EFFECT OF SODIUM CHLORIDE ON THE SELECTIVE
FLOTATION OF DOLOMITE FROM APATITE
DURSUN E. lNCE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA 1987
This work is dedicated to the
memory of my father, the late HIDIR INCE,
without whose inspiration, encouragement and special admiration for education this
work would have never been attempted.
I wish to express my sincere gratitude and respect to Professor B. M. Moudgil, my major advisor, for his invaluable help, guidance and encouragement during the course of this research.
I am very grateful to Professor F. N. Blanchard for his guidance, valuable comments, and to Professors 0. 0. Shah, E. D. Whitney, D. E. Clark, H. A. Laitinen and J. H. Simmons for very helpful discussions and comments.
I am most grateful to Professor C. T. Johnston and Mrs. L. D. Applewhite for discussions and help in FT-IR study.
With due gratitude I wish to acknowledge the encouragement,
helpful comments and constructive criticism of Drs. T. V. Vasudevan, H. Soto and Wen-Keng Shih.
I am thankful to M. May, A. Zutshi, J. Ransdell, Y. C. Cheng and J. Rogers for their help at various stages, and to Mrs. G. Keim for her help in preparing the manuscript.
Special love and appreciation is due to my wife Sevgi for her
support, help and encouragement and to my two lovely daughters, Elif and Ebru, for their patience throughout the course of this study.
Finally, I wish to thank Agrico Chemical Company and International Minerals and Chemicals Corporation for supplying the mineral samples used in this study and to acknowledge Florida Institute of Phosphate Research (Grants #82-02-023 and #85-02-067) for providing financial iii
support. Any opinions, findings, and conclusions or recommendations expressed in this work are those of the author and do not necessarily reflect the views of the Florida Institute of Phosphate Research.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS . . . . . . . . . . . . iii
LIST OF TABLES . . . . . . . . . . . . . ix
LIST OF FIGURES . . . . . . . . . . . . xi
ABSTRACT . . . . . . . . . . . . . . xiv
I INTRODUCTION . . . . . . . . . . . . 1
II BACKGROUND . . . . . . . . . . . . 6
Characteristics of Salt-Type Minerals . . . . . 6
Characteristics of Apatites in Florida Phosphorites . 7
Characteristics of Dolomite . . . . . . . 7
Solubility of Apatite and Dolomite . . . . . . 8
Surface Charge of Apatite and Dolomite . . . . . 9
Aging Behavior of Apatite and Dolomite . . . . . 10
Separation Studies . . . . ... . . . . . 11
Methods Based on Physical Properties . . . . . 11
Separation Based on Physico-Chemical Properties . . 12
Selective flocculation . . . . . . . 12
Selective flotation . . . . . . . 12
Flotation in the Presence of Salts . . . . . . 16
III EXPERIMENTAL . . . . . . . . . . . . 18
Materials . . . . . . . . . . . . 18
Minerals . . . . . . . . . . . 18
Apatite . . . . . . . . . . 18
Dolomite . . . . . . . . . . 18
Reagents . . . . . . . . . . . . 19
Other Chemicals . . . . . . . . . . 19
Methods . . . . . . . . . . . . . 20
Flotation . . . . . . . . . . . 20
Electrokinetic Measurements. .. ...... .......23
Solution Preparation .. ....... ...........24
Oleate Adsorption Tests. .. ....... ........24
Mineral Dissolution Tests. .. ...... ........27
Solubility Product Determination. .... ........29
FT-IR Tests. .. ...... ......... ......29
Contact Angle Measurements .. ....... .......31
Experimental Plan. .... ........ ...........31
Selection, Preparation and Characterization of
the Minerals. .. ... ........ .........32
Selection of the Surfactant. .. ...... .......32
Selection of the Experimental Techniques .. .......32
Flotation. .... ........ .........32
Electrokinetic measurements. ... .........33
Adsorption .. ... ......... ........33
Determination of the nature of the adsorbed
surfactant species. .. ...... ........34
Mineral dissolution studies. .... ........34
Contact angle. .... ........ .......34
Experimental Approach. .. ....... .........35
IV RESULTS. .... ........ ........ .......36
Characterization of Minerals. .. ...... .........36
Chemical Analysis .. ... ......... .......36
Surface Area and Porosity. .. ...... ........36
X-ray Analysis. .... ........ .........40
Surface Chemical Characterization .. ... ........40
Flotation Studies with Dodecylamine Hydrochloride ... 40
Single mineral flotation behavior. ..............40
Flotation of apatite-dolomite mixture with
dodecylamine .. ... ......... ......50
Flotation Studies Using Sodium Oleate as the
Collector. .. ...... ......... .......50
Single mineral flotation tests .. ... .......52
Mixed minerals .. ... ........ .......52
Flotation tests in the presence of NaCl using
dodecylamine as the collector. .. ... ......56
Flotation studies in the presence of NaCl using
sodium oleate as the collector .. .. .......59
Flotation studies in the presence of KCl and NaF 67
Electrokinetic Studies .. ... ........ ........67
Effect of Salt Addition on the Zeta Potential
Effect of Salts on the Zeta Potential of Dolomite . 70
Role of NaCl in the Reversal of Surface Charge
of Apatite......... ..............76
Substitution of sodium for calcium in the
apatite lattice .. ....... .........76
Adsorption Studies .. ... ........ ...........82
Single Minerals Adsorption Tests ..... ........... 82
Apatite-oleate system .... .............. ... 82
Dolomite-oleate system .... .............. ...85
Mixed Mineral Adsorption Studies ..... ........... 90
Characterization of the Adsorbed Oleate Species
by FT-IR Spectroscopy ........................... ..90
IR Spectra of Pure Oleate Species ............. ....93
Nature of the Adsorbed Species at pH 10 .... ........ 96
Apatite-oleate system .... .............. ... 96
Dolomite-oleate system .... .............. ...96
Nature of the Adsorbed Species at pH 4 .. ........ .. 99
Apatite-oleate system .... .............. ... 99
Dolomite-oleate system .... .............. ...99
Oleate Species Adsorbed in the Presence of NaCl .... 102
Apatite-NaCl-oleate system ..... ............ 102
Dolomite-NaCl-oleate system ..... ........... 102
V DISCUSSION ..... .... ........................ ...105
Solution Properties of Dodecylamine and Oleate ......... ...105
Apatite-Dolomite Flotation Using Dodecylamine as
the Collector ......... ...................... 109
Apatite-Dodecylamine System .... .............. ..109
Dolomite-Dodecylamine System .... ............. ..110
Flotation of Apatite and Dolomite Mixture with
Dodecylamine ................................ 111
Changes in the surface charge and surface
chemical composition .... .............. ..112
Surface coating ...... ................. ...114
Surfactant depletion by precipitation ... ...... 114
Flotation of Apatite and Dolomite Using Sodium Oleate
as the Collector ..... ..................... 116
Evaluation of the Results and Alternatives ... ...... 116
Effect of NaCl on the Selective Flotation of
Apatite Using Dodecylamine as the Collector ....... 119
Effect of NaCl on the Separation of Dolomite From
Apatite Using Sodium Oleate as the Collector ..... ...120
Mechanism of Selective Flotation of Dolomite from Apatite 121
Effect of NaCl on the Zeta Potential of Apatite
and Dolomite ...... ..................... ...121
Role of NaCl in the reversal of surface charge
of apatite ...... ................... ...122
Dolomite structure ....... ................ 126
Adsorption of Oleate on Apatite and Dolomite .... ........ 128
Effect of Conditioning pH .... ............... ...128
Effect of NaCl Addition on Adsorption ........... ...131
Adsorption in the Mixed Mineral System .. ........ .. 133
Effect of Salt on Adsorption in the Mixed Minerals
System ........ ....................... ...133
Nature of the Adsorbing Surfactant Species ........... ...134
Apatite-Oleate System ..... ................. ...134
Dolomite-Oleate System. .... ........ ....136
Preferential formation of magnesium oleate ....138
Contact Angle Studies .. ....... ........ ...145
Mechanism of Selective Flotation of Dolomite from
Apatite in the Presence of NaCl .. ........ ....147
VI CONCLUSIONS. .... ........ ........ ....149
VII SUGGESTIONS FOR FUTURE RESEARCH. .... ........ ..155
REFERENCES. .. ...... ........ ......... ...159
BIOGRAPHICAL SKETCH. .... ........ ........ ...168
LIST OF TABLES
Table No. Page
1 Sodium Oleate Distribution in Various Streams
During Adsorption Tests at pH 4.5 ............. ..25
2 Characteristics of Apatite and Dolomite Samples . 37
3 Flotation Results of 50:50 Apatite/Dolomite
Mixture Using Dodecylamine as the Collector ....... 51
4 Flotation Results of 50:50 Apatite-Dolomite
Mixture Using Sodium Oleate as the Collector ....... 54
5 Effect of NaCl on the Single Mineral Flotation
of Apatite and Dolomite with Dodecylamine
Hydrochloride as the Collector ............... ...57
6 Results of Mixed Mineral Flotation in the
Presence of NaCl Using Dodecylamine as the
Collector at pH 6.3 ..... ................. ...58
7 Results of Mixed Mineral Flotation Tests in the
Presence of NaCl Using Sodium Oleate as the
Collector ......... ...................... 60
8 Apatite Dissociation at pH 4 with and without
Added Sodium Chloride ....... ................ 79
9 Effect of NaCl on Unit Cell Dimensions of Apatite
Conditioned at pH 4 ..... ................. ...80
10 Determination of Substitution of Sodium for
Calcium in the Apatite Structure .............. ..81
11 Effect of Sodium Chloride Addition on Oleate
Adsorption on Apatite and Dolomite at pH 4.0 ....... 92
12 Dissolution of Calcium and Magnesium from
Dolomite at pH 4.0 with and without NaCl
Addition ....... ....................... ..113
13 Adsorption and Flotation Results with Zeta
Potential Values for Apatite and Dolomite
with and without NaCl at PH 4.0 .. ... .......132
14 Solubility Product of Calcium and Magnesium
Oleate .. .. ...... ........ .......140
15 Electrostatic Interaction Energy of Calcium and
Magnesium with Oleate. .. ...... ........144
16 Contact Angle and Flotation Recovery of Apatite
and Dolomite at PH 4.0 .. .. ...... .......146
LIST OF FIGURES
Figure No. Page
1 Location of Florida phosphate deposits .. .. ......2
2 A schematic diagram of a modified Hallimond Cell . 21
3 Hallimond cell flotation arrangement .. .. ... ..22
4 Apparatus for mixed mineral adsorption studies . . 28
5 SEM micrograph of apatite (65x100 mesh size
fraction), a) 200X; b) 1000X .. .. .. .. . ....38
