Professional Documents
Culture Documents
378.101, Florida Statutes) and empowered to conduct research supportive to the responsible
development of the state’s phosphate resources. The Institute has targeted areas of research
responsibility. These are: reclamation alternatives in mining and processing, including wetlands
reclamation, phosphogypsum storage areas and phosphatic clay containment areas; methods for more
efficient, economical and environmentally balanced phosphate recovery and processing; disposal and
utilization of phosphatic clay; and environmental effects involving the health and welfare of the people,
including those effects related to radiation and water consumption.
FIPR is located in Polk County, in the heart of the central Florida phosphate district. The Institute
seeksto serve as an information center on phosphate-related topics and welcomes information requests
made in person, or by mail, email, or telephone.
Research Staff
Executive Director
Paul R. Clifford
Research Directors
FINAL REPORT
Prepared for
March 1999
PERSPECTIVE
There is little doubt that a cost-effective crystal modifier would find widespread
application in the phosphoric acid industry and that a crystal modifier that would be
consistently effective for the more highly variable rock that will be mined in the future in
Florida would be a decided asset for the Florida phosphate industry.
iii
ABSTRACT
____________________
* Patent pending.
v
ACKNOWLEDGMENTS
vi
TABLE OF CONTENTS
Page
PERSPECTIVE .....................................................................................................iii
ABSTRACT............................................................................................................v
ACKNOWLEDGMENTS ......................................................................................vi
INTRODUCTION ..................................................................................................7
METHODOLOGY..................................................................................................9
vii
TABLE OF CONTENTS (CONTINUED)
Materials .........................................................................................................45
Procedure........................................................................................................45
Turbidity Measurements and Estimation of Induction Time (T)....................... 47
Calculation of Crystal Growth Efficiency (E) .................................................. 47
Calculation of Super-saturation (S)................................................................. 48
Effect of Surfactants on Crystal Growth (Turbidity) at
Different Super-saturation .........................................................................48
Effect of Super-saturation on Induction Time and Crystal
Growth Efficiency With and Without Surfactants.......................................51
Effect of Surfactants’ Concentration on Induction Time
and Crystal Growth Efficiency ...................................................................51
Correlation between Super-saturation and Induction Time ...............................52
Effect of Surfactants on Crystal Size Distribution of Gypsum...........................53
Effect of Surfactants on Gypsum Crystal Morphology .....................................55
COST-BENEFIT ANALYSES..............................................................................57
REFERENCES .....................................................................................................61
viii
LIST OF FIGURES
Figure Page
ix
LIST OF FIGURES (CONTINUED)
Figure Page
x
xi
LIST OF TABLES
Table Page
xii
LIST OF TABLES (CONTINUED)
Table Page
20. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%) ..... 30
xiii
LIST OF TABLES (CONTINUED)
Table Page
xiii
EXECUTIVE SUMMARY
• Bench scale testing of effect of two different surfactants (Crysmod and HiFlo-S5*)
on crystal modification and filtration of phosphogypsum produced from South
Florida high magnesium phosphate concentrate, at low-, medium- and high- sulfate
levels.
• Role of techniques and point of addition of surfactants; specific techniques involve
adding surfactants:
• As a mixture with recycle acid or water, and
• Before, during, or after nucleation
• Cost-benefit analyses
• Basic studies to elucidate the mechanisms controlling effect of surfactants under
different conditions.
Tests are carried out in which the Crysmod surfactant (1.5 kg/ton P2O5) is mixed
with recycle acid before nucleation for duration times of 30 and 5 minutes. The filtration
data in Table 1 show that, in general, mixing of the surfactant with recycle acid and its
addition before nucleation result in low filtration rates as compared to the baseline without
surfactant. It may be attributed to an inhibition of regular crystal growth as indicated by
the increase in the amount of fine crystals obtained in presence of surfactant as added
before nucleation.
Table 1. Effect of Adding Crysmod Surfactant/Recycle Acid Mixture Before
Nucleation on Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).
Surfactant Addition Technique
Item Baseline To RA† before To RA† before
(without Nucleation and Nucleation and
Surfactant) Mixed for 30 Min. Mixed for 5 Min.
Filtration rate, 4.24 1.39 3.51
Ton P2O5 /m2.day
Moisture % 28.4 33.3 30.2
Change in filtration rate, % - - 67 - 17
†RA: Recycle acid
_____________________
* Patent pending.
1
ADDITION OF SURFACTANT DURING NUCLEATION
Tests are conducted by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
either recycle acid or water and adding the mixture during nucleation. The results are
shown in Table 2 below.
The filtration data show that mixing surfactant with recycle acid during nucleation
leads to a decrease in filtration rate. On the other hand, adding surfactant / water results in
an increase in filtration rate, which corresponds to about 31% improvement over that
obtained in the absence of surfactant. Such improvement is attributed to change in
crystallization as indicated by a decrease in the amount of fine crystals produced in the
presence of surfactant/ water mixture.
A test was carried out by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
water, then adding the mixture continuously for 30 minutes. Feeding the surfactant
mixture started after addition of the phosphate concentrate and sulfuric acid. The filtration
data in Table 3 suggest that filtration rate is slightly improved (8%) due to addition of
surfactant after nucleation.
2
Table 3. Effect of Crysmod Surfactant Addition After Nucleation on
Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).
