Copyright 1999 by SME 1

SME Annual Meeting

March 1-3, 1999, Denver, Colorado
Preprint 99-3
PROCESSING OF CELESTITE FROM SIVAS-NASIR / TURKEY BY FLOTATION
T. Gler
Cumhuriyet Univ. Divrigi
Divrigi-Sivas Turkey
C. Hicyilmaz
Middle East Technical Univ.
Ankara Turkey
. Akdemir
Cumhuriyet Univ.
Sivas Turkey
ABSTRACT
Turkey has important celestite reserves in Sivas. These
reserves have gypsum and/or calcite as gangue minerals.
Celestite-calcite ore cannot be concentrated by gravity
methods due to the low efficiency of gravity methods in fine
sizes. Research was tried to concentrate celestite to a market
grade by flotation. Parameters such as pH, particle size,
dosage and types of collector and depressant were examined.
Cyanamid A-845 and Na-oleate combination as collector,
sodium silicate as depressant, and Aerofroth-65 as frother
were found as the optimum flotation reagents. By applying
3-stage cleaning, a concentrate assaying 95,34 % SrSO
4
was
obtained with 84,93 % recovery.
INTRODUCTION
Celestite (SrSO
4
) and calcite (CaCO
3
) are salt type
minerals in regard to their physico-chemical properties
(Hanna and Somasundaran, 1976). Therefore, separation of
celestite from calcite by flotation is difficult. In literature,
there are many studies made for separation of celestite and
calcite by flotation (Falconer and Crawford, 1944;
Glembotskii et al., 1961; Glembotskii and Uvarov, 1962;
Hanna and Somasundaran, 1976; Herninz and Calero, 1993;
Herninz and Calero, 1996; Hong, 1993; Houot et al., 1993;
Mishra, 1982; Taha et al., 1984; Taha et al., 1985; Uvarov
and Glembotskii, 1964; Yongxin and Changgen, 1983;
Zaman, 1985): In these studies alkaline salts of oleic acid,
particularly sodium oleate, are among the collectors
traditionally employed. Quebracho, sodium silicate, copper
sulfate and sodium metasilicate have all been used as
depressants with varying degrees of success in the celestite
flotation for improvement of selectivity. However, the high
grade market requirement cause the complication of process.
The aim of this study is to determine optimum flotation
conditions and reagents for a celestite-calcite ore from
Sivas/Nasir, Turkey, to obtain a salable concentrate having
more than 92 % SrSO
4
from an ore assaying 85.51 % SrSO
4
.
EXPERIMENTAL STUDIES
Mineralogical analysis of the ore showed that the major
constituents were celestite, calcite and quartz (Figure 1).
Table I shows the chemical analysis of the sample. The
sample taken from Sivas/Nasir deposit of Barit Maden Trk
A.S. was crushed by using jaw crusher, and ground in a
laboratory rod mill for flotation. The experiments were done
in 1.3 l flotation cell with an impeller speed of 1 100 rpm
while the pulp density was 23 % solid by weight. 60 g/t
Aerofroth-65 (Cyanamid) as frother were used during the
experiments. Conditioning time for pH regulator and
depressant was kept for 5 minutes while it was 3 minutes for
collector and 5 minutes for flotation time.
Figure 1. Result of X-Ray Diffraction
Copyright 1999 by SME 2
Table I. Chemical Analysis of the Sample
Element Content
Sr 40.78 %
Ca 3.68 %
Si 1.55 %
Al 0.55 %
Mg 2084 ppm
Fe 2059 ppm
Na 659 ppm
K 543 ppm
Ni 241 ppm
Pb 137 ppm
Co 77 ppm
Cr 69 ppm
Zn 62 ppm
Mn 48 ppm
Cu 7 ppm
RESULTS AND DISCUSSIONS
Effect of pH
A set of experiments were done to investigate the effect of
pH on flotation. pH was adjusted with NaOH or H
2
SO
4
. 375
g/t Na-Oleate was used as collector in flotation experiments
after the grinding of ore samples to 80 % minus 129 m.
Experimental results indicated that pH is very important
parameter for flotation of celestite. Grade and recovery of
concentrate increased up to pH 10 and then decreased.
Optimum flotation results were obtained at pH 10. Results of
flotation experiments to determine optimum pH value with
respect to concentrate grade and recovery are given in Figure
2. In the following experiments, pH of flotation pulp was
adjusted to 10.