6 SEM micrograph of dolomite (65x100 mesh size
fraction), a) lOOX; b) 100OX .. .. ...... ....39
7 X-ray diffractogram of apatite .. .. ...... ...41
8 X-ray diffractogram of dolomite. .. ...... ...42
9 Zeta potential of apatite as a function of pH . ..43
10 Zeta potential of dolomite as a function of pH .. 44
11 Flotation of apatite as a function of pH. .. ... ..46
12 Dolomite flotation as a function of pH. .. ... ...47
13 Flotation of apatite and dolomite (single minerals)
as a function of pH .. ... ......... .....48
14 Flotation of apatite and dolomite (single minerals)
as a function of pH .. ... ......... .....49
15 Flotation of apatite and dolomite (single minerals)
as a function of pH .. ... ......... .....53
16 Effect of sodium chloride addition on apatite and
dolomite flotation .. .. ...... ..........61
17 Effect of sodium oleate concentration on apatite
and dolomite (single minerals) flotation with and
without NaCl addition .. ... ........ ....62
18 Flotation of apatite and dolomite as a function of
sodium oleate concentration in the absence and
presence of sodium chloride (single minerals) . . 63
19 Effect of pH on flotation of apatite and dolomite
(single minerals) in the presence of sodium
chloride. .. ... ........ ........ ..65
20 Flotation recovery of apatite and dolomite (mixed
minerals) as a function of pH in the presence of
sodium chloride. .. ...... ......... ..66
21 Effect of KCl addition on flotation of apatite and
dolomite as a function of pH .. .. ...... ....68
22 Apatite and dolomite (single minerals) flotation
with and without NaF addition at pH 4 .. ... ....69
23 Zeta potential of apatite with and without NaCl
addition. .. .. ......... ........ ..71
24 Zeta potential of apatite as a function of pH
with and without KCl addition. .. ....... ...72
25 Zeta potential of apatite as a function of pH with
and without NaF addition. .. ... ........ ..73
26 Zeta potential of dolomite as a function of pH with
and without sodium chloride. .. ....... .....74
27 Effect of KCl on the zeta potential of dolomite . 75
28 Zeta potential of dolomite in the presence and
absence of NaF as a function of pH .. .. .... ...77
29 Oleate adsorption on apatite as a function of
conditioning pH. .. ....... ........ ..83
30 Oleate adsorption on apatite as a function of
initial oleate concentration at pH 4.0 and 10.0 . 84
31 Oleate adsorption on apatite at pH 4.0 in the
absence and presence of sodium chloride .. ... ...86
32 Adsorption of oleate on dolomite as a function of
conditioning pH. .. ...... ......... ..87
33 Oleate adsorption on dolomite as a function of
initial oleate concentration at pH 4.0 and 10.0 . 88 34 Adsorption of oleate on dolomite at pH 4.0, with
and without added sodium chloride. .. ...... ..89
35 Oleate adsorption on apatite and dolomite (single
and mixed minerals) as a function of pH .. ... ...91
36 Diffuse reflectance IR spectra of Mg-, Ca- and
Na-oleate (400-4000 cm-1 range). .. ....... ..94
37 Diffuse reflectance IR spectra of Mg-, Ca- and
Na-oleate (1200-1800 wavenumber region) .. .... ..95
38 Diffuse reflectance IR spectra of treated and
untreated apatite at pH 10.0 .. .. ...... .....97
39 IR spectra of untreated and treated dolomite at
pH 10.0 .. ... ........ ......... ..98
40 Diffuse reflectance IR spectra of treated and
untreated apatite at pH 4.0 .. ... ..........100
41 IR spectra of treated and untreated dolomite
at pH 4.0 .. ... ........ ........ ..101
42 Diffuse reflectance IR spectra of apatite
(treated and untreated) and the difference
spectrum at pH 4.0 in the presence of NaCl. .. .....103
43 Diffuse reflectance IR spectra of dolomite at
pH 4.0 in the presence of sodium chloride. .. ....104
44 Dodecylamine species distribution as a function
of pH. Total amine concentration, 1.6 x 10-2 M . 107
45 Oleate species distribution as a function of pH
Total oleate concentration, 4.0 x 10-5 M .. .......108
46 Crystal structure of dolomite, c-axis vertical
a) Layered structure b) Stereoscopic projection
of the hexagonal unit cell for dolomite (a = 4.81 A,
c = 16.00 A). .. ... ......... .......127
47 Correlation between oleate adsorption and flotation
for apatite as a function of pH .. ... .......129
48 Correlation between oleate adsorption and flotation
for dolomite as a function of pH. .. .. ........130
49 Difference IR spectra of apatite-oleate system at
pH 10, and at pH 4 in the presence and absence
of sodium chloride .. .. ...... ..........135
50 Difference IR spectra of dolomite-oleate system at
pH 10, and at pH 4 with and without sodium chloride
addition. .. .. ......... ........ ..137
Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EFFECT OF SODIUM CHLORIDE ON THE SELECTIVE FLOTATION OF DOLOMITE FROM APATITE
DURSUN E. INCE
Chairman: B. M. Moudgil
Major Department: Materials Science and Engineering
Selective flotation of dolomite from apatite was investigated
using dodecylamine hydrochloride and sodium oleate as the collector in the absence and presence of inorganic modifiers such as sodium chloride.
Separation of dolomite from apatite was anticipated from the single mineral experiments under various PH and collector concentrations. Flotation selectivity in the mixtures, however, was found to be limited. Attempts to control the flotation response of these minerals by addition of salts such as NaCl and KCl indicated that selectivity is possible. Upon optimization of the process parameters, the best selectivity was obtained at pH 4, in the presence of sodium chloride, using sodium oleate as the collector.
Electrokinetic and adsorption studies were conducted to elucidate the mechanism of observed selectivity. Zeta potential measurements xiv
demonstrated that the surface charge of apatite is reversed below its isoelectric point (pH 5.4) in the presence of sodium chloride. This was attributed to the increased rate of calcium dissolution and sodium substitution in the apatite structure.
Adsorption studies confirmed that the amount of oleate adsorbed on apatite decreases in the presence of sodium chloride relative to that in distilled water, while on dolomite the amount adsorbed remained unchanged.
FT-IR spectra of apatite in distilled water indicated the presence of oleic acid and calcium oleate on the surface. In the presence of NaCi, however, calcium oleate was not detected. This was ascribed to the surface charge reversal, which adversely influenced adsorption of anionic oleate species, in addition to the depletion of calcium sites by selective dissolution. FT-IR spectra of dolomite indicated the presence of magnesium oleate besides oleic acid, with and without the salt addition. Formation of magnesium oleate in preference to calcium oleate on the dolomite surface was explained in terms of higher charge density and electronegativity of magnesium ions,. and relatively higher rate of calcium dissolution.
Contact angle measurements on apatite and dolomite indicated a good correlation with adsorption and flotation results in the absence and the presence of sodium chloride.
The Florida phosphate rock deposits are located in the central and northern land pebble districts. The central land pebble district, as shown in Figure 1, constitutes the major phosphate rock producing region and underlies 2600 square miles in Polk, Hillsborough, Hardee, Manatee and DeSoto Counties. Until recently, the phosphate production has been confined to the Bone Valley of the central district. These deposits, however, are being depleted and the mining will shift to the southern extension of the central district. Phosphate rock from the southern extension (lower zone matrix) is lower in grade and contains significant quantities of dolomitic limestone (Ca, Mg carbonate) impurities in addition to quartz and clays (Lawver et al., 1982a). Beneficiation of Florida phosphate rock by a "double float" or Crago flotation process (Crago, 1940) has been in use commercially since 1937. However, with the conventional processing techniques, dolomite, usually reported as weight percent MgO, cannot be selectively removed from apatite, the phosphate mineral. On the other hand, it is generally agreed that more than 1% MgO in the final phosphate concentrate would present problems during chemical processing to manufacture phosphoric acid, an intermediate product used in the production of fertilizers. The presence of MgO in quantities greater than the specified amount would:
Hammon JAC SONVLLEI
1 Count Columbia
Hillb rough Polk Co nty County r-- Ak LAN TAM BA TOW
Manate T County .Hardee C
\Te Soto ounty
Central District MIAMI
Figure 1. Location of Florida phosphate deposits.
1. Increase the viscosity of the phosphoric acid resulting in
higher pumping costs (Cate and Deming, 1970);
2. Cause an increase in sulfuric acid consumption and
3. Form excessive sludge during phosphoric acid production;
4. Precipitate as complex salts (MgNH4P205) and lead to
clogging of filters (Becker, 1983);
5. Interfere in the production of certain super phosphates
Phosphate rock, which is essentially carbonate-fluorapatite,
(Fantel and Rosenkranz 1983; Whippo and Murowchick, 1967), contains magnesium impurities in one of the following forms:
1. Ionic substitution in the apatite structure;
2. Second phase dolomite in apatite;
3. Discrete dolomite particles.
The MgO content in francolite (carbonate-fluorapatite) occurs both as fine inclusions of dolomite and as Mg in the francolite structure, according to Lawver et al. (1982a). The range of "lattice Mg" calculated as MgO was found to be 0.40% from 31 samples studied by Lawyer and co-workers. Recently, Blanchard et al. (1986) have conducted a systematic investigation using 18 samples from Florida phosphate fields. Measurements of dolomite content by X-ray diffraction and chemical analyses of these samples indicated that the average excess MgO content above that accounted for by dolomite is about 0.57% by weight. This represents a reasonable estimate of the amount of MgO substituting in the apatite structure. Additionally, McClellan (1980)
showed that the "a" unit-cell dimension for carbonate-fluorapatite can also be used to estimate the Mg substitution for Ca. The average unit-cell dimensions measured for 10 samples corresponded to 0.55% (by weight) MgO in the apatite, in a good agreement with the average of
0.57% MgO estimated by Blanchard and co-workers (1986).
