produced as salable product in 1960. The
magnitude and severity of the slimes problem in Florida can
be estimated from the fact that almost 500,000,000 gallons of
Water per day are associated with the rejected phosphate
slimes.
The reason this waste continues is that, despite substantial
study, a low-cost method to dewater phosphate slimes has not
been developed. As previously noted, it is estimated that
disposal in ponds costs about $0.25 per ton of contained solids.
This is less than $0.10 per 1,000 gallons of water contained.
The characteristics of the slimes make them extremely diffi-
cult to process and there is no method for dewatering proposed
to date that has shown significant promise or has been developed
sufficiently to meet present economic limitations.
The problem is technically difficult because the solid particles
are almost all of colloidal size and extremely hydrophilic.
The solids are extremely slow to settle. Even after years
of settling, the solids content of the slime ponds does not
exceed 30 weight percent. At 30 percent solids, the phosphate
slimes are almost of jelly-like consistency. Centrifuging
Produces material containing only 35 percent solids.
199
-------
TABLE 23
CAPITAL AND OPERATING COSTS FOR SLIMES DISPOSAL
Capital Costs
Slime pump stations (3 complete units)
Pond spillways (10 required)
Excavator (for ditching)
Pumps (4 recirculating water)
Pipeline
Plant facilities
Plant utilities
Total cost (tax and insurance base)
Interest during construction
Subtotal (depreciation)
Working Capital
Total capital investment
(8)
Capital
Investment
$700,000
80,000
60,000
160,000
200,000
72,000
76,000
$1,348,000
33,700
1,381,700
280,200
$1,661,900
Operating Costs
Unit cost Total
Direct cost:
Raw materials and utilities:
Fuel - 13.3 gal x $0.15 x 40 hr x
52 weeks
Power - 2,150 kwhr/hr x 8,760 hr/yr x
$0.01 kwhr
Direct labort
16 man-hr/day - $3.00/man-hr x 365
day/yr
Supervision - 15 percent of labor
Maintenance:.
5 men - $7,000/yr
Supervision - 20 percent of main-
tenance labor
Material -
$4,100
188,300
17,500
2,600
35,000
7,000
17,500
Cost per
ton of
product1
Payroll overhead - 25 percent of payroll
Operating supplies - 20 percent of maintenance
Total direct cost
$192,400 $0.045
20,100
59,500
15,500
11.900
299,400
36,600
40,400
Indirect Cost:
40 percent of labor, maintenance, £ supplies
Fixed Cost:
Taxes and insurance - 3 percent of total cost
Depreciation - 5 percent of subtotal for deprecia-
tion and 10 percent for excavator 72,100
Dam construction 938,000
Total operating cost 1/386,500
Credit for recirculated water 358,100
Net operating cost $1,028,400
.005
.014
.004
.003
.071
.009
.010
.017
.223
.330
.085
•'•Cost per ton is based on 4,205,000 tons of product.
200
-------
The most immediate need is to develop means to dewater the
slime economically. Although the slimes contain substantial
phosphate values that might be recovered, utilization of slimes
will probably depend on initial dewatering. Even if the phos-
phate values could be recovered today by selective removal
from the slimes, disposal of the remaining slimes would still
be a major problem that must be resolved by means other than
the slime pond.
A discussion of the technology of treating the phosphate slimes
is not meaningful without considering the economics. One
consideration is that utilization of contained phosphate value
would almost certainly require drying. It is estimated that
drying fuel costs alone would be at least $3-4 per ton of
dry 30 percent P2°5 material. Considerably higher costs would
be incurred if multiple-effect evaporation could not be employed.
Actually, it is doubtful that the slimes could be dried for
less than $8 per ton equivalent to 30 percent P2Oc material.
Moreover, the P2°c content of the slime is less than half
that of the rock Being sold, and some means to upgrade to
over 30 percent P2°5 would be required. Florida phosphate
can be produced for considerably less than this estimated
drying cost. In view of the availability of lower grade Florida
and Western phosphates that would be an economically more
favorable source of P2°5 than the slimes, there appears to
be virtually no possibility that the P2G5 can be economically
recovered from the slimes.
Thus, the phosphate slimes problem is really a problem in
disposal. It has been estimated that the cost of slime pond
disposal in Florida is about $0.25 per ton of contained solids
or about $0.10 per 1,000 gallons of slime. Therefore, for
an alternate method to be of near term interest, it must offer
the possibility of dewatering at costs well below about $0.50
per ton of solids or $0.10 per 1,000 gallons of contained
water.
Experience in prior related investigations suggests that no
single process step can offer optimum processing. A combina-
tion of steps would probably be most economical for the range
of concentration of interest.