Effect of Grinding Time
The sample was ground in different periods to see the effect
of particle size on flotation. In the experiments, 375 g/t Na-
oleate was used as collector. Size analyses for different
grinding times are given in Figure 3.
pH
28
41
54
67
80
93
4 6 8 10 12
Conc.grade
Recovery
Figure 2. Effect of pH on Flotation
Screen size, m
C
u
m
u
l
a
t
i
v
e

p
a
s
s
i
n
g
,

%
0
20
40
60
80
100
0 50 100 150 200 250
15 sec
30 sec
1 min
3 min
5 min
7 min
Figure 3. Screen Analysis for Different Grinding Times
Experimental results indicated that as the grinding time
increases, recovery increases up to 1 minute grinding (80 %
passing 129 m screen size) while SrSO
4
grade of
concentrate decreases (Figure 4). For coarse grinding (by 15
sec and 30 sec grinding times), poor metallurgical results may
be due to coarse size of product and insufficient degree of
liberation. Fine grinding (by 3 min, 5 min and 7 min grinding
times) caused intruding of froth by fine gangue and so grade
of concentrate slightly decreased while recovery increased.
Therefore, in the following experiments, 1 min grinding time
was used. As seen in Figure 4, salable concentrate was
obtained from sample ground for 15 seconds. However,
flotation recovery was very low.
80 % passing, m
35
50
65
80
95
0 100 200 300 400 500 600
Conc.grade
Recovery
Figure 4. Effect of Particle Size on Flotation
Effects of Type and Amount of Collectors
A series of flotation experiments were carried out to
determine the effect of collectors on flotation recovery and
concentrate grade. These collectors were generally chosen
such that they were recommended for celestite-calcite
separation by flotation. In experiments, 375 g/t collector was
used. Results of experimental studies are given on Table II.
Table II. Effect of collector type on flotation
Collector Concentrate
Weight %
Conc.Grade
% SrSO
4
Recovery
%
Oleic Acid 80.13 88.65 80.26
Linoleic Acid 71.91 87.93 73.95
Cyanamid A-801 64.73 93.76 70.97
Cyanamid A-830 61.53 88.38 63.59
Cyanamid A-845 84.96 90.33 89.75
Na-Oleate 84.07 90.98 89.45
Copyright 1999 by SME 3
As seen from Table II, salable grade was only obtained with
Aerofloat-801 (A-801), but recovery was low. Therefore,
amount of A-801 was examined by further experiments.
However desired results were not obtained.
Reasonable recovery and grade values were obtained by
using A-845 and Na-oleate as seen in Table II. So a series of
flotation experiments were carried out to determine the effect
of amount of Na-oleate, A-845, and the combination of Na-
oleate and A-845 in different ratios on concentrate grade and
flotation recovery. Na-oleate consumption was changed from
125 g/t to 625 g/t. As seen from experimental results in Figure
5, when Na-oleate consumption increased, grade decreased
while recovery increased. High amount of Na-oleate caused
the flotation of celestite as well as gangue minerals which
increased recovery and decreased grade. The results indicated
that 375 g/t collector consumption gave optimum grade and
recovery.
Na-oleate, g/t
60
67
74
81
88
95
100 200 300 400 500 600
Conc.grade
Recovery
Figure 5. Effect of Na-Oleate Concentration on Flotation
A-845 consumption was changed from 250 g/t to 625 g/t.
According to experimental results given in Figure 6, when A-
845 consumption increased, grade decreases while recovery
increased up to a certain value. High amount of A-845 caused
to decrease in recovery. This may be due to micelle formation.
The results indicated that 375 g/t collector consumption gave
optimum grade and recovery.
In the case of Na-oleate and A-845 combination, as seen
from experimental results given Figure 7, optimum result was
obtained by using 50 % Na-oleate and 50 % A-845
combination.
Aerofloat-845, g/t
76
82
88
94
200 300 400 500 600 700
Conc.grade
Recovery
Figure 6. Effect of Aerofloat-845 Concentration on Flotation
Na-oleate, %
89
90
91
92
0 20 40 60 80 100
Conc.grade
Recovery
Figure 7. Effect of Percentage of Na-Oleate in Na-Oleate and
Aerofloat-845 Combination on Flotation.