Tests have been carried out with and without surfactants addition at low sulfate
contents. The amount of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and
0.75 kg/ton P2O5 produced, respectively. The filtration and reaction data are given in
Table 4. The filtration rate is 4.24 ton P2O5 /m2.day without surfactant and improved to
5.54 and 6.88 ton P2O5 /m2.day with the addition of Crysmod and Hiflo-S5 surfactants,
respectively. The percentages of filtration rate improvement correspond to about 31 and
62%, respectively. It is also important to note that the reaction efficiency is increased with
the addition of both surfactants. This may be attributed to decrease of both lattice (co-
crystallized) P2O5 and unreacted P2O5 losses. The lattice (co-crystallized) P2O5, generally,
represent about 60% of the total P2O5 losses. The increase in reaction efficiency is
significant as it approaches 3.9% and 4.1% and the increase in P2O5 recovery is 1.6% and
3.1% with Crysmod and Hiflo-S5 surfactants, respectively.
3
EFFECT OF SURFACTANTS AT MEDIUM SULFATE CONTENT (2.2-2.4%)
To cover the range of sulfate contents applied in industry, tests at medium sulfate
content (2.2-2.4%) are carried out with and without addition of surfactants. At medium
sulfate content, the filtration data are given in Table 5. The filtration rate has increased by
25% and 43% with the addition of Crysmod and Hiflo-S5 surfactants respectively. This is
attributed to modification and growth of crystals upon surfactant addition as confirmed by
the gypsum crystals’ size distribution.
It is most interesting that, at medium sulfate content, both P2O5 recovery and
reaction efficiency are higher with the addition of surfactants. The increase in reaction
efficiency is significant as it approaches 2.8% with Crysmod surfactant and 1.9% with
Hiflo-S5 surfactant as well as the increase in P2O5 recovery is 1.4% with Crysmod
surfactant and 1.8% with Hiflo-S5 surfactant.
The excess of sulfuric acid in the reaction slurry is the most effective factor
governing crystallization quality of phosphogypsum. Its effect is not only on the crystal
shape and size, but also on co-crystallized P2O5 losses and the unreacted P2O5 losses.
Therefore, it was decided to test the effect of adding surfactants at high levels of sulfate.
The results are given in Table 6. It is interesting to note that, in this case also, the filtration
rate has improved in the presence of surfactants.
The reaction efficiencies and P2O5 recoveries in the presence and absence of
surfactants are given in the same table. It is most interesting that, even at high sulfate
content, both P2O5 recovery and reaction efficiency are higher due to the addition of
surfactants.
4
Table 6. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%).
Most importantly, the cost-benefit analyses suggest a gain of about $1.0/ton P2O5,
if Crysmod is used. However, the gain will be about $4.0/ton P2O5 if HiFlo-S5 is used.
Such gain is in addition to the increase in filtration rates due to use of these surfactants.
5
INTRODUCTION
After leaching, the slurry is filtered and counter-current washed to separate the
acid from the phosphogypsum cake. Filtration operation represents a bottleneck in the
wet-process phosphoric acid industry. Using the same filter area, the production capacity
with lower operating costs can be achieved if the filtration rate is increased. It is known
that the filtration rate depends on the characteristics of the filter cake such as crystal size,
size distribution and morphology of the crystals. In other words, large, spherical and
narrow size distribution crystals give better filtration rate.
Recycle Acid 20% P2O5
Phosphate Product
Sulfuric WI
Concentrate Acid
Acid
28% P2O5
Wash Water W II
W II 5%P2O5 W I 10%P2O5
Figure 1. Simple Dihydrate Process Flowsheet.
W: Wash liquor
7
High magnesium phosphate concentrate from South Florida was tested for
phosphoric acid production with and without surfactants at different sulfate contents. The
additives are used as crystal modifiers for phosphogypsum. In this regard, there are a
limited number of published reports about industrial crystallization of phosphogypsum
using surfactants. The available data in literature is about nucleation and crystal growth of
gypsum from pure chemicals (Ibragimova 1986; Klima 1987; Witkamp 1990; Hatakka
1997). Also, the influence of different inhibitors on the growth rate of calcium sulfate
dihydrate crystals has been studied (Ibragimova 1986; Klima 1987; Witkamp 1990;
Hatakka 1997). In addition, many research papers are published explaining the effects of
additives and impurities on crystallization of calcium sulfate (Davey 1982; Botsaris 1982;
Budz 1986; Van Rosmalen 1987; Hasson 1990; Martynowicz 1996). There are, however,
numerous reports about improving the filtration rate using polymers (Cody 1991; Liu
1973a; Liu 1973b; Liu 1975; MacCartney 1958; Sarig 1975; Smith 1970; Smith 1971;
Taylor 1989; Shoulian 1996; Moudgil 1995). At the same time, studies dealing with effect
of surfactants on filtration of phosphogypsum are limited (Rocha 1995; Schroeder 1975;
Kopyleva 1987). Even in these studies, crystallization of calcium sulfate dihydrate from
pure chemicals is reported but no data was available about surfactant addition technique.
South Florida phosphate concentrate has a relatively high impurities content (MgO
+ Fe2O3 + Al2O3 + Na2O + K2O = 4.82 %). Generally, some of these impurities lead to
decrease in the filtration rate during the processing of this particular phosphate using the
dihydrate process. Also, excessive foaming is experienced during acidulation due to high
carbonate content. To study this concentrate, dihydrate process conditions were simulated
(Abdel-Aal, 1984, 1989) and two different surfactants were tested. The size distributions
of the gypsum crystals were studied to determine the effect of the surfactants on crystal
growth at different sulfate levels.