REVIEW OF PRIOR RESEARCH RELATED TO PHOSPHATESLIMES DISPOSAL
The Tennessee Valley Authority worked for many years on various
methods of partial dewatering of slimes. Appendix A provides
an outline of their work. Although no economic method of de-
watering was found, three methods were encouraging, viz.:
201
-------
1. Thickening with slow stirring (to 30 percent solids)
2. Freezing (solids settled rapidly to 45 percent,
filtered rapidly to 60 percent)
3. Electrophoresis (to 50 percent solids).
The United States Bureau of Mines did extensive research on
combined dewatering and recovery of phosphate and/or other
values from the slimes by physical beneficiation or chemical
extraction, I12' Appendix B outlines the scope of their program.
The work by Stanczyk, et.al., on dewatering by combination
of electro-osmosis and vacuum filtration '13) is also signi-
ficant.
In a pilot plant run on electrical dewatering of phosphate
tailing, Houston, et.al.,^14) report power consumption of
65 kwhr per ton of solids removed, with a deposition rate
of 8 Ib per sq ft per hour. A rotating drum type of continuous
dewatering machine with a cast iron anode was used. The process
is not yet competitive with the use of settling pond.
Elaborate work by Thompson ^15) and later by Thompson and Vilbrant*16^
on use of ultrasonic energy for coagulation and settling of
solids in phosphate tailing illustrates the range of concepts
tried for a solution of the problem. Near the isoelectric
point, mild intensity (below caviation level of 0.3 volt-
ampere per sq cm) insonation increased the coagulation and
subsequent settling rate by a factor of 5 as compared to a
control without ultrasonic energy. Intense insonation above
certain limits had the opposite effect, increasing the thixo-
tropic gel structure and, consequently, retarded settling.
Studies by Sun and Smit(17) On flotation reclamation of phos-
phate from slime produced some interesting results. Phosphate
flotation of Bartow washer slime by fatty acid (Pamak-1) was
improved by classification at 10.0, 5.0, 2.5, and 1.25 microns.
Flotation effectiveness was approximately proportional to
size. As would be expected, the flotation efficiency decreased
drastically for particles under 5-micron size.
The elegant analysis by Derjaguin and Dukhin * ' on the mechanics
of fine-particle flotation reveals some of the difficulties
involved in colloid flotation.
(19)
Studies by Greene and Duke x ' on selective flotation of ultra-
fines as applied to both the Florida and Tennessee phosphatic
slime demonstrated some practical possibilities. Addition
of extraneous material like sulfur as a carrier improved flo-
tation of phosphates from the Florida slime. Tennessee slime
did not respond to such treatment when sulfur was used as
the carrier.
202
-------
A recent patent by the same authors ' describes a process
for tumbling phosphate slimes with 10-35 mesh angular silica
sand for several hours and then floating the pulp after removing
silica sands by screening. The process shows significant
improvement in flotation of phosphate fines (>9 microns) .
A 30 percent increase in recovery of phosphate (85.4 percent
recovery) was demonstrated in comparison with operations without
the sand tumbling.
Haseman's1 ' proposal of selective flocculation of colloidal
phosphate with starch created considerable interest which
resulted in more intensive study of flocculation, subsidence
and filtration of phosphate slimes by La Her, et.al. (22-27)
They have noted significant improvement in filtration by addi-
tion of potato starch. Appendix C shows comparative results
for the 12 most promising flocculant gums and polyelectrolytes
in a group of 150 tested. Dow Chemical's "Separan 2610" showed
the best results. Theoretically, it seems highly doubtful
that dewatering with hydrophilic sols would succeed beyond
20-30 percent solids.
(28)
Mellgren and Shergold l ' describe a method for recovering
ultrafine mineral particles by extraction with an organic
phase. By using a suitable surfactant, the contact angle
of the solids in the liquid phase can be modified to let the
mineral particles concentrate either at the liquid-liquid ,-Q.
interface or in the organic phase. Tests by Shulman and Lejas^ '
on an emulsion stabilized by barium sulfate confirm the feasi-
bility of the principle.
Takauwa and Takamori ' have investigated the possibility
of a phase inversion technique for quick appraisal of laboratory
flotation results. Agglomeration of fine solids in liquid
suspension by preferentially wetting the solids with another
liquid which acts as a bridging agent between the particles
has been the subject of a patent. <3D The technique was fur-
ther elaborated in a paper by Pernand, et.al. I32)
Of significance to the problem of colloid flotation of phos-
phatic slime are works by Clanton, <33) Magoffin and Clanton,
and Grieves, et.al. (35-41)
Clanton and co-worker's attempts were directed mainly at the
removal of sols from textile wastes. Ferric oxide sols were
floated with kerosene sodium oleate combination to almost 100
percent removal. Similar tests were unsuccessful in floating
chromium oxide sols.