In conclusion, 50 % Na-oleate (187.5 g/t) and 50 % A-845
(187.5 g/t) combination was found as optimum collector types
and amount for flotation of this celestite ore sample. In the
following experiments this combination was used as collector.
Effects of Type and Amount of Depressants
A series of flotation experiments were carried out to
determine the effect of different depressants (caustic starch,
calgon, dextrin, albumin, gelatine, sodium silicate, gum arabic
and quebracho) on the concentrate grade and recovery.
During the using of caustic starch, calgon, dextrin, albumin
and gelatine, concentrate grade and flotation recovery
decreased as the depressant amount increased. These results
indicated that these depressants depress celestite more than
calcite and a salable product could not be obtained. In the
case of using sodium silicate, gum arabic and quebracho,
flotation recovery decreased, while concentrate grade
increased, indicated that calcite was depressed more than
celestite with these depressants. Results plotted in Figures 8
to 10 show that marketable concentrate can be obtained with
these depressants.
Figure 8 points out that when sodium silicate was used in
the amount of 500 g/t, 1500 g/t, 2000 g/t, 2500 g/t, salable
concentrate was obtained with 92.06 %, 92.94 %, 93.04 %,
95.52 % SrSO
4
grades and 90.39 %, 91.32 %, 85.12 %, 77.81
% recoveries, respectively. In the case of using gum arabic
in 100 g/t, 250 g/t, 500 g/t dosages, salable concentrate was
obtained with 92.07 %, 92.54 %, 93.10 % SrSO
4
grades and
87.81 %, 89.51 %, 58.62 % recoveries, respectively (Figure
9).
Sodium silicate, g/t
77
82
87
92
97
0 500 1000 1500 2000 2500
Conc.grade
Recovery
Figure 8. Effect of Sodium Silicate on Flotation
Copyright 1999 by SME 4
Gum Arabic, g/t
58
67
76
85
94
0 100 200 300 400 500
Conc.grade
Recovery
Figure 9. Effect of Gum Arabic on Flotation
Qubracho, g/t
69
78
87
96
0 50 100 150 200 250
Conc.grade
Recovery
Figure 10. Effect of Quebracho on Flotation
As seen from Figure 10 when quebracho was used in 50 g/t,
100 g/t, 250 g/t dosages, salable concentrate was obtained
with 92.01 %, 92.44 %, 94.78 % SrSO4 grades and 88.47 %,
89.97 %, 70.90 % recoveries, respectively.
Cleaning of the Rougher Concentrate
To increase the economic value of concentrate, 3 stage
cleaning was investigated. Optimum conditions (187.5 g/t Na-
oleate and 187.5 g/t A-845 combination as collector, 1500 g/t
Na
2
SiO
3
as depressant, grinding to 80 % minus 129 m)
found throughout this research were used in rougher stage.
Solid weight percentages in first, second and third cleaning
stages were 20.15 %, 19.37 % and 18.95 %, respectively.
Collector, frother and depressant were not used in cleaning
stages. Only pH was adjusted to 10. Flotation time was
applied as 5 minutes in rougher and cleaner stages to take
celestite in concentrate as high as possible. Results of
cleaning experiments were given in Table III. The 3-stage
cleaning of concentrate increased the grade while decreased
the recovery.
Table III. Results of 3-stage cleaning experiments.
Product Weight
%
Grade
% SrSO
4
Recovery
%
Final Concentrate 76.17 95.34 84.93
Rougher Tailing 15.98 46.44 8.68
First Cleaner Tailing 4.03 63.98 3.01
Second Cleaner Tailing 2.12 72.16 1.79
Third Cleaner Tailing 1.68 80.92 1.59
Feed 100.00 85.51 100.00
CONCLUSIONS
In the light of the flotation tests of celestite-calcite ore,
following conclusions can be drawn:
1. The highest SrSO
4
grade and recovery was obtained at
pH 10.
2. Grinding to minus 129 m gave optimum concentrate
grade and recovery.
3. A concentrate with sufficiently high grade (91.56 %
SrSO4) and recovery (90.30 %) were obtained using Na-
oleate and Aerofloat-801 combination in equal amount.
Salable concentrate was also obtained by using
Aerofloat-801, however recovery was low.
4. Sodium silicate gave the highest selectivity in calcite
depression and a concentrate assaying 92.94 % SrSO
4
was obtained with 91.32 % recovery.