The main aim of this study is to enhance the filtration rate of phosphogypsum
using cost-effective reagents. In addition, clarification of the mechanisms controlling the
effect of such additives is another important objective in this study. Achieving these goals
will allow the following advantages:
8
METHODOLOGY
MATERIALS CHARACTERIZATION
Pure (95.5%) sulfuric acid of 1.835 g/ml specific gravity is used for digestion. The
recycle (return) acid is accumulated while carrying out the tests. It is adjusted to the
required P2O5 content before recycling. For the first test (start-up), the required amount of
recycle acid (about 18% P2O5) is prepared by double-stage acidulation, the first with water
and sulfuric acid while the second with about 10% P2O5 phosphoric acid produced from
the previous run and sulfuric acid. The surfactants used are Crysmod (a mixture of C6-C22
Sorbitan esters) supplied by Polimeros Sinteticos, Mexico City, Mexico and Hiflo-S5
(patent pending) surfactants.
Constituent % Constituent %
P2O5 27.22 Al2O3 0.95
CaO 42.79 Na2O 0.73
MgO 1.58 K2O 0.12
Fe2O3 1.56 Insolubles 8.05
9
APPARATUS
PROCEDURE
1. Add the required amount of the return (recycle) acid R.A. to the reactor.
2. Wait until the temperature of R.A. goes up to 80 oC.
3. Add the phosphate concentrate, sulfuric acid, and the surfactant/water
suspension, continuously, for 30 minutes.
4. Continue the reaction for another 2.5 hours.
5. After the required retention time, the slurry is weighed and poured into the
filter.
6. Apply vacuum and record the time from the moment of vacuum application to
just before the surface appears dry to avoid the effects of channeling and
cracking (wall effect).
7. Wash with the required amount of wash I (~10 % P2O5) and repeat step 6.
8. Repeat step 6 with wash II (~5% P2O5), and wash water or simulated pond
water. Then allow the cake to filter for 5 seconds to dry (taking off as much as
possible of the residual liquor from the cake).
9. Weigh the wet cake and determine its moisture by drying at 60 oC for 24 hr.
10. After 0.5, 1.0, 2.0 and 3.0 hours from the start-up of the test, take 3 ml slurry
and disperse in 100 ml methanol. Then sieve using 106 Fm screen and
determine the size distribution using Coulter laser diffraction analyzer.
In industry the filtration rate is expressed in tons P2O5 produced per square meter
per day. So the same expression was used. The filtration rate was calculated applying the
following equation:
10
where:
F.R: Filtration rate, ton P2O5 /m2.day
SW: Slurry weight, g
SC: Solid content, %
F: Filtration factor
T: Total time of filtration, washing and drying, sec
After the required retention time, the produced phosphogypsum slurry was filtered
and washed under the following conditions:
11
efficiency is calculated from the following equation that obtained form IMC-Agrico
Company (Moudgil, 1995):
where:
Also, P2O5 recovery (overall or plant efficiency, or P2O5 yield) is defined as the %
of P2O5 passing from the phosphate concentrate into the produced phosphoric acid. It is
calculated as follows (Moudgil 1995):
The main objective of the washing operation is to extract, with wash liquor, as
much as possible of the phosphoric acid held by the capillary forces in the interstices of the
gypsum cake. Wash liquor displaces the impregnating phosphoric acid. The washing
(filtration) efficiency is defined as the % of water-soluble P2O5 passing from the slurry into
the produced phosphoric acid. It is calculated applying the following relation (Moudgil
1995):
Different tests are carried out to find out the best methodology for sampling
gypsum crystals for Laser diffraction size analyzer including:
• The first test is conducted using about 1.0 g dry gypsum cake. Drying is
o
performed at 60 C for 24 hours. The sample is, then dispersed in 100 ml
methanol.
• The second test is carried out using about 1.0 g air dried gypsum cake and
dispersed in 100 ml methanol.
• In the third test, a 3.0 ml slurry sample is withdrawn at different time intervals
and dispersed in 100 ml methanol.
12
. The fourth test is performed by taking 3.0 ml slurry sample at different time
intervals, and dispersed in 100 ml methanol, then screened using 106 m sieve
before conducting size analyses by Laser Diffraction Technique.
In this report, size distribution analyses have been done as outlined in the fourth test
above.
For preparation of phosphogypsum slurry the following data were used for the
calculation of feeding amounts of reactants:
Operating parameters:
13
Retention time = 180 min
P2O5 in recycle acid = 18.32 %
Free sulfate in filter acid = 2.0 %
P2O5 in filter acid = 27.5 %
Assumptions:
From the industrial point of view, it is very important to add the surfactant without
negative effects on process performance. Thus, different methodologies for surfactants’
addition were studied. For example, effect of surfactant addition to the reactor after its
mixing with recycle acid was tested. Also, mixing of surfactant with water instead of
recycle acid was studied. It should be mentioned that, the slurry solid content and amount
and P2O5 concentration of recycle acid have been kept similar to that of the baseline
(without surfactant). In these tests, the dosage of Crysmod surfactant is 1.5 kg/ton P2O5.
The following is a detailed description of these techniques:
• Mix surfactant with recycle acid for 30 minutes in the reaction vessel before
addition of reactants (phosphate concentrate and sulfuric acid). Then, add reactants
continuously for 30 minutes.
14
. Mix surfactant with recycle acid for 5 minutes before nucleation. Then add the
reactants continuously for 30 minutes.
• Mix surfactant with a volume of recycle acid equivalent to 10% of the total
volume of recycle acid. The suspension is then added continuously during the
nucleation for 30 minutes.
• Mix surfactant with water (volume of water is 10% of the total volume of recycle
acid). The suspension is then added continuously during the nucleation for 30 minutes.