Physical methods of separation obviously cannot remove the
substituted magnesium. However, the second phase dolomite and discrete particles upon liberation can be separated from apatite by physical and physi co-chemi cal processes.
Separation of dolomite from apatite by selective flotation has been the focal point of research in the last decade, because physical methods such as gravity and magnetic separation have not shown much promise in beneficiating dolomitic phosphate ores. Flotation studies by the phosphate industry (Lawyer et al., 1978; Snow, 1979; Dufour et al., 1980; Lawyer et al., 1980; Lawyer et al., 1982b), the U.S Bureau of Mines (Llewellyn et al., 1982 and 1984) and the Tennessee Valley Authority (Lehr and Hsieh, 1981) have resulted in the development of a number of processes. It should be pointed out that these studies evolved from engineering applications, and an understanding of the fundamentals which govern the selectivity of the proposed processes were not fully established. This is believed to be a serious limitation in the optimization of these processes. Nevertheless, past efforts have provided a direction for developing suitable separation techniques.
A systematic study involving an apatite/dolomite-anionic surfactant system was conducted by Moudgil and Chanchani (1985a, 1985b and 1985c), and Chanchani (1984) which has resulted in the development of a "two-
stage conditioning" process. In this process, the feed is conditioned at pH 10 followed by reconditioning at a pH lower than 4.5 before flotation. Dolomite is selectively floated out following the two-stage conditioning process. Bench-scale optimization of this process using natural samples appears promising.
Flotation separation of apatite from dolomite using cationic collectors has also been studied by researchers at International Minerals and Chemicals Company (Lawver et al., 1980; and Snow, 1979), followed by Soto and Iwasaki (1985 and 1986). The role of the dissolved mineral species, flotation pH and solution chemistry of the surfactant, however, was not taken into account in the above studies.
The surface modifiers used in the past for apatite separation from carbonates are mostly phosphate salts, which are uneconomical. Also, changes in the characteristics of the matrix (ore), even from the same location, have been observed. It is conceivable that a given flotation scheme would be applicable to a specific ore. It is imperative, therefore, to develop flotation processes which possibly would have applicability to a wide variety of ores. Additionally, understanding the mechanism of selectivity would be helpful in overcoming the difficulties in processing different ores and ensuring the usefulness of the method for treating ores of different characteristics.
Characteristics of Salt-Type Minerals
Salt-type minerals such as apatite and dolomite are characterized by solubilities higher than those of most oxides and silicates, but lower than simple salt minerals such as halite and sylvite. Flotation of such minerals from the associated gangue and from each other is of major practical importance. For example, apatites constitute the largest tonnage of any raw material beneficiated by froth flotation techniques in the United States as well as several other countries (Hanna and Somasundaran, 1976).
Separation of the salt-type minerals from oxide and silicate
minerals has been achieved and used commercially. However, separation of these minerals from each other is complex and the problems involved remain unresolved in many cases. It has been reported that the differences between flotation characteristics of various salt-type minerals may not be greater than those between samples of a single mineral from different deposits (Sorensen, 1973). The similarities in the flotation response of these minerals is generally attributed to their comparable surface chemical behavior. In addition, interaction of dissolved mineral species with collector molecules is considered to contribute to the poor selectivity. Consequently, it has been suggested that the use of inorganic or organic modifying agents might result in
the desired selectivity in these systems (Klassen and Mokrousov, 1973; Joy and Robinson, 1964).
Characteristics of Apatites in Florida Phosphorites
Phosphorites are sedimentary rocks with 15 to 20% P205 content. In Florida deposits, apatite is the most abundant phosphorite mineral generally found in the form of carbonate-fluorapatite (Calo(P04)6X(C03)X FO4xF2) (McClellan, 1980). The carbonate-fluorapatite (also known as francolite) can have extensive substitutions such as carbonate for phosphate; and other cations such as Mg, Na, Mn and K for calcium (Lehr et al., 1967; McConnell, 1952, and McConnell and Gruner, 1940). Apatite has a hexagonal lattice structure and lattice parameters are dependent on the extent of substitution. It is generally composed of microcrystals which vary in size from 0.02 to 0.20 microns (Lehr et al., 1967; Smith and Lehr, 1966). High surface area of apatites has therefore been attributed to this crystal size. In addition, much of the south Florida phosphate rock is composed of grains which are mixtures of apatite, dolomite and other constituents. Phosphorite sediments are complex because they are the product of several different sedimentary systems, and are formed by intermixing of phosphates, carbonates, organic matter, glauconite, terrigenous and siliceous sediments (Riggs, 1979).
Characteristics of Dolomite
Dolomite along with calcite is the most abundant carbonate mineral. The crystal structure of dolomite is similar to that of calcite. The
description of the dolomite [CaMg(C03)21 structure is provided by retaining the calcite (CaC03) structure, but simply substituting Mg atoms for the Ca atoms in every other cation layer. The alternating CaMg arrangement of dolomite has some similarities with calcite, but the c-glide present in calcite is destroyed. Dolomite has a rhombohedral crystal structure with a Ca-0 bond length of 2.38 A, and Mg-O bond length of 2.08 R. This results in oxygen lying closer to Mg than Ca in the dolomite structure (Reeder and Sheppard, 1984)
Solubility of Apatite and Dolomite
Charge characteristics of the solid/liquid interface and chemical composition of the aqueous phase depends on the solubility of the minerals. When the minerals come into contact with water, the constituent species such as Ca++, Poi- and F from apatite, and Ca++, Mg and COJ- from dolomite will be transferred from the structure into,"/ the solution. This dissolution will continue until the chemical potentials of the species in solution and the solid phases reach an equilibrium. The solubility of the salt-type minerals often varies over a wide range, because the chemical potentials of the species in solution and in the solid are affected by a number of factors such as degree of hydration, solid solution formation and presence of other components in solid or solution. It is known that both apatite and dolomite, in contrast to simple salts, dissolve with their ions undergoing various hydrolysis and complex formation reactions.
There have been a number of publications on the solubility of apatite with marked differences in the data reported. Hanna and
Somasundaran (1976) have attributed these discrepancies to crystal structure modifications, the presence of impurities, and added electrolytes. Saleeb and de Bruyn (1972), however, have shown that a constant solubility product can be obtained if a stoichiometric compound is prepared. These investigators have reported a pK50 value of 119.1 for fluorapatite.
In the case of dolomite, the lack of understanding of its precipitation process has contributed to the discrepancy of the solubility products reported by different investigators. The published pK50 values are in the range of 16.5 to 19.5 for dolomite (Stumm and Morgan, 1981).
Surface Charge of Apatite and Dolomite
It is known that in the case of insoluble oxides the surface charge is mostly determined by OH- and H+, i.e., by the pH of the suspension in the absence of any specifically adsorbing ions. The charge of the ionic solids such as silver iodide is considered to be the result of preferential dissolution or adsorption of Ag+ or I-. The magnitude and sign of the surface charge is determined by the solution concentration of the constituent ions.
Surface charge development on minerals such as apatite and dolomite is much more complex. It is controlled by preferential dissolution of constituent species, and their hydrolysis products in addition to OH- and H+ (Saleeb and de Bruyn, 1972) ions. It has also been shown that the impurity ions present as substituents for lattice ions produce significant changes in the electrokinetic characteristics.
Somasundaran (1968) obtained an isoelectric (IEP) point of pH 4 for natural apatite that was partially saturated with fluoride. It has been reported that the nature of pretreatment plays an important role on the IEP of apatite (Somasundaran, 1972). The IEP values given in the literature for fluorapatite and hydroxyapatite vary between pH 3.8 and
8.5. Chanchani (1984) found the IEP of Florida apatite to be at pH 5.5. In the case of dolomite, Predali and Cases (1973) have determined the IEP at pH less than 7 using a natural sample from Kosice, Czechoslovakia. Chanchani (1984) has reported two IEPs (pH 5.5 and pH 10.5) for dolomite from Florida. The second IEP at pH 10.5 was attributed to the precipitation of hydrolysis products such as magnesium hydroxide on the dolomite surface.
Aging Behavior of Apatite and Dolomite
The soluble minerals such as apatite and dolomite exhibit aging phenomenon, i.e., a change in the pH of the aqueous slurry as a function of time. Aging behavior of apatite and dolomite samples from Florida has been examined by Chanchani (1984) using 1 wt % suspension of these minerals. It was shown that in the case of apatite equilibrium is reached after about 600 minutes. Initial pH of 4 and 10 was observed to shift to a value of 6 and 7. respectively, after equilibration. Similar pH shifts for apatite were also observed by Somasundaran (1968) which were attributed to the dissolution of the mineral. In the case of dolomite, the equilibrium pH was found to be between pH 8.2 and 8.5.
The past research efforts to develop a suitable technique for separation of carbonates from apatite can be divided into two broad categories, those in which differences in physical properties such as specific gravity, conductivity, hardness, etc., were utilized and those where surface chemical properties of the minerals were exploited to achieve the desired separation. A brief discussion of these efforts is presented below.
Methods Based on Physical Properties
Apatite and dolomite have relatively close physical properties such as specific gravity and hardness, thereby making it difficult to achieve their separation based on these properties. Both of the minerals are also nonmagnetic. Hence, the separation techniques based on such properties are not feasible. However, the "apparent" densities of the two minerals have been found to be sufficiently different for the possible application of heavy media separation (Lawver et al., 1982a). The difference in the specific gravity is attributed to the highly porous nature of the large dolomite particles as compared to the same size apatite particles. The effectiveness of the heavy media separation process is reported to be limited to the coarse size (pebble size) only, which is found in lesser quantity in the southern district. Selective attritioning has also been attempted by Soto and Iwasaki (1986). Results from these studies indicated that only 40-60% of the dolomite can be eliminated and that flotation is necessary for further reduction of MgO in the apatite concentrate.
Separation Based on Physico-Chemical Properties Selective flocculation
Selective flocculation of dolomite from apatite was attempted for the first time by Moudgil and Shah (1986) using polyethylene oxide (PEO) as the flocculent. It was shown that PEO flocculates dolomite, but not apatite in single mineral tests. However, mixed mineral tests did not exhibit the expected selectivity. The loss in selectivity in the case of mixed minerals was attributed to polymer-induced entrapment, which occurs due to a limited affinity of the polymer for the inert mineral (apatite). Incorporation of the polymer-coated apatite particles into dolomite flocs was explained on the basis of the higher probability of polymer bridging due to the larger number of "active" sites on dolomite as compared to apatite particles. Further efforts are underway to reduce the adsorption of the polymer on apatite so as to achieve the desired separation.