5. Celestite concentrate of 95.34 % SrSO
4
with a recovery
of 84.93 % was obtained with 3-stage cleaning by using
the combination of 187.5 g/t Na-oleate and 187.5 g/t
Aerofloat-845 as collectors, 1,500 g/t sodium silicate as
depressant at pH 10.
REFERENCES
Brobst, D.A., 1990, "Barium and Strontium Minerals,"
Surface Mining, B.A. Kennedy, ed., Colorado, pp. 116-119.
Falconer, S.A., and Crawford, B.D., 1944, "Froth Flotation
of Some Nonsulphide Minerals of Strategic Importance,"
AIME.
Fulton, R.B., 1983, "Strontium," Industrial Minerals and
Rocks, Vol. 2, S.J. Lefond, ed., AIME, New York, pp. 1229-
1233.
Glembotskii, V.A., and Uvarov, V.S., 1962, "Alkali
Treatment of Some Minerals in Relation to Their Flotation,"
Dokl. Akad. Nauk Tadzh. SSR, Vol. 3, No. 158, pp. 93-95.
Glembotskii, V.A., Uvarov, V.S., and Solozhenkin, P.M.,
1961, "Some Data on Celestite Flotation," Izv. Akad. Nauk
Tadzh. SSR, Otd. Geol. Khim. Tekh. Nauk, No. 1, pp. 51-56.
Gonzlez-Caballero, F., Cabrerizo, M.A., Bruque, J.M., and
Delgado, A., 1989, "Flotation Properties of Celestite in
Aqueous Solution of Ionic Surfactants," International Journal
of Mineral Processing, Vol. 26, pp. 51-63.
Hanna, H.S., and Somasundaran, P., 1976, "Flotation of salt
type minerals," Flotation: A.M. Gaudin Memorial Volume,
Vol. 1, M.C. Fuerstenau, ed., AIME, New York, pp. 197-272.
Herninz, F., and Calero, M., 1993, "Influence of quebracho
and sodium silicate on flotation of celestite and calcite with
sodium oleate," International Journal of Mineral Processing,
Vol. 37, pp. 283-298.
Copyright 1999 by SME 5
Herninz, F., and Calero, M., 1996, "Flotation Rate of
Celestite and Calcite," Chemical Engineering Science, Vol.
51, No. 1, pp. 119-125.
Hong, W., 1993, "Celestite and strontianite: Review of ore
processing and exploitation," Industrial Minerals, Vol. 309,
pp. 55-59.
Houot, R., Joussement, R., and Fabre, F., 1993, "Celestite
Ore Upgrading: From Laboratory to Mill," XVIII.
International Mineral Processing Congress, Sydney, pp. 1053-
1059.
Kirikoglu, M.S., 1990, Endstriyel Hammaddeler, ITU
Matbaasi, Istanbul.
Mishra, S.K., 1982, "Electrokinetic properties and flotation
behavior of appetite and calcite in the presence of sodium
oleate and sodium metasilicate," International Journal of
Mineral Processing, Vol. 9, pp. 59-74.
Ober, J.A., 1995, "Strontium," American Ceramic Society
Bulletin, Vol. 74, No. 6, pp. 155-157.
Taha, F., Hluvieva, G.U., and Megakhed, A.A., 1984,
"Effects of Various Reagents Used in Flotation on
Electrokinetic Potential and Flotability of Barite and Calcite,"
Izv. Vyssh. Uchebn. Zaved., Tsvetn. Metall., No. 1, pp. 8-10.
Taha, F., Hluvieva, G.U., and Megakhed, A.A., 1985,
"Dependence Between Electrokinetic Potential and Flotability
of Barite and Calcite," Izv. Vyssh. Uchebn. Zaved., Tsvetn.
Metall., No. 1, pp. 8-10.
Uvarov, U.S., and Glembotskii, V.A., 1964, "Effect of
Sodium Carbonate and Some Modifiers on the Flotation and
the Properties of the Surface Layer of Celestite and
Anhydrite," Dokl. Akad. Nauk Tadsh. SSR, No. 7, pp. 16-19.
Yongxin, L., and Changgen, L., 1983, "Selective flotation of
sheelite from calcium minerals with sodium oleate as collector
and phosphates as modifiers," International Journal of
Mineral Process, Vol. 10, pp. 205-218.
Zaman, S., 1985, "Strontium," SME Mineral Processing
Handbook, Vol. 2, N.L. Weiss, ed., AIME, New York, pp.
29.18-29.19.