• Mix surfactant with a volume of water (equivalent to 10% of the total volume of
recycle acid). The suspension is then added continuously starting 30 minutes after
nucleation for 30 minutes addition time. In these experiments, it is assumed that, the
nucleation begins with the start-up of the reaction.
15
RESULTS AND DISCUSSION
South Florida phosphate concentrate (high magnesium) was tested for phosphoric
acid production with and without Crysmod and Hiflo-S5 surfactants. Surfactant addition
techniques are studied in details (before nucleation, during nucleation and after nucleation)
using Crysmod surfactant. During industrial crystallization of phosphogypsum, the
presence of free (excess) sulfuric acid in the reaction medium is the most effective factor
governing crystal quality. Its effect is not only on the crystal shape and size, but also on
co-crystallized P2O5 losses and the unreacted P2O5 losses (Becker 1989). Free sulfate
concentration affects the super-saturation, which is the driving force of both nucleation
and crystal growth. At low sulfate content, the super-saturation is low and the rate of
both nucleation and crystal growth are low and vice versa. The range of free sulfate
industrially applied is 2.0-3.0% (Becker 1989). However, sulfate as low as 1.5% and as
high as 3.5% can be encountered. Consequently, South Florida phosphate concentrate is
tested at different sulfate contents (1.5%-3.5%).
Tests were carried out in which the Crysmod surfactant (1.5 kg/ton P2O5) is mixed
with recycle acid before nucleation for duration times of 30 and 5 minutes. The data are
given in Tables 9 and 10 and in Figures 2 and 3. The results suggest, in general, mixing of
the surfactant with recycle acid and its addition before nucleation result in low filtration
rates as compared to the baseline without surfactant. It may be attributed to an inhibition
of regular crystal growth as indicated by the increase in the amount of fine crystals
obtained in presence of surfactant as added before nucleation.
Figures 2 and 3 show that % fine crystals of produced gypsum (<10 Fm) after 3
hours are increased to about 83% and 69% for tests of high mixing time (30 min) and low
mixing time (5 min), respectively as compared to lower value of 64% for the baseline. The
reasons for formation of very fine crystals are not yet understood. It may be attributed to
retardation or inhibition of regular crystal growth by surfactant/recycle acid mixture. On
the other hand, addition of surfactant has resulted in an increase in reaction efficiencies,
P2O5 recoveries and the washing efficiencies as compared to the baseline, where surfactant
is not added (see Table 10).
17
Table 9. Effect of Adding Surfactant/Recycle Acid Mixture Before Nucleation on
Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).
18
95
80
75
70
65
60
After 1 hr. After 2 hr. After 3 hr.
19
Addition of Crysmod Surfactant During Nucleation
Tests were conducted by mixing the Crysmod surfactant (1.5 kg/ton P2O5) with
either recycle acid or water and adding the mixture during nucleation. The
surfactant/liquid suspension is added continuously for 30 minutes with the reactants
(during nucleation). The amount of either recycle acid or water is equivalent to 10% of
total volume of recycle acid. The results are given in Tables 11 and 12 and Figs. 4 and 5.
The filtration data in Table 11 show that mixing surfactant with recycle acid during
nucleation gives low filtration rate. On the other hand, adding surfactant/water mixture
during nucleation results in a higher filtration rate (5.54 ton P2O5/m2.day). This increase in
filtration rate corresponds to about 31% improvement.
Data in Figs. 4 and 5 show that mixing surfactant with recycle acid results in an
increase in the amount of fine crystals. On the other hand, less fines are produced due to
adding surfactant/water mixture.
The reasons of formation of very fine crystals upon the addition of surfactant
/recycle acid mixture during nucleation are not yet understood. However, it may be
attributed to reaction of the surfactant with the recycle acid leading to formation of some
compounds that inhibit crystal growth.
The reaction efficiencies, P2O5 recoveries and the washing efficiencies of the
phosphate concentrate with surfactant additions during nucleation are given in Table 12,
together with the data obtained without surfactant addition (baseline). It is clear that the
reaction efficiency is increased with the addition of surfactant. This may be due to
decrease of lattice (co-crystallized) P2O5 losses, which represent the majority of the total
P2O5 losses.
20
Table 11. Effect of Surfactant Addition During Nucleation on
Phosphogypsum Filtration at Low Sulfate Levels (1.5-1.8%).
21
Without Surfactant
80
70
60
50
40
30
20
After 1 hr. After 2 hr. After 3 hr.
22
Addition of Crysmod Surfactant After Nucleation
A test was carried out by mixing the surfactant (1.5 kg/ton P2O5) with water then
adding the mixture continuously for 30 minutes. Feeding the surfactant mixture started
after addition of phosphate concentrate and sulfuric acid. The amount of surfactant
corresponds to 1.5 kg/ton P2O5 and the amount of water is equivalent to 10% of total
volume of recycle acid. The results are given in Tables 13 and 14 and Figures 6 and 7.
The filtration data in Table 13 show only 8% improvement in the filtration rate.
Results in Figures 6 and 7 suggest that the amount of fine gypsum crystals (<10
Fm) is decreased to about 37% upon addition of surfactant with water after nucleation.
Note that -10 micron fraction in the baseline without surfactant is 64%.
The reaction efficiencies, P2O5 recoveries and the washing efficiencies obtained
with surfactant addition after nucleation are given in Table 14 together with those
achieved without surfactant (baseline). In this case also, the reaction efficiency has
increased with the addition of surfactant. This may be attributed to decrease of lattice (co-
precipitated) P2O5 losses as mentioned above.