Grieves, et.al. 's work centered on removal of surfactants
from liquids by foam fractionation. While studying the effect
of suspended material on foam separation of anionic and
203
-------
(37)
cationic surfactants, ' they observed that colloidal particles
could be removed along with the foam. Using this principle,
they removed ferric hydroxide sols with dodecalsodium sul-
fate.'38' The volume of foam removed was a significant weight
fraction of the original mass, but the recovery of the colloids
was not sufficient to produce fully clarified water,
RECOMMENDATIONS FOR RESEARCH TO DEVELOP IMPROVED MEANS FOR
DISPOSAL OF EFFLUENTS FROM PHOSPHATE MINING AND BENEFICIATION
It is recommended that research be undertaken to better estab-
lish the merits and technical-economic limitations of various
methods known to provide some basis for successful development
of an improved slimes-disposal system. It is anticipated that,
upon completion of the effort, the most profitable directions
for emphasis on subsequent development programs would be known.
Such work should result in the development of an economic de-
watering process, through analysis of the slimes system and
definition of the merits and limitations of processing known
to be technically effective in slimes recovery and dewatering.
The promising process concepts that can be applied to concen-
trate and dewater the slimes can be classified as: (1) surface
chemical, (2) electrochemical, (3) thermal, and (4) combina-
tions of these. Comments follow concerning the background and
research effort believed warranted within the classifications
listed.
(1) Surface Chemical Research, in general, the lowest cost
separation method will be based on physical separation. How-
ever, a limitation on physical separation in colloid suspension
is dictated by surface chemistry factors. To better understand
flocculation, filtration, and flotation processes, it will
be necessary to know more about reagent adsorption on the
slime components and capability to alter or eliminate the
hydrophilic nature of the clays. A key focus for the work
on surface chemistry should be to determine possibilities
to reduce requirements for flocculant reagents by use of sel-
ective adsorption on the surface of components present in
the slimes in only minor amounts or by selective adsorption
on only certain types of adsorption sites. A practical objec-
tive would be to achieve satisfactory flocculation using 1-
2 parts per million (ppm) quantities of flocculants rather
than the 100-200 ppm range required for flocculation with
polyaerylamides and potato starches.
It was noted in preliminary experiments recently made at Battelle
that chelating agents can be used to render the slimes hydrophobicr
and thus floatable. The amount of reagent required, from
a practical standpoint, was excessive in these experiments.
204
-------
However, the possibility that certain metal ion sites can
be chelated selectively with trace quantities of agents which
can bridge to a flotation collector deserves attention as
a possible method to reduce reagent requirements for initial
partial dewatering by bulk colloid flotation or selective
separation.
In summary, the required activity for surface chemistry research
is to study adsorption of reagents which promote flocculation
and/or alter the hydrophilic nature of the slimes solids.
The objective of this work would be to better understand factors
influencing the minimum reagent requirements.
(2) Electrochemical Research. Application of an electric
field to a suspension of fine particles causes directional
particle migration. Solids could thus be removed from suspen-
sion, either in bulk or perhaps selectively. This technology
is termed electrophoresis. It is also possible to electrochem-
ical ly dewater a suspension by causing movement of the medium
(water) through a membrane by force of an electrical field
(electro-osmosis).
(14)
Electrophoresis: Houston, et.al. reported thickening of
Tennessee phosphate slimes from 12.5 to 50 percent solid con-
tent by electrophoresis. The overall cost of dewatering Tenne-
ssee phosphate slimes by electrophoresis was estimated in
1952 by Tyler and Waggaman to be $2 per ton of dry slime versus
$1 per ton for pond disposal. Since that time the technology
on electrophoresis has advanced and the costs can probably
be reduced substantially.
Work has shown that when surface properties of these "slimes"
are modified (for example, by addition of a dispersant such
as sodium silicate) the efficiency of electrophoresis increases.
Superimposition of vacuum filtration with drum-type electro-
phoretic apparatus enhances the efficiency of the operation
and can reduce the operating cost.
However, there are no published data on the dewatering of
Florida phosphate slimes by electrophoresis. Experiments
to obtain basic data to enable determination of optimum oper-
ating conditions and process costs are, therefore, warranted.
Electro-Osmosis: An attractive feature of electro-osmosis
is that virtually clear water can be obtained for immediate
reuse. Stanczyk and Feld^13) of the U.S. Bureau of Mines
compared dewatering of Bartow phosphate slimes by electro-
osmosis, vacuum filtration, and a combination of the two pro-
cesses. The power requirements for the combination were
205
-------
found to be less than for electro-osmosis alone, and a function
of applied voltage and solids content. Under the conditions
studied, the power requirement for increasing the slimes solids
from 15 to 35 percent was equivalent to about 15 kwhr per
ton of dry slime solids.
(3) Thermal Processing Freezing. It has been known for some
time that simple freezing and thawing of phosphate slimes
causes irreversible changes in hydration so that solids will
settle rapidly and to a relatively high solids content without
addition of flocculants. Experiments made at Battelle with
fresh Bartow slime samples demonstrated fast settling to 36
percent solids from a slime with an initial solids content of
2.7 percent. Pressure filtration then yielded a cake of 43
percent solids.