Separation of dolomitic limestone from sedimentary phosphates has been the subject of studies by various researchers. Due to the complex structure and the presence of different amorphous and porous phosphates in these sedimentary deposits, separation of carbonates such as calcite and dolomite from phosphates has not always been feasible. The amount and type of impurities present have shown considerable variation even within the same ore deposit. Despite these problems, development of a
number of flotation processes has been reported to separate dolomitic impurities using both cationic and anionic collectors.
Cationic flotation of apatite from dolomite. Flotation of apatite (francolite) from dolomite using cationic collectors was investigated by the International Minerals and Chemicals Corporation in the late seventies and early eighties (Snow, 1979; Baumann and Snow, 1980; and Lawver, 1980). These investigators developed a process which reduces the MgO content of the conventionally floated (double flotation) material to 1% or less yielding more than 90% BPL recoveries. This process involved a rougher float followed by several cleaner stages using a primary aliphatic amine in combination with kerosene as the collector.
The above process was later studied by Soto and Iwasaki (1985 and 1986) to elucidate the mechanisms involved. These researchers concluded that there is a stronger chemical interaction between the cationic collector and the phosphate ions present at the apatite surface. The selectivity was attributed to the lower solubility of the reagentphosphate compound as compared to that of the reagent-carbonate complex formed. Other noteworthy studies with cationic collectors have been conducted to separate calcite from apatite (Hanna, 1975, and Samani et al., 1975).
Anionic flotation of dolomite from apatite. Separation of
calcite and in some cases dolomite from phosphate ores using anionic collectors has been studied extensively in the past decade. A detailed review of these studies has recently been presented by Moudgil and
Somasundaran (1986) and Chanchani (1984). A brief summary of recent developments is presented below.
Flotation of dolomite from apatite in the presence of inorganic depressants such as phosphates and fluorides at pH 5.6-6.2 was studied by Lawyer et al. (1984) using fatty acids and their soaps, including petroleum sulfonates. Reportedly, the best results were obtained with sodium tripolyphosphate and hexametaphosphate using a proprietary anionic collector. This process resulted in phosphate concentrates containing less than 1% MgO at 50-90% BPL recoveries from a feed of less than 48 mesh size fraction containing about 2% MgO.
Hsieh and Lehr (1985) at TVA used diphosphonic acid to depress apatite while floating dolomite with oleic acid. This process reduced the MgO content of the concentrate to less than 1% from a feed containing 1.9% MgO at 83% apatite recovery. In another process developed by TVA (1983). H2S04 was added to the concentrate to separate calcareous phosphate ores. Selective flotation of carbonates was attributed to differential desorption of the fatty acid collector on the phosphate mineral. No experimental data was, however, presented to support this hypothesis.
Llewellyn et al. (1982) at U.S. Bureau of Mines depressed
dolomite by the addition of sodium silicate and floated apatite at pH 9.2-9.6. In cases where the MgO content of the final concentrate was not reduced to less than 1%, further removal of dolomite by S0 leaching was recommended. Rule et al. (1970 and 1985), also at USBM, depressed apatite with fluosilicic acid while floating carbonaceous impurities using fatty acid emulsion under slightly acidic pH conditions.
Dufour et al. (1980) at Minemet Recherche, France, depressed phosphate at pH 5.5 and floated dolomite after attritioning; however, the reagents used were not disclosed. The apatite recoveries ranged from 48 to 85 percent with the final product containing less than 1% MgO.
Johnston and Leja (1978) also selectively floated dolomite at pH 6, using oleic acid as the collector by depressing fluorapatite by phosphate ions. The difference in flotation was explained on the basis of adsorption of phosphate ions onto apatite by hydrogen bonding. Dolomite flotation was attributed to hydrogen bonding of oleic acid on dolomite in addition to CO2 gas evolution.
In an effort to develop new reagents that interact selectively, Eu and Somasundaran (1986) used alizarin red S, a dye that stains calcite, and achieved effective separation of calcite from apatite with sodium oleate as the collector. They found that alizarin red S adsorbs more on apatite than calcite and consequently acts as an apatite depressant.
Other noteworthy studies include that of Bushel et al. (1970); Onal (1973); Ratobylskaya et al. (1975); Lawyer et al. (1978, 1980, 1981); Kiukkola (1980); Dufour et al. (1980); Baumann and Snow (1980); Houot and Polgaire (1980); Lehr and Hsieh (1981); Clerici et al. (1984); Rao and coworkers (1985), and Atalay et al. (1985).
It should, however, be noted that most of the above studies have evolved from engineering studies. Hence, the fundamental principles governing these processes were not studied which are required for developing more efficient techniques.
A systematic and thorough study of apatite-dolomite separation using anionic collectors has been conducted by Moudgil and Chanchani (1985a and 1985b). The surface chemical and dissolution characteristics of the minerals were taken into account in addition to solution properties of the surfactant (sodium oleate), and a "two-stage flotation" process was developed. This process involves conditioning the feed at pH 10 followed by reconditioning at a pH less than 4.5 to selectively float out dolomite. Bench scale testing of this process has yielded favorable results (Moudgil, 1987).
Flotation in the Presence of Salts
As mentioned above apatite-dolomite separation has been studied
(Atalay et al., 1985; Rule et al., 1985; Hsieh and Lehr, 1985; Lawyer et al., 1984; Llewellyn et al., 1982; Johnston and Leja, 1978; and Onal, 1973) using phosphate salts as depressants for apatite. However, no systematic studies are reported on the role of the added salts on the flotation behavior of these minerals. It should be noted that in these studies phosphate minerals were depressed in the presence of phosphate salts.
Maslow (1971) and Strel'tsyn et al. (1967) have reported the adverse effect of salt addition on apatite flotation in apatitenepheline ore. These workers found that when NaCl addition exceeded 0.5 kg/ton, the ore was suppressed at pH 9, but it was overcome by addition of NaOH and sodium silicate. Additionally, Gruber et al. (1986) have reported successful separation of carbonates such as calcite from apatite in sea water from the Santo Domingo phosphate deposit in Baja
California Sur. Apatite was found to be depressed at pH less than 5.5 in sea water when fatty acid was used as the collector. In the presence of fresh water, apatite was found to be naturally depressed at pH 3.0 and below. No efforts, however, were made to study the mechanism involved in such a separation.
A sample of high-grade phosphate rock (apatite) was procured from Agrico Chemical Company (Mulberry, Florida). This sample (16x150 mesh) was screened to obtain a 65x100 mesh fraction, which was deslimed, dried and passed through an electrostatic separator after heating to 140 C to remove the silica grains. The 65x100 mesh sample was used for flotation and adsorption studies. FT-IR and electrokinetic studies, however, were conducted on a portion of this sample which was ground to
This sample, supplied by International Minerals and Chemicals Corporation (Bartow, Florida), was hand picked and crushed using a Chipmunk crusher and hand ground to maximize the yield of 65x100 mesh fraction. The dolomite was deslimed and dried at 140 C before removing silica by electrostatic separator. Electrokinetic and FT-IR studies were conducted on a portion of this sample ground to -325 mesh. The samples were stored in a glass jar and used as required.
Samples of completely crystalline New Jersey dolomite and
fluorapatite obtained from Geology Department, University of Florida (Blanchard, 1987) were used for contact angle measurements. Chemical analysis of the apatite sample indicated 37.51% P205. 0.14% MgO and
1.15% acid insoluble. Crystalline New Jersey dolomite analyzed 17.73% MgO, 0.36% P205 and 0.95% acid insoluble.
Dodecylamine hydrochloride obtained from Eastman Kodak Company, and purified sodium oleate purchased from Fisher Scientific Company, were used as the cationic and anionic collectors, respectively, for flotation experiments.
A mixture of unlabeled and 14C labeled oleic acid was used for the adsorption experiments. Unlabeled oleic acid (gold label grade) was obtained from Aldrich Chemical Company. 14C labeled oleic acid was purchased from ICN Pharmaceuticals in nitrogen-sealed ampules of 0.1 mCi.
All other chemicals such as calcium and magnesium standards, HN03, KOH, NaOH, NaCl, KCl, NaF, etc., were of reagent grade purchased from Fisher Scientific Company.
Triple distilled water of less than 1.2 micromhos conductivity was used in the flotation experiments. All other solutions were prepared using triple distilled water deaerated by bubbling nitrogen for two hours.
Flotation experiments were conducted using a modified Hallimond cell (Fuerstenau et al., 1957; and Modi and Fuerstenau, 1960) shown in Figure 2. This cell consists of two parts connected by a ground glass joint. The lower part of the cell is fitted with a fritted glass, having uniform pore size. A Teflon-coated magnetic stirring bar is used to maintain the particles in suspension. In the Hallimond cell technique, the hydrodynamic variables such as agitation, airflow rate, and flotation time can be controlled.
The complete Hallimond cell flotation assembly is illustrated in
Figure 3. The details of the set-up and operation are similar to those used by Moudgil (1972), Ananthapadmanabhan (1980) and Chanchani (1984) and are only briefly described below.
Nitrogen gas was used for the flotation experiments, which was
first purified by passing through an ascarite column (D) for removing the CO2 (see Figure 3). Subsequently, the gas passes through a water trap (T1) to remove any ascarite fines carried over. Traps (T2 and T3) are provided to prevent back suction of water to the ascarite column. The purified gas is passed to a 50 liter glass reservoir (R), which acts as a buffer tank to supply the gas at a constant pressure. The pressure in the tank is measured by the manometer (M). The outlet of the gas reservoir is connected to the Hallimond cell (HC) through a solenoid valve (V) and flow meter (F). The solenoid valve is energized using a
29/42 1 cm
9mm Tubing Jointd
Cork Stopper Nitrogen
Figure 2. A schematic diagram of a modified Hallimond Cell.
4-1 cl; Z- 00 LU Z-1
timer, which can be set to allow the gas flow and stirring for any length of time.