23
Table 14. Effect of Surfactant Addition After Nucleation on Reaction Efficiency,
P2O5 Recovery and Washing Efficiency.
100
Volume % of Phosphogypsum Crystals,
Without Surfactant
90
70
60
50
40
30
20
After 1 hr. After 2 hr. After 3 hr.
24
Figure 7. Effect of Surfactant Addition, After Nucleation, on
Phosphogypsum Size Distribution.
Tests have been carried out, with and without surfactants addition at low sulfate
contents. The amount of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and
0.75 kg/ton P2O5 produced, respectively. The results are given in Tables 15-17 and Figure
8.
The filtration and reaction data are given in Table 15. The filtration rate is 4.24 ton
P2O5 /m2.day without surfactant and improved to 5.54 and 6.88 ton P2O5/m2.day with the
addition of Crysmod and Hiflo-S5 surfactants, respectively. The percentages of filtration
rate improvement correspond to about 31 and 62%, respectively. This is attributed to
modification and growth of produced crystals upon surfactant addition. The size
distributions of gypsum crystals (given in Tables 16 and 17 and Fig. 8) show that the
volume percentages of fine crystals (<10Fm) is decreased to about 31% and 9% upon
addition of Crysmod and Hiflo-S5 surfactants, respectively, while it is as high as 64%
without surfactant addition. Also, the mean of crystal size increases with addition of
surfactants.
The data for reaction efficiencies, P2O5 recoveries and washing efficiencies with
and without additives are given in Table 15. It is important to note that the reaction
25
efficiency is increased with the addition of both surfactants. As mentioned before, this may
be attributed to decrease of both lattice (co-crystallized) P2O5 and unreacted P2O5 losses.
The lattice (co-crystallized) P2O5 losses, generally, represent about 60% of the total P2O5
losses. In this regard, it has been reported that, presence of a surfactant has resulted in
reduction of P2O5 in gypsum and this is attributed to the surfactant effect in reducing the
capture of phosphate ions from solution by gypsum crystals. Regarding the decrease of
unreacted phosphate, it was found that addition of surfactant increases the phosphate
concentrate solubility and consequently increases reaction efficiency. In another study, it
has been found that the presence of surfactant increases the solubility of calcium
phosphate in thermal and wet-process phosphoric acid of 10-40% P2O5 concentration and
at temperature range of 40-75 oC.
Table 15. Filtration and Reaction Data at Low Sulfate Content (1.5-1.8%).
Table 16. Comparative Size Analyses of Gypsum Crystals (At Low Sulfate Content
and After 3hr).
26
Table 17. Statistics of Phosphogypsum Crystal Size Distribution
(At Low Sulfate and After 3 Hr).
27
Testing at Medium Sulfate Content (2.2-2.4%)
To cover the range of sulfate contents applied in industry, tests at medium sulfate
content (2.2-2.4%) were carried out with and without addition of surfactants. At medium
sulfate content, the obtained filtration data are given in Table 18. The filtration rate is
increased by 25% and 43% due to the addition of Crysmod and Hiflo-S5 surfactants,
respectively. This is attributed to modification and growth of produced crystals upon
surfactant addition as confirmed by the size distribution of gypsum crystals given in Table
19 and Figure 9. It is clear that the volume percentage of coarse phosphogypsum crystals
(>75Fm) is increased with surfactants addition. On the other hand, the volume percentage
of fine phosphogypsum crystals (<38Fm) is decreased with surfactants’ addition.
Table 18. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%).
28
Table 19. Comparative Size Analyses of Gypsum Crystals (at Medium Sulfate
Content and After 3 Hr).
29
Washing with Simulated Pond Water. Two tests were carried out to study
effect of using simulated pond water on the filtration rate without surfactant and with
Hiflo-S5 surfactant. The simulated pond water has specific gravity of 1.02 g/ml and
contains 1.5% P2O5 and saturated with phosphogypsum. The results are reported in Table
20 and compared with the results using wash water. Filtration rate was slightly decreased
upon washing with simulated pond water with and without surfactant. It is interesting to
notice a slight decrease in filtration rate whenever pond water is used for washing instead
of tap water.
Table 20. Filtration and Reaction Data at Medium Sulfate Content (2.2-2.4%).
Table 21. Comparative Size Analyses of Gypsum Crystals (Without Surfactant and
at Medium Sulfate Content).
30
Table 22. Comparative Size Analyses of Gypsum Crystals (With Crysmod and at
Medium Sulfate Content).
Table 23. Comparative Size Analyses of Gypsum Crystals (With Hiflo-S5 and at
Medium Sulfate Content).
31
35
Without Surfactant
WithHiflo-S5 Surfactant
25
20
15
10
0
After 0.5 hr After 1 hr After 2 hr After 3 hr
32
100
Without Surfactant
70
60
50
40
30
20
10
0
After 0.5 hr After 1 hr After 2 hr After 3 hr
33
Figure 13. Comparative Size Distribution of Phosphogypsum Crystals
(With Crysmod Surfactant at Medium Sulfate Content).
34
Testing at High Sulfate Content (3.2-3.5%)
The excess of sulfuric acid in the reaction slurry is the most effective factor
governing crystallization quality of phosphogypsum. Its effect is not only on the crystal
shape and size, but also on co-crystallized P2O5 losses and the unreacted P2O5 losses.
Sulfate concentration affects super-saturation of the medium. At high sulfate content, the
super-saturation is high and the rate of both nucleation and crystal growth are high. The
amounts of applied Crysmod and Hiflo-S5 surfactants correspond to 1.5 and 0.75 kg/ton
P2O5 produced, respectively.