The economics of a slimes dewatering process based on freezing
at first seem prohibitive. However, recovery of fresh potable
water by freeze-desalting of sea water has received extensive
study. Published cost estimates predict, perhaps optimistically
costs for a 1-million-gallon-per-day freeze-desalting facility
as $0.40 per 1,000 gallons, t4-2) A representative of the Office
of Saline Water has stated that freezing appears to be the
most economical desalting process up to capacities of about
1 million gallons per day.*
It should be recognized that in freeze-desalting there is
a brine reject containing not more than about 7 percent salt.
Thus, freeze-desalting must process almost twice the amount
of water that is produced. This would not be the case for
slimes processing. Also, a freeze-desalting process is a
far more complex process than the simple freezing and melting
required for slimes. With slime processing there would be
no need for washing the ice to remove occluded brine. If
the $0.40 per 1,000 gallon estimate can be achieved for desalt-
ing, it is anticipated that costs less than $0.20 per 1,000
gallons, or about $1.00 per ton of dry slime solids, are possi-
ble. It is noteworthy that large-scale operations have been
installed in England for freeze-based dewatering of water
filtration and sewage sludge.
Careful study of the economic potential of freeze-based de-
watering applied to phosphate slimes appears warranted be-
cause it provides potential for increased water reuse (fatty
acids and amines from flotation would be removed and water
recycled to flotation without adverse effect} and use of the
dewatered slime solids for light-weight aggregate.
*Private communication.
206
-------
COORDINATION OF RECOMMENDED RESEARCH
Through discussion with representatives of the producing com-
panies and various government agencies, it is well recognized
that efforts are presently being made that relate to resolu-
tion of this difficult technical-economic problem. In addition
to the basic objectives of the industry to improve phosphate
recovery, minimize water usage, and minimize the land require-
ments and land maintenance costs, there are factors involved
pertinent to state and federal agencies concerned with land
and water resource management, water pollution, and solids
waste disposal. Knowledge already developed and current research
in progress could provide meaningful information if there
is a coordinated program intended to resolve the slimes prob-
lem. Such coordinated effort is considered by Battelle to
be the best procedure, if not the only procedure, that can
lead to a near-term resolution of the problem. Implementation
of a coordinated research program is therefore recommended.
207
-------
REFERENCES
1. Enigh, G. P., Mining Engineer, 23^ 1 59 (1970).
2. Levitsky, S. L., "Sulphur Phosphate and Potash", Mining
Congress Journal (February 1970).
3. Sweeney, J. W. and Hasslacher, R. N., "The Phosphate
Industry in the Southeastern United States and Its Rela-
tionship to the World Mineral Fertilizer Demand", Bureau
of Mines 1C 8459 (1970).
4. T.V.A., "New Developments in Fertilizer Technology"
(October 1970).
5. Town, J. W., Gruzensky, W. G., and Banker, P. E., "Effects
of Certain Mineralogical Features on Phosphate Recovery",
Bureau of Mines RI 6749 (1966).
6. Harre, E. A., "Fertilizer Trends 1969", T.V.A. (1969).
7. Beall, J. V. and Merritt, P. C., "Phosphate and Potash
Minerals to Feed the World", Mining Engineer (October 1966).
8. Boyle, J. R., "Waste Disposal Costs of a Florida Phosphate
Operation", Bureau of Mines 1C 8404 (1969).
9. Specht, R. C., "Effect of Waste Disposal of the Pebble
Phosphate Rock Industry in Florida on Condition of Receiving
Streams", AIME Trans, Vol. 87 (July 1950).
10. Timberlake, R. C., "Building Land With Phosphate Wastes",
Florida Section AIME (November 1969).
11. Business Week, January 2, 1971.
12. Gary, J. H., Feld, I. L., and Davis E. G., "Chemical and
Physical Beneficiation of Florida Phosphate Slimes", Bureau
of Mines RI 6163 (1963).
13. Stanczyk, M. H., and Feld, I. L., "Electro-Dewatering of
Florida Phosphate Rock Slime", Bureau of Mines RI 6451 (1964)
14. Houston, E. C., Jones, V. J., and Powell, R. L., "Electrical
Dewatering Phosphate Tailing", Trans AIME 184, pp. 365-369
(October 1949).
15. Thompson, D., "Ultrasonic Coagulation of Phosphate Tailing",
Bull Virginia Poly Tech Inst, Eng. Expt Stn. Series No. 75
63, 5 (July 1950).
209
-------
16. Thompson, D., and Vilbrant, F. C., "Effect of Ultrasonic
Energy on Settling of Solids in Phosphate Tailing", Indust
and Eng Chem, 4£ 1172-1180 (June 1954).
17. Sun, Shiou-Chuan, and Smit, F. J., "Reclamation of Phosphate
from a Florida Washer Slime by Flotation", Trans AIME, 225
454-461 (December 1963).
18. Derjaguin, B. V., and Dukhin, S. S., "Theory of Flotation
of Small and Medium-Size Particles", Trans Inst Min Met,
70 221-246 (1960).
19. Greene, E. W., and Duke, J. B., "Selective Froth Flotation
of Ultrafine Minerals or Slime", Trans Min Eng, 225 389-395
(December 1962).