The apatite, dolomite, or the mixture of these minerals were aged in 100 ml volumetric flask for 20 minutes and conditioned after reagent addition for a period of 5 minutes by slow tumbling at 8 rpm. Whenever required, minerals were aged in the salt solution of specified concentration. The PH of the suspension was measured before aging and after conditioning in all cases and the latter is reported as the PH of flotation. The suspension was transferred to the Hallimond cell with minimum turbulence and floated for one minute using nitrogen gas at a flow rate of 50 ml/min. The floated and unfloated fractions were dried at 50 OC, and flotation recovery was calculated based on the weight of the dried samples. In the case of the mixtures, float and sink fractions were pulverized and leached in aqua-regia before analyzing for P205 and MgO contents.
El ectroki neti c Measurements
Zeta potential measurements were conducted using a Pen Kem Model 501 Laser Zee Meter. Suspensions of 0.1 wt % apatite or dolomite were prepared using -325 mesh sample. These were stirred with a Tefloncoated magnetic bar for approximately 21 hours at natural PH. Fifty milliliter samples of the aged suspension were equilibrated at the desired pH for 3 hours before they were transferred to the cell for zeta potential measurements.
Stock solutions of 5.0x10-3 kmol/m3 sodium oleate or dodecylamine hydrochloride were prepared in deaerated distilled water as required for flotation tests. The solution pH was adjusted to approximately 11.5 and
8.0 for oleate and dodecylamine, respectively, using NaOH or HCl. It was diluted daily and used as needed, and the stock solution was used for 5 days only.
Carbon-14 labeled radioactive oleic acid solution was also prepared as described above for the unlabeled sodium oleate solution after evaporating benzene. The stock solutions were refrigerated and used up to one week. Solutions of desired concentrations were obtained by mixing 14C labeled oleic acid with unlabeled oleate solution.
Oleate Adsorption Tests
To ensure correct adsorption measurements, the solid samples were rinsed with distilled water of the same pH value after the adsorption experiments to remove any precipitated or entrapped oleic acid from the solids. Control tests were conducted to check the oleic acid coating on the surface of the vials. The results presented in Table 1 indicated that approximately 20% of the initial collector present in the solution reported as possible coating on the vial surface. Out of the total, only 5% was determined to adsorb on the mineral (apatite) and less than 2% was found to be in the rinse solution at pH 4.5. This revealed that there was not any significant desorption of oleic acid as a consequence of rinsing of the solids. It should be noted that if the solution depletion method were to be followed in this case, the amount adsorbed
0 C 1
j c r- Ln
E 4-3 r- rS.. 0 LO cl
4-) 4- V)
(n LO 0 C C
.- la -P
41 V) Ln
-j 4-) C LO
F- 4-3 LLS- CL 4-) So C)
V) C) C) C)
cr 5- C C
(D a) CIA C\j
4-3 Cn co c
a) .- M: CY) LO 0
o :5 0 cl
4-) -P u
0 cm C) C)
(D E -4 -4
u x x
C 00 m
0 0 00 Ln a)
(-> E 4-)
on the mineral would have been calculated to be five-fold higher, resulting in misleading conclusions. In addition, adsorption of oleate on quartz was measured at pH 4 to ensure that the results obtained with apatite and dolomite do not include any coating of oleic acid on the surface. Since quartz is not expected to absorb oleate at pH 4, any adsorption of oleic acid can be presumed to be physical coating. Results of these tests indicated that there was no physical coating of oleate on quartz, for up to a sodium oleate concentration of 6.0x10-4 kmol/m3.
The amount of oleate or oleic acid adsorbed was therefore
determined in this study by direct counts on the solids in the oleic acid precipitation region, i.e., at pH 8.0 and below. Above this pH value, the amount adsorbed was determined by the solution depletion method because test results indicated that oleate adsorption on the vials is less than 0.5% of the total amount present.
Single mineral adsorption tests were conducted by adding 11 ml of sodium oleate solution of the desired pH to 0.1 g of 65xlOO mesh mineral sample in a 20 ml polypropylene vial. The samples were then conditioned by tumbling at 8 rpm for 5 minutes. The amount adsorbed on the minerals at pH 8 and below (oleic acid precipitation region) was determined by direct counts on the solids after cocktail addition. To determine the amount adsorbed above pH 8, as mentioned earlier the solution depletion method was followed. The solids were first allowed to settle for 5 minutes and the supernatant was analyzed for the amount of surfactant depleted. This method was followed because no major surfactant precipitation was anticipated at pH 8 and above.
The mixed mineral adsorption tests were conducted in a glass cell
arrangement as shown in Figure 4. This cell design allows mixed mineral conditioning of apatite and dolomite in the same medium without coming in contact with each other. A 0.5 g, 65xlOO mesh sample was used for these tests. The sample volume in each side of the cell was maintained at 50 ml and the conditioning was identical to that of single mineral experiments with respect to pulp density, conditioning time, and tumbling speed. Adsorption tests in the presence of sodium chloride were conducted in the same manner after aging the mineral samples in
2.0x10-2 kmol/m3 NaCl.
Analysis of the 14C labeled oleic acid was performed using a Beckman Model LS 1800 liquid scintillation counter after mixing the solids or solution with a scintillation cocktail, "Scintiverse-II" (obtained from Fisher Scientific Company). In the case of mixed minerals, the amount adsorbed was also determined directly on the solids in the entire pH range since the solution depletion method could not be followed.
Mineral Dissolution Tests
The dissolution of ions from apatite or dolomite at a given pH
value was determined by agitating one gram of 65xlOO mesh solids in 100 ml of distilled water for a known time interval using a slow speed (8 rpm) tumbler. At the end of the tumbling period, a 10 ml sample of the supernatant was withdrawn for analysis. The supernatant was centrifuged at 734 G (2500 rpm) for 30 minutes to remove fines created during agitation, before calcium and magnesium ion analysis was carried out
using a Perkin Elmer Plasma II inductively coupled plasma (ICP) emission spectrometer.
Solubility Product Determination
The solubility products of calcium and magnesium oleate at pH 10
were determined using the nephelometric method. Known concentrations of calcium and magnesium solutions were prepared from 1000 ppm standards and were mixed with the sodium oleate solution of desired concentration. The mixtures were initially stirred for a period of one minute and then allowed to stand for 10 minutes before making turbidity measurements using a Hach Model 2100 turbidimeter. The instrument was calibrated by measuring the turbidity of solutions containing known concentrations of sodium oleate and calcium/magnesium ions. The solubility product (Ks) was obtained using the concentrations of calcium/magnesium and oleate ions, at which a sharp increase in turbidity was observed.
Samples used for FT-IR analysis were prepared as follows: A one gram, -325 mesh size fraction, sample was suspended in 100 ml of distilled water in a 100 ml volumetric flask and the pH was adjusted to the desired value using HN03 or KOH. After 20 minutes of aging, the supernatant was partly replaced by the surfactant solution such that a concentration of 5.0x10-3 kmol/m3 sodium oleate was obtained. The suspension, with the added surfactant, was then tumbled for one hour at 8 rpm. At the end of the conditioning period, the solids were separated from the supernatant by centrifuging the sample at 734 G (2500 rpm) for
30 minutes. Next, the solids were resuspended twice in distilled water of the same pH value as the conditioning solution and centrifuged again to remove the entrapped or free oleate solution from the solids. The pure apatite and dolomite samples were also treated identically and centrifuged the same way to remove water. The characterization of adsorbed oleate species in the presence of sodium chloride was carried out by aging the solids in 2x10-2 kmol/m3 NaCl solution, followed by conditioning in a 5x10-3 kmol/m3 sodium oleate solution of the same NaCl concentration used during aging. The solid samples were then dried at 50 OC for 24 hours and stored in a vacuum desiccator until used.
Freshly ground, dried KCl was used as a reference throughout these experiments. The calcium and magnesium oleate precipitates were obtained by mixing sodium oleate solution with those of calcium and magnesium chloride, respectively. The reagentized pure mineral and the precipitate samples were first mixed with 90% KC1 before introduction into the diffuse reflectance cell.
The diffuse reflectance IR spectra were obtained on a BOMEM DA3.10 Fourier Transform Infrared Spectrometer equipped with a 25 cm path length Michelson interferometer fitted to a KBr beam splitter. The optical interferometer is connected to a high speed vector processor, which performs the Fourier Transform and numerical fitting of the collected interferograms. The spectra were recorded with a spectral resolution of 0.5 cm/sec. Typically, 64 to 256 scans were obtained under vacuum conditions to minimize interference from atmospheric moisture and carbon dioxide.
Contact Angle Measurements
The samples for contact angle measurements were prepared by cutting the rock using a diamond saw and mounting them on an epoxy base. The surface of these samples were abraded using sandpaper with 400 grids and polished using diamond cloth. The reagentization of samples was similar to the method used for FT-IR analysis, except that aging and conditioning in this case were performed by simply immersing the solid in solution for 10 minutes. The solution was agitated using a magnetic stirrer bar. After conditioning, the solid samples were dried in an oven at 30 OC before measuring the contact angle. The water droplets were dispensed onto the sample surface using a micro-syringe. The contact (tangent) angle formed between a "sessile" drop and the mineral surface was determined directly, using a NRL Contact Angle Goniometer Model 100-00. Following each test, the samples were abraded and repolished to remove the surface coating before making subsequent contact angle measurements.
The experimental plan employed in this investigation involved a) Selection, preparation, and characterization of the minerals; b) Selection of surfactant;
c) Selection of the experimental techniques; and d) The experimental approach.
Selection, Preparation, and Characterization of the Minerals
Natural apatite and dolomite samples from Florida phosphate
deposits were selected. Both apatite and dolomite samples were crushed and ground to obtain 65xl00 mesh size fractions for flotation experiments. A portion of these samples was ground to -325 mesh for electrokinetic and FT-IR spectroscopic studies. The samples were characterized for chemical composition, surface area and porosity, and surface charge behavior. X-ray analysis and Scanning Electron Microscopy studies were also carried out on the samples.
Selection of the Surfactant
Sodium oleate (anionic) and dodecylamine hydrochloride (cationic) surfactants, which constitute the active components of fatty acid and fatty amine collectors, respectively, were used in this study. It should be noted that both of these surfactants hydrolyze significantly. Nature of the dominant species present therefore would be governed by the pH of the solution.