The results obtained are given in Tables 24-26 and Figure 15. It is interesting to
note that the filtration rate (6.77 ton P2O5 /m2.day without additive) has improved to 8.33
and 8.98 ton P2O5/m2day with the addition of Crysmod and Hiflo-S5 surfactants,
respectively. The increase in filtration rate is related to the growth of crystals upon
surfactants addition and formation of high volume percent of coarse crystals (>75 µm).
The size distribution of phosphogypsum crystals given in Tables 25 and 26 and Figure 15
shows that, the volume percentages of coarse crystals (>75 µm) are increased to about
15% and 24% upon addition of Crysmod and Hiflo-S5 surfactants, respectively while it is
as low as 10% without surfactant addition. Also, the mean of crystal size increases with
addition of surfactants.
Table 24. Filtration and Reaction Data at High Sulfate Content (3.2-3.5%).
35
Table 25. Comparative Size Distribution of Gypsum Crystals (At High Sulfate
Content and After 3 Hr).
Table 26. Statistics of Phosphogypsum Crystal Size Distribution (At High Sulfate
and After 3 Hr).
36
Figure 15. Comparative Size Distribution of Phosphogypsum Crystals
(at High Sulfate Content and After 3 Hr).
The relationship between amount of phosphogypsum fines and the filtration rates
are given in Tables 27 and 28. These results show that, at each sulfate level, higher
filtration rates are obtained as the fine crystals (-10 Fm) decrease, and the mean and the
median of crystal size distribution increase.
37
Table 27. Relationship Between % Phosphogypsum Fines and Filtration Rate At
Low Sulfate Content.
38
Table 28. Relationship Between % Phosphogypsum Fines and Filtration Rate at
Different Sulfate Contents.
Gypsum Morphology
One of the most important factors affecting filtration rate is gypsum morphology
(size and shape of the crystals). For best filtration and washing, crystals of uniform sizes
are most desirable. At low sulfate content, shape of the crystals is changed from tabular
crystals to clusters by addition of the surfactants. On the other hand, at high sulfate
content large clusters are formed upon using the surfactants as shown in SEM
photomicrographs, Plates 1-8. These clusters are composed of many small crystals while
small clusters are formed without addition of surfactants. This may be due to adsorption of
surfactants on crystal surfaces rendering it hydrophobic. Thus, clusters may form due to
hydrophobic interactions between crystals. Similar findings have been reported in literature
(Sirianni and others 1969).
39
Plate 1. SEM Photomicrograph of Phosphogypsum without Surfactant
(Magnification 500).
40
Plate 3. SEM Photomicrograph of Phosphogypsum with Crysmod Surfactant
(Magnification 200).
41
Plate 5. SEM Photomicrograph of Phosphogypsum with Crysmod Surfactant
(Magnification 2000).
42
Plate 7. SEM Photomicrograph of Phosphogypsum with Hiflo-S5
Surfactant (Magnification 500).
43
BASIC STUDIES: PRIMARY NUCLEATION OF
CALCIUM SULFATE DIHYDRATE
MATERIALS
Pure chemicals including phosphoric and sulfuric acids, and calcium hydrogen
phosphate monobasic (from Fisher Scientific) are used to prepare the following solutions:
PROCEDURE
1. Add 500 ml of solution # 1 in 800 ml beaker and heat to 80° C using a constant
temperature water bath.
2. Then add simultaneously the following solutions (see Tables 29-30): Calcium
Hydrogen Phosphate Monobasic (which is prepared by dissolving the required
amount in 100 ml of solution #2 and the required amount of sulfuric acid, solution
#3, and 50 ml of deionized water or water/surfactant solution.
3. Keep reaction at 80° C with constant agitation.
4. Measure turbidity of the resulting suspension at different time intervals during the
course of the reaction using a HACH 2100A Turbidimeter.
5. Take samples (5 ml) from the pregnant liquor for particle size distribution analysis
after 1 minute, 5 minutes, 15 minutes and at the end of the experiment (when the
turbidity value exceeds 1000 NTU).
45
6. For size distribution measurements, disperse the samples in about 50 ml of
methanol and measure the size distribution using a Coulter Laser Size Analyzer.
Table 29. Amounts of Calcium Hydrogen Phosphate Monobasic and Sulfuric Acid
for Different Super-saturation.
46
TURBIDITY MEASUREMENTS AND ESTIMATION OF INDUCTION TIME (T)
600
500
400
300
200
100
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
Time (min)
Figure (3). From the Figure Time = 130 min
Figure 16. Example of Turbidity Data Plotting and Induction Time Estimation.