20. Duke, J. B., and Green, E. W., "Preparation of Ore Slimes
for Froth Flotation", U.S. Patent No. 3,259,326 (July 1966).
21. Baseman, J. F., "Selective Flocculation of Colloidal Phos-
phate in Presence of Clay", U.S. Patent No. 2,660,303 (November
24, 1953).
22. La Mer, V. K. and Smellie, R. H., "Flocculation, Subsidence
and Filtration of Phosphate Slimes", J. Colloid Sci., 11
704-709 (1956).
23. Op cit 710-719.
24. Op cit 719-731.
25. La Mer, V. K., Smellie, R. H., and Pui-Kum Lee, "Flocculation,
Subsidence, and Filtration of Phosphate Slimes", J. Colloid
Sci, 1£ 230-239 (1957).
26. Op cit 566-574.
27. Smellie, R. H., and La Mer, V. K., "Flocculation, Subsidence,
and Filtration of Phosphate Slimes", J. Colloid Sci., 23
589-599 (1958).
28. Mellgren, O. and Shergold, H. L., "Method for Recovering
Ultrafine Mineral Particles by Extraction with an Organic
Phase", Trans Inst Min Met 7J5 C 267-268 (1966).
29. Schulman, J.H. and Leja, J., "Control of Contact Angles
at the Oil-Water-Solid Interfaces", Trans Faraday Soc,
50 598-605 (1954).
210
-------
30. Takakuwa, T. and Takamori, T., "Phase Inversion Method for
Flotation Study", Proc 6th Int Min Process Congress, Cannes
(English Edn) 1-10, Peragamon Press, New York (1965).
31. Puddington, I.E., Pernand, J. R., and Smith, H. M., "Aggrega-
tion and Separation of Solids Suspended in Liquid Medium",
Canada Patent No. 647,378 (August 28, 1962).
32. Pernand, J. R., Smith, H. M., and Puddington, I. E., "Spher-
ical Agglomeration of Solids in Liquid Suspension", Can.
J. Chem. Eng., 94-97 (April 1961).
33. Clanton, R. B., "Fundamental Properties of Textile Wastes
V, Flotation from Ferric Oxide Sols", Textile Research,
81 301-4 (1938).
34. Magoffin, J. E. and Clanton, B. R., "Fundamental Properties
of Textile Wastes VIII, The Flotation of Colloidal Suspen-
sions", Textile Research, £357-63 (1938).
35. Grieves, R. B. and Bhattacharya, D., "Foam Fractionation
of Colloid-Surfactant System", AIChE J., 11 (2), 274-279
(March 1965).
36. Grieves, R.B., Bhattacharya, D., and Crandall, C. J.,
"Foam Separation of Colloidal Particulates", J Appl. Chem.,
r? 163-168 (June 1967).
37. Grieves, R. B. and Bhattacharya, D., "Effect of Colloidal
Particulates on Foam Fractionation", Nature, London, 207
476 (1965).
38. Grieves, R. B. and Bhattacharya, D., "Foam Fractionation
of the Ferric Oxide-Dodecylsodium Sulphate System", Can.
J. Chem. Eng. 43 286 (1965).
39. Staff, "Particles Control Foam", Chem. & Eng. News, 22
(September 1966).
40. Grieves, R. B. and Wilson, T. E., "Flotation of Dichromate
Ion", Nature, London, 205 1066 (1965).
41. Grieves, R. B. and Ettelt, G. A., "Continuous Dissolved-
Air Ion Flotation of Hexavalent Chromium", AIChE J., 13_
(6), 1167-1171 (November 1967).
42. Saline Water Conversion Report, United States Department
of the Interior (1966).
211
-------
43. Anon., "Board Proposes Restrictions On Water Usage
By Lee Creek Operations/1 Eng. Mining J., 169 (October)
pp. 109-110 (1968).
44. J. E. Hobbie, "Phosphorus Concentrations In the Pamlico
River Estuary of North Carolina," North Carolina
Water Resources Research Institute Report No. 33,
(March 1970).
212
-------
LITERATURE NOT CITED
1. Spedden, H. R., "Impact of Environmental Controls on
Nonferrous Metals Extraction", Mining Congress Journal
(December 1970).
2. Davenport, J. E. and Watkins, S. C., "Beneficiation of
Florida Pebble Phosphate Slime", I&EC, Vol. 8, p. 533
(October 1969).
3. Davenport, J.E., Carroll, Frank, and Tarbutton, Grady,
"Beneficiation of Florida Aluminum Phosphate Ore", I&EC,
Vol. 46 (August 1954).
4. Davenport, J. E. and Haseman, J. F., "Laboratory-Scale
Flotation of Brown Rock Phosphate", AIME Tech Pub N-2239
(November 1947).
5. Ruhlman, R. E., "Phosphate Rock", Bureau of Mines 1C 7814
(January 1958).
6. Waggamen, W. H. and Ruhlman, E. R., "Conversation Problems
of the Phosphate Industry", Ind Eng Chem, Vol 48 (1956).