Selection of the Exoerimental Techniques
The experimental techniques selected are described in the following section.
Microflotation tests using a Hallimond cell was selected for
flotation studies. This technique enables close control of hydrodynamic variables such as the agitation, air flow rate, and flotation time. It
should be noted that a direct correlation of Hallimond cell results to a laboratory cell in terms of percent recovery, etc., may not be appropriate, but the trends obtained are expected to be similar.
Zeta potential measurements with and without modifiers in combination with dissolution and flotation can lead to a better understanding of the various species adsorbed at the solid/liquid interface. Zeta potential of the minerals was measured by the electrophoretic mobility technique using a Pen Kem Model 501 Laser Zee Meter.
These tests were conducted to obtain information on the amount of oleate/oleic acid adsorbed on the mineral surface under different experimental conditions. Oleate adsorption measurements using 14C labeled oleic acid were conducted using a Beckman Model LS 1800 liquid scintillation counter.
To avoid experimental artifacts in the adsorption studies, the following precautions were taken:
(1) Oleate adsorption tests were conducted using the same 65x100 mesh
size fraction that was employed for flotation experiments.
(2) Reagentizing time of 5 minutes was maintained for both adsorption
and flotation to avoid variations due to kinetics involved.
(3) The exact PH value of oleic acid precipitation depends on the
total oleate concentration and the reaction constants used. Hence,
the low solubility of oleic acid in the neutral and acidic PH range
was taken into account to ensure correct analyses.
Determination of the nature of the adsorbed surfactant species
Adsorption studies are not expected to yield any information about the nature of the various oleate species such as ionic oleate monomers or acid-soap complex. Such information can be helpful in explaining the differences in flotation behavior of these minerals under different PH conditions. FT-IR spectroscopy was employed to study the nature of the adsorbed species, using a BOMEM 0A3.1O model instrument.
Mineral dissolution studies
These tests can provide information about the dissolution
characteristics of various ionic species from the minerals which can be helpful in explaining the surface charge and adsorption behavior. Analysis of the dissolved ions such as calcium and magnesium was carried out using a Perkin Elmer Plasma II, inductively coupled plasma (ICP) emission spectrometer.
These measurements under different experimental conditions can
yield information about relative hydrophobicity of the mineral surfaces. Contact angle measurements were conducted with and without salt addition, using a NRL Contact Angle Goniometer (Model 100-00).
The Experimental Approach
The experimental approach involved the following steps:
(1) Flotation tests to identify the conditions for selectivity.
a) Study of the flotation response of apatite and dolomite
individually to determine the reagent (dodecylamine and oleate) concentration and pH values where differences in the flotation
behavior are maximum.
b) Flotation of the mixed minerals to evaluate the selectivity
predicted by the single mineral tests, and investigation of the
reasons if observed selectivity is not achieved.
c) To select appropriate experimental conditions based on (b) to
achieve the desired separation.
(2) Elucidate the mechanisms by conducting the following studies:
a) Electrokinetic measurements as a function of pH with and without
chemical additives to establish the surface charge.
b) Adsorption tests under conditions selected on the basis of
c) A study of the nature of the surfactant species by FT-IR under
the conditions used for adsorption tests.
d) Evaluation of the relative hydrophobicity of the mineral
surfaces by contact angle measurements.
Characteristics of Minerals
Chemical analyses of the apatite and dolomite samples were
conducted using a Perkin Elmer Plasma Il inductively coupled plasma (ICP) emission spectrometer. It is clear from the data presented in Table 2, that the apatite sample is essentially free of dolomite and vice versa. The major impurity occurring in these samples is silica, which is reported as acid insoluble.
Surface Area and Porosity
The surface area and pore size distribution of the minerals were determined using nitrogen gas as the adsorbate with a Quantachrome Autosorb-6 unit. Surface area measurements of apatite and dolomite presented in Table 2 indicated that these samples are highly porous. Pore size distribution of apatite and dolomite samples revealed that as much as 95% of their surface area is contributed by pores less than 400 A in diameter. Average pore radius for both apatite and dolomite was determined to be 82 A. SEM micrographs of 65xlOO mesh apatite and dolomite samples presented in Figures 5 and 6, respectively, further confirm the high surface porosity. It is observed from these micrographs that the pores in the apatite sample extend to the surface.
< tm LO C)
E V 0
V) c C\j
W 00 t.0
LL.j 4-3 u C) C\j co
co C co
V) LO C\j 0)
(\i Ln C
4-) u ro S..
L) C) C)
tio C) C)
S- V) -4 r-I
LL. a) x x
b) Figure 5. SEM micrograph of apatite (65x100 mesh size fraction),
a) 200X; b) 1000X.
b) Figure 6. SEM micrograph of dolomite (65x100 mesh size fraction), a)
IOOX; b) 100OX.
Dolomite, on the other hand, appears to have a relatively rough surface morphology, but does not reveal any features of its porosity.
X-ray diffraction analysis of apatite, presented in Figure 7,
indicated the presence of a very small amount of quartz, however, even after repeated scans no dolomite was found to be associated with apatite. The dolomite sample exhibited the characteristic peak of dolomite along with minor peaks for quartz and feldspar as seen in Figure 8.
Surface Chemical Characterization
The zeta potential measurements were made using a Pen Kem Model 501 Laser Zee Meter. The isoelectric point (IEP) of apatite, as shown in Figure 9, is at pH 5.4. This value is similar to that reported by Chanchani (1984) and Somasundaran (1968). In the case of dolomite, two isoelectric points, at pH 5.3 and pH 11.1, are observed (Figure 10). The second IEP at pH 11.1 has been attributed to the presence of hydroxylated magnesium species on the dolomite surface (Chanchani, 1984; Iwasaki and Krishnan, 1983; and Balajee and Iwasaki, 1969).
Flotation Studies with Dodecylamine Hydrochloride Single mineral flotation behavior
Flotation response of single mineral apatite and dolomite was examined at two concentrations of dodecylamine hydrochloride as a
CM 4-CL 0
0 to S
C44 C4 co cc 04
0 cl to Col Vva vo
N CL 00
o 0 0
A Us "IVIJLN3.LOd VJL3Z
0 XO CL
Aw 'IVIJLN3.LOd V.L 3 Z
function of pH. These tests were followed by mixed mineral flotation tests under selected experimental conditions.
Apatite-dodecylamine system. Results of apatite flotation as a
function of pH at two levels of dodecylamine concentration are presented in Figure 11. It is observed that at a dodecylamine hydrochloride concentration of 1.0x10-3 kmol/m3, apatite recovery is 100% between pH 4 and 10, and decreases precipitously beyond pH 10. However, at a dodecylamine concentration of 1.6xi0-4 kmol/m3, apatite recovery exhibits two maxima at pH 6 and 9.8, the flotation of apatite at pH 6 being about 50% of that at pH 9.8.
Dolomite-dodecylamine system. Flotation response of dolomite as a function of pH at a dodecylamine hydrochloride concentration of 1.0x10-3 and 1.6x10-4 kmol/m3 is illustrated in Figure 12. The amount floated is observed to increase sharply at a dodecylamine concentration of 1.0x10-3 kmol/m3 above the first IEP at pH 5.3 and to reach a 100% level at pH
5.8. However, flotation recovery starts decreasing above pH 9, descending to 25% at pH 11, the value at which dolomite exhibits the second IEP (refer to Figure 10).
The amount of apatite and dolomite floated as a function of pH at a dodecylamine hydrochloride concentration of 1.0x10-3 kmol/m3 is compared in Figure 13. It is indicated from these results that apatite can possibly be recovered selectively from its mixture with dolomite at pH less than 4.5. At a dodecylamine concentration of 1.6x10-4 kmol/m3, another region of selectivity occurs at pH 9.8, as illustrated in Figure 14.
c %-. x x
E 0 0
0 cm OL
2 CO I..
LL. fu LL
%-LM '031VOIJ LNnowv
uj 1 1
%.LM 'a3.LVOI:l lNnowv
CO 4E 0
c i m u
06 0 x
%.Lm 'c3ivoiq LNnowy
%-LM 'G 3 1 V 0 -1:1 1 Nnowv
Attempts were therefore made to achieve the separation of apatite from dolomite by conducting flotation tests using synthetic mixtures of apatite and dolomite under the above specified pH conditions.
Flotation of apatite-dolomite mixture with dodecylamine
Results of 50:50 apatite and dolomite mixed mineral flotation
experiments under selected pH conditions and collector concentrations are summarized in Table 3.
Although a preferential flotation of apatite was observed at pH
9.8 and pH 4.1 at both levels of dodecylamine concentrations, the magnitude of the selectivity predicted by the single mineral experiments was not realized.
The difference in the flotation response of these minerals at pH
9.8 and at a dodecylamine concentration of 1.6x,0-4 kmol/m3 also did not correspond to the single mineral test results. In general, results of mixed mineral experiments with dodecylamine as the collector indicated depression of apatite and activation of dolomite in the selectivity ranges predicted by the single mineral tests.
Flotation Studies Using Sodium Oleate as the Collector
Flotation behavior of apatite and dolomite using sodium oleate as the collector was examined as a function of pH. Mixed mineral tests, were also conducted under specific pH conditions.
Flotation Results of 50:50 Apatite/Dolomite Mixture Using Dodecylamine as the Collector
Flotation Collector Chemical Analysis of Apatite Dolomite
pH Conc. Float Fraction, % Recovery Reject
(kmol/m3) P205 MgO (Weight %) (Weight %)
4.10.2 1.0x10-3 26.5, 26.6 4.2, 4.2 72.3, 74.7 53.4, 55.7
9.80.2 1.6x10-4 25.4, 25.9 3.8, 4.0 53.3, 51.8 37.6, 38.2
Feed: 1 gram 65x100 mesh fraction 18.0% P205, 9.5% MgO
Single mineral flotation tests
Results of flotation tests as a function of pH using sodium oleate as the collector are presented in Figure 15. At a sodium oleate concentration of 4.OX10-5 kmol/m3, flotation recovery of dolomite was observed to be 100% in the acidic pH range (pH 4.0-5.5). Dolomite recovery remained at 10-15% range between pH 7 and 10 and was seen to increase above pH 10.