Induction times are calculated as described above for the data collected in presence
and absence of surfactants. From the estimated values, crystal growth efficiency (E) is
calculated by the following equation as reported by Georgiev and others (1995):
E = (T0 – T1)
T0
where:
T0 = Induction time in absence of surfactant
T1 = Induction time in presence of surfactant
47
CALCULATION OF SUPER-SATURATION (S)
M.W. CaO
Mass CaO = Mass CaH4(PO4)2•H2O ×
M.W. CaH4(PO4)2•H2O
Mass CaO
% CaO = × 100
Total Mass
% CaO x M.W. CaSO4•2H2O
c = Calcium sulfate dihydrate concentration =
M.W. CaO
c* = Calcium sulfate dihydrate (solute) solubility at the given temperature = 0.83 (Becker,
1989)
c
S = Super-saturation =
c*
48
1200
Without Surfactant
1000
Crysmod Surfactant
800
Turbidity (NTU)
Hiflo-S5
600
400
200
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
Time (min)
1200
Without Surfactant
1000
Crysmod Surfactant
600
400
200
0
0 10 20 30 40 50 60 70 80 90 100
Time (min)
49
1200
Without Surfactant
1000
Crysmod Surfactant
Turbidity (NTU)
800
400
200
0
0 5 10 15 20 25 30
Time (min)
1200
Without Surfactant
1000
Crysmod Surfactant
Hiflo-S5 Surfactant
800
Turbidity (NTU)
600
400
200
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
Time (min)
50
EFFECT OF SUPER-SATURATION ON INDUCTION TIME AND CRYSTAL
GROWTH EFFICIENCY WITH AND WITHOUT SURFACTANTS
Induction time and crystal growth efficiency are determined at different super-
saturation with and without surfactants and given in Table 32. The surfactant
concentration used is 100 ppm. These results confirm that Crysmod consistently lowers
the induction time to a greater degree than Hiflo-S5 when 100 ppm of surfactant is used at
different super-saturation. The induction time (determined from Figures 17-20) of the
solution without surfactant is always higher than that with Crysmod or Hiflo-S5
surfactant. In all three cases, as the super-saturation is increased, the induction time is
decreased.
Table 32. Effect of Super-saturation on Induction Time (T) and Growth Efficiency
(E) of Gypsum Crystals (Using 100 ppm Surfactant).
Super-saturation
Surfactant 1.018 1.222 1.502 1.979
T* E T E T E T E
None 140 - 80 - 24 - 2.9 -
Crysmod 130 0.071 76 0.050 17 0.292 1.70 0.414
Hiflo-S5 137 0.021 77 0.038 17 0.292 2.75 0.052
* in minutes
51
Table 33. Effect of Surfactant Concentration on Induction Time (T) and Growth
Efficiency (E) of Gypsum Crystals (Using 1.502 Super-saturation).
The induction time without and with 100 ppm of surfactant is plotted as a function
of super-saturation, Figure 21. When the natural log of the induction time is plotted
against super-saturation in Figure 22, a linear relationship is observed. This confirmed that
the plot of Figure 8 is indeed exponential. Note that the induction time is inversely
proportional to the super-saturation. Also, Figure 22 shows that, at high supersaturation
the efficiency of the surfactants is higher. It may be inferred from these data that in the
presence of surfactants, retention times may be reduced leading to high throughputs.
However, this needs to be confirmed on plant scale testing.
140
W ithout Surfactant
120 Crysmod Surfactant
Hiflo-S5 Surfactant
Induction Time (min)
100
W ithout Surfactant
Crysmod Surfactant
80
Hiflo-S5 Surfactant
60
40
20
0
1 1.2 1.4 1.6 1.8 2
Supersaturation
52
6
Without Surfactant
Crysmod Surfactant
5
Hiflo-S5
Linear (Hiflo-S5)
ln tind
0
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
Super-saturation
53
Figure 23. Differential Volume of Gypsum Crystals Obtained after 5 Minutes at a
Super-saturation of 1.502 (Surfactant Concentration = 100 ppm).
54
EFFECT OF SURFACTANTS ON GYPSUM CRYSTAL MORPHOLOGY
Gypsum crystal morphology was investigated without and with 100 ppm
surfactant. The photomicrographs are taken using an Olympus BX60F5 microscope. The
calcium sulfate dihydrate crystals (gypsum) in these photomicrographs are precipitated at
a super-saturation of 1.502. These photomicrographs confirm that Crysmod and Hiflo-S5
surfactants have significantly altered the morphology of the gypsum crystals. The crystals
shown in Plate 9, formed in the absence of surfactant, are needle-like shape with high
aspect ratio. Needle-like crystals give low filtration rate and maintain high amounts of
liquid. On the other hand, Plates 10 and 11 show crystals that are much larger in size and
tabular in shape with low aspect ratio. These crystals are expected to allow high filtration
rates. It is important to mention that the results of the fundamental studies confirm the
data obtained in the digestion work using phosphate concentrate.
55
Plate 10. Photomicrograph of Gypsum Crystals Formed with Crysmod
Surfactant (100 ppm).
56
COST-BENEFIT ANALYSES
1. In Case of Crysmod:
Cost* = $4.50/ton P2O5
Benefits:
- Gain in recovery
(1.4% @ $395.00/ton P2O5) = $5.53/ ton P2O5
- Increase in filtration rate = 25-30%
2. In Case of HiFlo-5S:
Cost = $1.5/ton P2O5
Benefits:
- Gain in recovery
(1.4% @ $395.00/ton P2O5) = $5.53/ ton P2O5
- Increase in filtration rate = 30-40%
_________________
*
Cost includes cost of surfactant at the delivery price as quoted by manufacturer. Feeding
and handling costs are neglected since only a feeding tank and dosing pump are needed for
this application.
57
CONCLUSIONS AND RECOMMENDATIONS
1. Bench scale testing of effect of two different surfactants (Crysmod and HiFlo-S5)
on crystal modification and filtration of phosphogypsum produced from South
Florida phosphate concentrate containing high magnesium, at different sulfate
levels (low, medium and high);
3. Cost-benefit analyses
The data obtained in the above studies lead to the following conclusions:
• Most importantly, surfactants are found to increase reaction efficiency and P2O5
recovery. The increase in reaction efficiency ranges between 1.2% to 4.1% and the
increase in P2O5 recovery ranges from 0.5% to 3.1% depending on surfactant type
and sulfate level.