7. Waggamen, W. H. and Bell, R. E., "Western Phosphates", Ind
Eng Chem, Vol 42 (1950).
8. Swanson, R. W., McKelvey, V. E., and Sheldon, R. P.,
"Progress Report on Investigations of Western Phosphate
Deposits", Geol Survey Giro 297 (1953).
9. Smith, R. W. and Whitlatch, G. I., "The Phosphate Resources
of Tennessee", Tenn. Dept. of Cons., Div. of Geol. (1940).
10. Kibler, D. G., Jr., "Mining and Preparation of Florida
Hard-Rock Phosphate", Min & Met, Vol 25, No. 454 (1944).
11. Grissom, R. J., "Mining and Washing Phosphate Rock in
Tennessee", Min & Met, Vol 25, No. 454 (1944).
12. Butner, D. W., "Phosphates of Tennessee", Am Chem Soc Mon.
34, ided, Reinhold Pub Corp, New York, New York (1952).
13. Davenport, J. E., Carroll, F., Kieffer, G. W., and Watkins,
S. C., "Beneficiation of Florida Hard-Rock Phosphates",
I&EC, Vol 8 (October 1969).
14. Cox, J. L. and Falkie, T. V., "Phosphate Wastes", Mineral
Waste Utilization Symposium, IIT Research Inst. (March 1968)
213
-------
15. U.S. Department of the Interior, "Surface Mining and Our
Environment", A Special Report to the Nation (1967).
16. Hee, Olman, "A Statistical Analysis of U.S. Demand for
Phosphate Rock, Potash, and Nitrogen", Bureau of Mines
1C 8418 (1969).
17. Town, J. W., Clark, C. W., Sanders, C. W., and Sullivan,
E. E., "Batch and Continuous-Circuit Beneficiation of
Western Phosphate Ore", Bureau of Mines RI 6930 (1967).
18. Mabie, C. P. and Hess, H. D., "Phosphate Study and Classi-
fication of Western Phosphate Ores", Bureau of Mines
RI 6468 (1964).
19. Coffman, J. S. and Service, A. L., "An Evaluation of the
Western Phosphate Industry and Its Resources", Part 4,
Bureau of Mines RI 6934 (1967).
20. Service, A. L. and Petersen, N. S., "An Evaluation of the
Western Phosphate Industry and Its Resources", Part 5, Bureau
of Mines RI 6935 (1967).
21. Custred, V. K., "New Mining Methods Rehabilitate Florida's
Strip Mines", Mining Engineer (April 1963).
22. Corrigan, P. A., Lyons, V. E., Barnes, G. D., and Hall,
F. G., "Conductivity Measurements Monitor Waste Streams",
Environmental Science and Technology, Vol 4 (February 1970).
23. Davenport, J. E., Kistfer, G. W., and Brown, E. H.,
"Disposal of Phosphate Tailings", TVA Division of Chemical
Development Research Branch (1953).
24. "Fertilizer Producers See Break-Even Year", C&EN (February
1971).
25. Heylin, M., "Pollution Control Instrumentation", C&EN
(February 1971).
26. Town, J. W. and Sanker, P. E., "Evaluation of Phosphate
Fines From Southeastern Idaho", Bureau of Mines RI 7205
(1968).
27. Harre, E. A., Kennedy, F. M., Hignett, T. P., and McCune,
D. L., "Estimated World Fertilizer Production Capacity as
Related to Future Needs 1970 to 1975", TVA (June 1970).
28. Mukai, S., "The Floatability of Fine Particles I", Journal
of the Mining Inst. Japan, 68, 313-18 (July 1952).
214
-------
29. Mukai, S., "The Ploatability of Fine Particles II", Journal
of the Mining Inst. Japan, 69 291-295 (August 1953).
30. Mukai, S., and Masakiajo, M., "Effect of Electrolytes on
the Pine Particles of Zincblende", Journal of the Mining
and Metallurgical Society of Japan, 73 289-294 (May 1957).
31. Jaycock, M. J. and Ottewill, R. H., "Adsorption of Ionic
Surface Active Agents by Charged Solids", Trans. Inst. Min.
Met., 67_ 497 (1963).
32. Creighton, H. J., Principles and Application of Electro-
chemistry, John Wiley and Sons, Inc., New York, New York
1^ 158 (1935) .
33. Schwerin, B. V., U.S. Patent No. 1,133,967 (March 1915).
34. Sollner, K., "Sonic and Ultrasonic Waves in Colloid Chemistry",
Colloid Chemistry, J. Alexander Ed. 5_ 337-42, Reinhold
Publishing Company, New York, New York, 1st Ed. (1944).
35. Sollner, K. and Bondy, C., "Mechanism of Coagulation by
Ultrasonic Wave", Trans. Faraday Soc., 32 616 (1936).
36. Tyler, P. M. and Waggaman, W. H. "Report on the Possible
Utilization of Phosphate Rock Slime", 31 (June 29, 1953).