Apatite flotation in the acidic pH range, on the other hand,
exhibited a maximum at pH 4. The recovery of apatite increased sharply above pH 6, and was observed to be 100% between pH 7 and 10.5. No apatite flotation was observed between pH 5 and 6 at this level of sodium oleate addition.
As observed from Figure 15, selective flotation of dolomite from apatite or vice versa was predicted in the following pH ranges.
1) Flotation of dolomite from apatite at pH 5 to 6.
2) Flotation of apatite from dolomite between pH 7 and 10.
Results of the mixed mineral flotation tests using a 50:50 apatite and dolomite mixture under given pH conditions are summarized in Table
4. It is observed that at pH 5.3, even though dolomite recovery was greater than 95%, a significant amount of apatite (33%) also reported in the float fraction. Consequently, apatite recovery in the sink fraction was reduced to approximately 67%. In the alkaline pH range, apatite recovery decreased to 68% from 100% observed in the single mineral tests. On the other hand, dolomite flotation remained at the level
0 -Oc G 0 4C
Iwo 0 C
I- M 'C31V 141 n
Flotation Results of 50:50 Apatite-Dolomite
Mixture Using Sodium Oleate as the Collector
Flotation Chemical Analysis Apatite Dolomite
pH of Concentrate, % Recovery Reject
P205 MgO (Weight %) (Weight %)
5.20.2 32.82, 33.36 1.22, 1.28 67.8, 66.1 96.0, 95.3
8.10.2 31.40, 30.64 1.32, 1.36 69.2, 68.6 92.4, 91.6
Collector Conc., 4.0x10-5 kmol/m3
Flotation Feed: I gram, 65x100 mesh size fraction, 18.0% P205, 9.5% MgO
predicted by the single mineral results. It appears that the changes in the flotation response of apatite are the major cause for the loss of selectivity in the mixed mineral system when sodium oleate is used as the collector.
It is clear from the results presented that the loss of selectivity in the mixed mineral system is primarily due to the activation or depression of either apatite or dolomite depending on whether dodecylamine hydrochloride or sodium oleate is used as the collector. Moudgil and Chanchani (1985a and 1985b), and Soto and Iwasaki (1985 and 1986) have provided detailed explanations for the loss of selectivity in this system.
In the case of the apatite/dolomite-amine system, Soto and Iwasaki (1985) proposed that the adsorption of amine on dolomite is mainly controlled by the electrostatic attraction in addition to weak chemical interaction. The fact that flotation recovery of both apatite and dolomite decreased sharply below their respective IEPs (at pH -5.4), when dodecylamine was used (refer to Figure 14) and increased in the case of sodium oleate (Figure 15), indicated that the coloumbic forces play a significant role in the adsorption of the collector on the respective substrates. It should, therefore, be possible to modify flotation behavior of dolomite by the addition of indifferent electrolytes such as sodium chloride.
Flotation tests in the presence of NaCi using dodecylamine as the
Single minerals flotation tests. Experiments conducted in the natural pH range (pH 6.7) as shown in Table 5, indicated 85-90% flotation recovery of apatite at a dodecylamine concentration of
1.6x,0Q4 kmol/m3 and a NaCl concentration of 5.0x10'1 kmol/m3. Apatite recovery was observed to be only 50% under identical conditions without NaCl addition (refer to Figure 11). In contrast, dolomite recovery under similar experimental conditions remained at the 15-20% level. It should be noted that the change expected in the flotation behavior of dolomite in the presence of sodium chloride was not observed, but more selective flotation of apatite was realized.
Mixed mineral flotation tests. Flotation of 88:12 apatitedolomite mixtures was conducted to determine the selectivity predicted by the single mineral experiments in the presence of NaCl at natural pH value. Results presented in Table 6 demonstrated that at a dodecylamine concentration of 1.6x10-4 kmol/m3 apatite can be selectively recovered from the mixture leaving dolomite in the sink fraction.
It should be noted that with dodecylamine hydrochloride, apatite, the major mineral, is floated leaving dolomite in the sink. In practice, however, flotation of the minor constituent is desired. Further test work therefore, was conducted using sodium oleate as the collector, which under slightly acidic pH conditions is known to yield flotation of dolomite leaving apatite in the sink fraction.
Effect of NaCl on the Single Mineral
Flotation of Apatite and Dolomite with
Dodecylamine Hydrochloride as the Collector
Mineral Amount Floated, Weight %
Without Salt With Salt
Apatite 50.0, 48.0 84.5, 90.8
Dolomite 17.0, 20.0 15.7, 20.2
Collector Conc., 1.6x10-4 kmol/m3
NaCl Conc., 5.0x10-I kmol/m3
Feed: 1 gram, 65x100 mesh size fraction
Flotation pH: 6.7
Results of Mixed Mineral Flotation in the Presence of
NaCl Using Dodecylamine as the Collector at pH 6.3
Test Chemical Analysis Apatite Recovery Dolomite Reject
of Float Fraction, % in Float in Sink
No. P205 MgO (Weight %) (Weight %)
1 33.65 0.97 79.0 71.8
2 33.49 1.02 79.8 69.8
3 33.59 1.03 81.2 69.1
Collector Conc., 1.6x10-4 kmol/m3
NaCl Conc., 5.0x101- kmol/m3
Feed: 1 gram 88/12 Apatite-Dolomite Mixture
Flotation studies in the presence of NaCi using sodium oleate as the
Results of mixed mineral flotation tests summarized in Table 7
indicated selective recovery of dolomite from the apatite and dolomite mixture at pH 4 in the presence of 5.0x101l kmol/m3 NaCl and at a sodium oleate concentration of 4.0x10-5 kmol/m3. Single mineral tests without NaCi addition indicated 60% apatite and 95-100% dolomite recovery at the same pH and collector concentration (see Figure 15). In the presence of NaCl apatite was found to be depressed without any significant effect on the flotation of dolomite. In order to achieve maximum separation of dolomite single mineral tests were conducted to determine the optimum pH, salt and collector concentration.
Optimum salt concentration for apatite depression. Mixed mineral results indicated the best selectivity at pH 4, in the presence of NaCl (see Table 7). Therefore, further tests to determine the optimum concentration of NaCl were also conducted at this pH value. It was determined that optimum results are obtained in the presence of 2.0x10-2 kmol/m3 sodium chloride addition (see Figure 16). It is to be noted that dolomite flotation is not affected by NaCl addition under the given experimental conditions.
Optimum collector concentration. Results presented in Figure 17 and 18, indicated that at pH 10, apatite requires less collector to float with and without salt addition, as compared to dolomite, and vice versa at pH 4. The difference in the amount of sodium oleate required to float 100% apatite or dolomite without NaCl addition was determined to be about five-fold, and it more than doubled in the presence of sodium chloride. From the data in Figure 17, the optimum sodium oleate
Results of Mixed Mineral Flotation Tests in the
Presence of NaCl Using Sodium Oleate as the Collector
Flotation Chemical Analysis, % Apatite Recovery Dolomite Reject
pH P205 MgO Weight % Weight %
10.9 18.0 9.5 100.0 0.0
8.2 18.5 9.5 100.0 0.0
6.6 18.5 9.5 100.0 0.0
4.0 33.3, 33.8 0.64, 0.48 96.0, 94.8 96.6, 97.6
Collector Conc., 4.0x10-5 kmol/m3 NaCl Conc., 5.0x10-1 kmol/m3 Flotation Feed: 1 gram 50:50 Apatite-Dolomite Mixture, 18.0% P205, 9.5% MgO
(L) *.% 40
0 E 4-1
E E E
E E x
%JLAR '03JLVOI:f LNnonv
C; c 0
0 x 0
z W 0
Cb V U
0 m o t a) 41
Ct) -j IV 3:
0 -0 0 0111,
0 W 40 V)
L) S4- r
%IM 'G31VOI:f INnowv
C M E
u mU 10 E
concentration was determined to be 4.0x10-5 kmol/m3, which incidentally coincides with that used for single mineral experiments conducted without NaCl addition. It is to be noted that at this level of sodium oleate concentration at pH 4, apatite appears to be completely depressed whereas dolomite flotation approaches 100%.
Optimum pH for separation. Results of single mineral apatite and dolomite flotation tests as a function of pH are plotted in Figure 19. It is observed that apatite remains depressed up to pH 4.2, but becomes activated above this pH value.
Apatite flotation behavior in the presence of NaCl at pH 9.5 and above is significantly different from that observed in distilled water (refer to Figure 15). On the contrary, dolomite recovery, for the most part, appears to be the same as that obtained in distilled water.
Following the above tests, flotation response of a 50:50 apatitedolomite mixture was studied as a function of pH in the presence of
2.0x10-2 kmol/m3 NaCl and at a sodium oleate concentration of 4.0x10-5 kmol/m3.
In general, the selectivity predicted by the single mineral tests (see Figure 19) was maintained in the mixed mineral systems (see Figure 20). In addition, during the flotation of the mixed minerals, a more effective apatite depression was realized in the pH range of 4.0-4.5 as compared to single mineral test results. Optimum separation was obtained at pH 4.00.2 where more than 95% apatite was recovered with 95% or more dolomite rejection. The MgO content of the sink fraction (concentrate) was analyzed to be less than 0.7% from a feed containing
0 CL E
0 0 j c
CL 4-- E
0 a) V)
%.LM 'C31VOIJ I Nnov4v
E 4E Z
EE cm a
X C) x
Q) o CL
C C\j CID 0 N
%JLM '031VOIA I Nnowv U-
Flotation studies in the presence of KCl and NaF
To examine the effect of additional salts such as KCl and NaF on
the flotation behavior of apatite and dolomite, single and mixed mineral tests were conducted. Results of the experiments with KCl in the pH range of 3.0-5.5, using sodium oleate as the collector, are illustrated in Figure 21. It is seen that in the case of single minerals about 80% of dolomite is floated at pH 4 with only 5-10% apatite reporting in the float fraction. The best results with the 50:50 mixture, however, were obtained at pH 4.5, where 75% of dolomite and 5-10% apatite was recovered in the float fraction. These tests were conducted in the presence of 2.0x10-2 kmol/m3 KCI (the optimum concentration for NaCl).