59
• The fundamental study with pure chemicals indicates that:
• Most importantly, the cost-benefit analyses suggest a gain of $1.03 / ton P2O5, if
Crysmod is used. However, the gain is up to $4.03/ ton P2O5 if HiFlo-S5 is used.
Such gain is in addition to the increase in filtration rates due to use of these
surfactants.
60
REFERENCES
Abdel-Aal EA. 1984. Comparative study on phosphoric acid production from Egyptian
phosphate ore concentrates by the wet process. M. Sc. thesis, Dept. of Chemistry,
Faculty of Science, Cairo University.
Abdel-Aal EA. 1989. Industrial simulation for wet process phosphoric acid production.
Ph.D. dissertation, Dept. of Chemistry, Faculty of Science, Cairo University.
Becker P. 1989. Phosphates and phosphoric acid: raw materials, technology and
economics of the wet process. New York: Marcel Decker Inc.
Budz J, Jones AG, Mullin JW. 1986. Effects of selected impurities on the continuous
precipitation of calcium sulphate (gypsum). J. Chem. Tech. Biotechnology 36:
153-161.
Cody AM, Cody RD. 1991. Crystal habit modifications of gypsum from epitaxial-like
adsorption of stereospecific growth inhibitors. Journal of Crystal Growth 113:
508-519.
Dyatlova NM, Evseev AM, Nikolaeva LS, Samakeav RKh. 1984. Simulating the
inhibition of calcium sulfate crystallization in the presence of EDTP. Zhurnal
Prikladnoi Khimii 57(10):2237-2240.
Georgiev GS, Djarova MD, Petkova JI, Georgiev GI, Koseva NS. 1995. Effect of poly
(acrylic acid-co-N-vinylpyrrolidone) on calcium oxalate crystallization. Polymer
Journal 27(4): 441-444.
Hatakka H, Oinas P, Reunanen J, Palosaar S. 1997. Induction time and reaction kinetics
in batch precipitation of calcium sulfate. Acta Polytechnica Scandinavica,
Chemical Technology Series (244): 102-104.
He S, Oddo J, Tomson M. 1994. The inhibition of gypsum and barite nucleation in NaCl
brines at temperatures from 25 to 900. Applied Geochemistry 9:561-567.
61
Ibragimova AN, Ibragimov RN, Mezhidov VKh. 1986. Determination of the nucleation
parameters of calcium sulfate under frictional conditions. Zhurnal Prikladnoi
Khimii 59(5): 965-968.
Klima WF, Nancollas GH. 1987. The growth of gypsum. AIChE Symposium Series
83(253): 23-30.
Liu ST, Nancollas, GH. 1973a. Linear crystallization and induction-period studies of the
growth of calcium sulfate dihydrate crystals. Talanta 20: 211-216.
Liu ST, Nancollas GH. 1973b. The crystal growth of calcium sulfate dihydrate in the
presence of additives. Journal of Colloid and Interface Science 44(3): 422-429.
Liu ST, and Nancollas GH. 1975. A kinetic and morphological study of the seeded
growth of calcium sulfate dihydrate in the presence of additives. Journal of
Colloid and Interface Science 52(3): 593-601.
MacCartney ER, Alexander AE. 1958. The effect of additives upon the process of
crystallization: I. Crystallization of calcium sulfate. Journal of Colloid Science 13:
383-396.
Martynowicz ETMJ, Van Rosmalen GM, Witkamp GJ. 1996. The effect of aluminum
fluoride on the formation of calcium sulfate hydrates. Hydrometallurgy 41:171-
186.
Rocha SDF, Ciminalli VST. 1995. Effect of surfactants on calcium sulfate crystallization
in phosphoric acid solution. Minerals and Metallurgical Processing 12(3): 166-
171.
Sarig S, Kahana F, and Leshem R. 1975. Selection of threshold agents for calcium
sulfate scale control on the basis of chemical structure. Desalination 17: 215-229.
62
Shoulian Z. 1996. Modification of crystal size distribution for enhanced filtration of
phosphogypsum using poly(ethylene oxide). Ph.D. dissertation, Dept. of
Materials Science and Engineering, University of Florida, Gainesville, FL.
Smith BR, Alexander AE. 1970. The effect of additives on the process of crystallization
II. Further studies on calcium sulfate. Journal of Colloid and Interface Science
34(1): 81-90.
Smith BR, Sweett F. 1971. The crystallization of calcium sulfate dihydrate. Journal of
Colloid and Interface Science 37(3): 612-618.
Tavare, NS. 1995. Industrial crystallization: process simulation analysis and design.
The Plenum Chemical Engineering Series. New York: Plenum Press,
p 14.
Taylor JF. 1989. The effect of polymeric additives on the filtration characteristics of
gypsum. M. Sc. thesis, Dept. of Materials Science and Engineering, University of
Florida, Gainesville, FL.
Van Rosmalen GM, Daudey PJ, Marchee WGJ. 1982. Quantitative description of the
influence of the inhibitor concentration on the growth rate of calcium sulfate.
Industrial Crystallization 81: 147-153.
Van Rosmalen GM, Weijnen MPC. 1984. The role of additives and impurities in the
crystallization of gypsum. Industrial Crystallization 84: 61-66.
Van Rosmalen GM, Witkamp GJ, de Vreugd CH. 1987. Additive and impurity effects in
the crystallization process. Industrial Crystallization 87: 21-25.
Witkamp GJ, Van der Eerden JP, Van Rosmalen GM. 1990. Growth of gypsum.
Journal of Crystal Growth 102: 281-289.
63