37. Matijevic, E., and KerKer, M., "The Charge of Some Heter-
opoly Anions in Aqueous Solutions as Determined by
Coagulation Effect", J. Phys. Chem., 62 1271 (1958).
38. Matijevic, E. and KerKer, M., "Detection of Metal Ion
Hydrolysis by Coagulation III. Aluminum", J. Phys.
Chem., 65_ 826 (1961).
39. Matijevic, E. and KerKer, M., 148th Am. Chem. Soc. Meeting
(September 1964).
40. La Mer, V. K. and Healy, T. W., "Adsorption-Flocculation
Reactions of Molecules at the Solid-Liquid Interface",
Review of Pure and Applied Chem., 13 112 (1964).
41. Rutz, L. O.and Goldberger, W. M., "Review of Ion and
Precipitate Flotation", Battelle Memorial Institute Report
(December 31, 1967).
42. Tyler, P. M. and Waggaman, W. H., "Phosphatic Slime",
IEC, 46 1052 (May 1954).
215
-------
43. Meloy/ T. P., "The Treatment of Fine Particles During Flota-
tion", Froth Flotation (50th Anniversary Vol.) 247-257,
Published by AIME, New York, New York (1962).
44. Kaufman, A. and Nadler, M., "Water Use in the Mineral Industry",
Bureau of Mines 1C 8285 (1966).
216
-------
APPENDIX A
OUTLINE OF VARIOUS BASIC PROCESSES AND VARIATIONS TRIED
BY TVA FOR PARTIAL DEWATERING AND SETTLING OF SLIME
217
-------
APPENDIX A
OUTLINE OF VARIOUS BASIC PROCESSES AND VARIATIONS TRIED
BY TVA FOR PARTIAL DEWATERING AND SETTLING OF SLIME*
(After Tyler & Waggaman)
SEDIMENTATION
Chemical treatments
Dispersants
Plocculants
Modifiers, designed to delay gel formation or to produce
more compressible gel
Heteropolar organic compounds to render clay hydrophobic
Positively charged in organic sols—e.g., aluminum and
ferric oxide sols
Dyes, selectively adsorbed by clay
Dissimilar metal powders, galvanic action
Agents for removing calcium ion from solution, by
precipitation
Physical Treatments
Ultrasonic irradiation
Freezing
Weighting with sand or fly ash
Dehydration, concentrated salt solutions/ extraction
with ehter
Evacuation, to remove dissolved gases
Magnetic field to promote settling
Heating matrix or treating with silicone prior to washing
to prevent hydration of the clay
THICKENING, WITH SLOW STIRRING
Rate of stirring
Type of stirring element
Presettling
Chemical treatments
FILTRATION
Filter aids
Heat
Chemical treatment
* Tyler, P.M., and Waggaman, Wm.H., "Phosphatic Slime, A
Potential Mineral Asset", Indust. & Eng. Chem. 46,
1049-1056 (May 1954).
218
-------
CENTRIFUGING
Chemical treatment
DRYING
High temperature
Ambient temperature
Forcing compressed air through suspension
ELECTRICAL METHOD
Electrophoresis, direct current
Electrode spacing
Voltage gradient
Intermittent current
Chemical treatment
Electrophoresis combined with sedimentation
Electrophoresis combined with filtration
Alternating current
Voltage
Frequency
Magnetically induced current, by flow of suspension through
a magnetic field
219
-------
APPENDIX B
OUTLINE OF DEWATERING, CHEMICAL, AND PHYSICAL
BENEFICIATION OF FLORIDA PHOSPHATE SLIMES BY USBM
220
-------
APPENDIX B
OUTLINE OF DEWATERING, CHEMICAL, AND PHYSICAL
BENEFICIATION OF FLORIDA PHOSPHATE SLIMES BY
I. DEWATERING
(a) Slime Flocculation and Thickening: 29 flocculants
were tested for static flocculataon tests. The maxi-
mum settling over a 24-hour period was obtained with
citric acid (to 10% solid).
(b) Pressure Filtration; Filtration tests carried out in
the laboratory at pressures ranging from 10 to 50 psig
and 2 to 40 pounds of filter aid (diatomite). Also
temperature was varied from 30 to 165 °C.
(c) Extraction of organic material before filtering;
Organic solvents like acetone, ethyl alcohol, ethelyene
glycol, 2,3,4-trimethyl pentane, and carbon tetra-
chloride were used as extractant for organic materials
believed to be deleterious for filtration. Filtra-
tion rate did not show any improvement.
II. PHYSICAL BENEFICIATION
(a) Selective Oiling; Kerosene in combination with fatty
acid collector were mixed with the slime and allowed
to settle.
(21)
(b) Selective Flocculation; Baseman's technique were
investigated.
(c) Flotation; The froths obtained were easy to handle
and flotation was rapid. Consumption of reagents
were high.
(d) Cyclone Sizing; Phosphate distribution in various
size fraction did not show any particular pattern of
interest.