Comparison of apatite and dolomite flotation with and without
2.0x10-2 kmol/m3 NaF, as shown in Figure 22, indicated that dolomite depression is relatively higher than that of apatite. Due to the greater depression of dolomite with NaF, the difference in the floatability gap of these minerals narrowed. In fact, some exploratory mixed mineral experiments conducted at pH 4 using sodium oleate as the collector indicated that in the presence of NaF, either both of these minerals were depressed or floated together.
Effect of Salt Addition on the Zeta Potential of Apatite
Zeta potential of apatite was measured as a function of pH in the presence of 2.0x10-2 kmol/m3 NaCl (optimum concentration for apatite depression), KCI and NaF. As seen from the results plotted in Figures
"E N 0 0
E E F
0 0 0
co z 0
14 44- 0 0 c
(L) u 4
LLJ LO.CM U
%.L /A 'G3JLVO'l:i LNnonv
c E cm
10 E c x 1 0 o z E
0 E d) UJ
0 An "a CL
%-LM 'C3JLVOI:f LNnowv
23-25, the zeta potential of apatite in the presence of added salts exhibited negative values in the entire pH range (4-11) examined, indicating reversal of the surface charge below the IEP (pH 5.4). It should be noted that ionic strength was not maintained constant because, as discussed in a previous study by Chanchani (1984), KNO3 additions of up to 1.0x10-2 kmol/m3 did not significantly affect flotation recovery of either apatite or dolomite.
The zeta potential of apatite in the presence of NaCl in the pH range of 6 to 9 was found to be less negative than that in distilled water. In the presence of KC1, at pH 6 and above, the zeta potential of apatite was measured to be nearly the same as that in distilled water. The fact that both NaCl and KCl reversed the surface charge indicated that they are not indifferent electrolytes for apatite. On the other hand, because F- is a lattice ion and therefore a potential determining ion (Somasundaran and Wang, 1984)' NaF was not expected to act as an indifferent electrolyte for apatite. NaF decreased the zeta potential of apatite in the entire pH range (4-11) examined. It also rendered the surface more negative as compared to NaCl or KCl.
Effect of Salts on the Zeta Potential of DolomiteThe zeta potential of dolomite was measured in the presence of salts as a function of pH. Zeta potential versus pH curves for dolomite in the presence of NaCl and KCl, shown in Figures 26 and 27, respectively, indicated that the IEP of dolomite (at pH 5.3) is not influenced by NaCl or KCl addition. The zeta potential values measured below IEP did not appear to show any measurable change. Above the IEP,
cm N N C4
Aw '"lVIJLN3.LOd VI 3 Z
J9 c CL
c c !d
CL CL CL
(NJ C4 4)
C4 SAul '-IVI.LN3.LOd V13Z
x CL UCY cl
10 C CL
TM 0 4x 0
0 C4 0
co C C
U. 40 CL CL co 0
0 CV) OL
A -IVI.LN31.0d V13Z UL
0 0 0 0 C\i
A W '-IVI.LN3.LOd VJAZ
C c .0
CIO C4 C4 CIO
A W 6-IVI.LN3.LOd V13Z
however, both NaCi and KCl reduced the zeta potential of dolomite from
-25 mV to -15 mV between pH 8 and 10. This decrease in the value of zeta potential can be attributed to the compression of the electrical double layer. It is suggested that the compression of the electrical double layer occurred in the presence of the added electrolytes also below the IEP. However, due to the low zeta potential values, these changes could not be measured. In the case of sodium fluoride, however, zeta potential of dolomite, as shown in Figure 28, was found to be reversed below its IEP, as was the case for apatite. This is possibly due to the adsorption of fluoride ions on the surface Ca++ and Mg++ sites for dolomite since the solubility of CaF2 and MgF2 (1.6x10'10 and
8.4x10-8, is rather low.
It should be mentioned that the maximum flotation selectivity was achieved in the presence of NaCl, and the minimum with NaF. Further studies were therefore, conducted to understand the reason for charge reversal in the presence of only NaCl.
Role of NaCl in the Reversal of Surface Charge of Apatite
Charge reversal of apatite in the presence of sodium chloride could be due to the selective dissolution of calcium and/or the substitution of sodium for calcium sites in the apatite structure.
Substitution of sodium for calcium in the apatite lattice
Calcium, in addition to POJk, OH- and H+, is the potential
determining ion (PDI) for apatite (Somasundaran, 1968; Samani et al., 1975, and Somasundaran and Wang, 1984). Selective dissolution of
E C4 (D
E E 0 0
r14 CO) C14 cl
Aw 'IVIJLN3.LOd VJL3Z U-
calcium from apatite can therefore render the surface more negatively charged. Results of Ca++ and P0J- dissolution from apatite as a function of time with and without NaCi addition are summarized in Table
8. An increase in Ca++ and a decrease in P0J-dissolution in NaCl solution relative to that of distilled water is observed from these results. It is therefore possible that selective dissolution of calcium can lead to surface charge reversal of apatite. This preferential dissolution, however, was believed to be occurring as a result of sodium substitution in the apatite structure, since dissolution was found to be incongruent in the presence of added salt. This possibility was studied by 1) measuring the change in the unit cell dimensions of apatite; 2) analyzing the mineral sample for sodium content before and after treatment with NaCl solution.
The change in the lattice parameters of apatite was studied by
computerized X-ray diffraction method. Results of the tests, summarized in Table 9, indicated 0.0060.001 A decrease in the unit cell "a" dimension of apatite when conditioned with NaCl solution. This change was anticipated to be small, because the ionic radius of Na+ (0.95 A) is only slightly smaller than that of Ca++ (0.99 A) (Brescia et al., 1966). These tests were followed by determination of sodium uptake by apatite. Results shown in Table 10, indicated that the amount of sodium present in apatite after conditioning and rinsing is stoichiometrically equivalent to the amount of calcium dissolved from apatite suggesting mole per mole substitution of Na+ for Ca++ in the apatite structure.
Apatite Dissociation at pH 4 with and without Added Sodium Chloride
Time Amount of Ions Dissolved, kmol/m3
(iue)In Distilled Water In 2.0x10-2 kmol/m3
Ca++ POj- Ca++ POj05 8.75x,0-4 3.05x10-4 1.05xj103 9.45x10-5
30 1.60x10-3 6.00xj104 2.23x10-3 2.21x,0-4
60 2.25x,0-3 8.40x10-4 3.53xj103 3.57x10-4
Effect of NaCl on Unit Cell Dimensions
of Apatite Conditioned at pH 4
Salt Conc. Measured Unit Cell Dimension ( )
(kmol/m3) "a" "c"
None 9.348, 9.346 6.893, 6.894
7.5x10-4 9.339, 9.340 6.895, 6.893
2.0x10-2 9.342, 9.340 6.892, 6.894
r- CD C)
4.3 0; a;
< 0E --c%) CY)
4-) E I
C) C) C>
4-) Iq -4 -4
0 0 x x x
V) E Oll 00 r00 C C
OCY) CY) CY)
LLJ E E- E I I
-j :3 :3 4-) C) C)
cc .- a) r-I --q
< -0 -a- 0 0 x x
F- 0 0 CL E m r- ::IV)
4- qr C\j
V) a) S- 0
0) 0) a)
4- c c V)
0 0 0
o 4-)CY) C\j C\j
M E I I
S- -- a) C) C)
4-) c 1-4 -.1
a 0 0 x x
E a) E M C) C)
In order to investigate the mechanism of observed flotation
selectivity, adsorption of sodium oleate on apatite and dolomite was studied. Adsorption experiments were conducted under conditions identical to the flotation tests so that a meaningful correlation between the two could be established.
Single Minerals Adsorption Tests
The amount of oleate adsorbed on apatite as a function of
conditioning pH is plotted in Figure 29 at three levels of sodium oleate concentration. Two peaks, a larger peak at pH 8 and a smaller one around pH 5, are observed. It is seen that the amount of oleate adsorbed on apatite increased with an increase in the amount of sodium oleate added.
Adsorption data on apatite as a function of sodium oleate concentration at pH 4 and pH 10 was also obtained. Results presented in Figure 30 indicated that the amount of oleate adsorbed at pH 10 is about five-times higher than that at pH 4, i.e., at a sodium oleate concentration of 2.0x10-4 kmol/m3). At a lower sodium oleate concentration, however, this ratio appears to be different.
Effect of NaCl on sodium oleate adsorption on apatite. Selective flotation of dolomite from apatite was found to be enhanced in the presence of sodium chloride at pH 4 (see Figure 19 and 20). Adsorption experiments were therefore conducted as a function of sodium oleate concentration in order to determine the effect of NaCl (2.0x10-2
0 4Q 4 u
40 CL 4
40 0 0 L) c .0
C 4 x x x 0
c u V)
IL o C4 03
a/iotur( 'a3suosay 3.Lv3,io mvnicos
0 E 40
u C) Z
41 CL C) S
z 15110WW'038UOSCIV 3.LV3-10 U-
kmol/m3) on the amount of oleate adsorbed on apatite. Results of the adsorption experiments illustrated in Figure 31, indicated a decrease in the amount of oleate adsorbed on apatite in the presence of sodium chloride as compared to that obtained in distilled water.
Dol omi te-ol eate system
The amount of oleate adsorbed on dolomite as a function of pH is plotted in Figure 32 at three levels of sodium oleate concentration. Adsorption behavior similar to that of apatite (refer to Figure 29) is observed, however, the maxima in the alkaline pH range is shifted to pH 10 as compared to pH 8 for apatite. The amount of oleate adsorbed on dolomite at pH 10 is comparable to that on apatite at pH 8. In the acidic pH range, a smaller peak is observed at pH 4. Oleate adsorption on dolomite was found to be minimal between pH 5.5-8.5. The amount adsorbed was higher at higher dosages of sodium oleate.
Tests were also conducted at pH 4 and pH 10 to determine the amount adsorbed on dolomite as a function of sodium oleate concentration. It is observed from Figure 33, that the amount adsorbed at pH 10 is approximately three-fold higher than that at pH 4 under most of the concentration range examined.
Effect of NaCl on sodium oleate adsorption on dolomite. Results of adsorption experiments, at pH 4 in the presence of 2.0x10-2 kmol/m3 NaCl, as shown in Figure 34, indicated no measurable change in the amount adsorbed on dolomite as compared to that in distilled water. It should be noted that oleate adsorption on dolomite is nearly twice as