III. HYDROMETALLURGICAL BENEFICIATION
(a) Ion Exchange; Anonic resin. 5% extraction 70 hours
contact time at pH 6 to 7.
(b) Acid Leach; Sulfuric, hydrochloric and nitric acids
were tried'. Phosphorus extraction similar by the
three acids (1 lb acid per Ib of slime dry basis:
Extraction 82-96% phosphate).
221
-------
IV BACTERIOLOGICAL STUDIES
(a) Microorganism in the slime
(b) Aggregation of slime by microorganisms
(c) Assimilation of phosphorus by microorganisms
222
-------
APPENDIX C
IMPROVEMENT IN FILTRATION RATES EFFECTED BY FLOCCULANTS
223
-------
APPENDIX C
IMPROVEMENT IN FILTRATION RATES EFFECTED BY FLOCCULANTS*
Represents 12 of the best products out of a total of 150
commercial samples tested.
The Filtration Improvement Factor (F.I.F.) is the ratio of the
filtration rate for a given flocculated slime to that of
untreated slime.
Flocculant
Amer, Cyanamid Co,
Reagent S-3000
Polyacrylamide PAM-50
Polyacrylamide PAM-75
Polyacrylamide PAM-100
Dow Chemical Co.
Separan 2610
General Aniline & Film Co.
PVM/MA(Copolymer of methyl vinyl
ether and maleic anhydride)
series PVM/MA
A. F. Goodman & Sons
Potato starch
Monsanto Chemical Co.
F.I.F.
16
20
12
50
70
30
9.3
Lytron 886 (partial Ca salt of copolymer
of vinyl acetate and maleic
anhydride) 9.0
Lytron 886 with 0.05% added CaCl2 50
Lytron 887 (copolymer of vinyl
acetate and maleic anhydride) 9
Morning Star Nicol
Starch ether:Solvitose H4 8
613-45 (Cationic starch derivative) 8
Optimum Flocculant
Cone, in ppm
200
400
250
250
200
80
400
30
80
30
800
1500
LaMer, V.K., Smellie, R.H., and Pui-Kum Lee, "Flocculation,
subsidence and filtration," Journ. Coll. Sci., 12, 238 (1957)
224
-------
ACKNOWLEDGEMENTS
This report was prepared by H. T. Fullam of Battelle-
Northwest 'and B. P. Faulkner of Battelle-Columbus.
The authors are indebted to a number of individuals who
helped contribute to the preparation of this report.
The study was sponsored by the Environmental Protection
Agency. Mr. T. N. Sargent of the Southeast Water Lab-
oratory Athens, Georgia was the EPA Project Officer.
The Preliminary draft of the report was reviewed on
behalf of the Environmental Protection Agency by Mr. Sargent,
Mr George Rey, and Dr. Robert Swank.
Dr. J. W. Bartlett of Battelle-Northwest and Dr. W. M.
Goldberger of Battelle-Columbus helped review and edit
the report.
A great many individuals and companies in the fertilizer
industry were most cooperative in supplying much of the
information and data needed to complete the study.
225
-------
1
Accession Number
w
2
Subject Field & Group
05
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
f I Organization
Battelle-Memorial Institute,
Battelle-Northwest
Richland, Washington
Title
Inorganic Fertilizer and Phosphate Mining Industries - Water Pollution
and Control
1A I Profect Designation
EPA Grant No. 12020 PPD
10
Authors)
Fullam, Harold T.
Battelle-Northwes t
Faulkner, Bobby P.
Battelle-Columbus
21 Wore
22
Citation
23
Descriptors (Starred First)
^Fertilizers, *Industrial production, *Industrial waste, Waste
water disposal, Waste water treatment, Water pollution control,
Costs, Nitrogen compounds, Phosphates
25
Identifiers (Starred First)
•Fertilizer production, *Phosphate rock production, Survey
27
Abstract
A state-of-the-art survey was made of the water pollution problems
"which result from the production of inorganic fertilizers and phosphate
rock. Information required to complete the study was obtained through
an extensive literature search, questionnaires sent to the major fertil-
izer producers, and visits to selected production plants. Ninety eight
plants representing thirty three different companies were surveyed.
Production figures since 1940 and estimates of production through 1980
were accumulated for phosphate rock and the major fertilizer products.
The specific production operations which are the principal generators
of contaminated waste waters were identified, and the waste water vol-
umes and compositions for each operation were determined wherever possi-
ble. The capability of current technology to treat and control the
contaminated waste waters generated by the fertilizer industry was evalui-
ated. Problem areas where additional research and development effort
is needed to provide adequate control of waste water discharge were
identified. (Fullam - Battelle-Northwest)
Abstractor
pullam
Institution
Battelle-Northwest
WRllOS IREV JULY <»»»>
WRSIC
• END WITH COPY OF DOCUMENT, TOl WATER RESOURCES SCIENTIFIC INFORMATION CENTER
' U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
* «POI I970-»»e-»»0
-------
|