Journal of Minerals & Materials Characterization & Engineering, Vol. 8, No.5, pp 405-416, 2009
jmmce.org Printed in the USA. All rights reserved
405
Pilot Scale Froth Flotation Studies To Upgrade Nigerian Itakpe Sinter Grade
Iron Ore To A Midrex-Grade Super-Concentrate
1*
Ola, S. A.,
1
Usman G.A.,
1
Odunaike, A.A.,
2
Kollere, S.M.,
1
Ajiboye, P.O.,
3
Adeleke, A. O.
1
National Metallurgical Development Centre (NMDC), P.M.B. 2116, Jos, Nigeria.
2
Raw Materials Research and Development Council, Abuja, Nigeria.
3
Department of Materials Science and Engineering, Obafemi Awolowo University,
Ile-Ife, Nigeria.
Abstract
The sinter grade concentrate of Itakpe iron ore found in the North Central Kogi state of Nigeria
with cumulative undersize of 6.34% and 2.12% passing 90µm and 63µmsieves, respectively, and
that assayed 63.63% Fe, 5.90% silica and 0.72% alumina was ground in ball mill to produce a
pilot plant froth flotation feed with cumulative undersize of 74.37% and 42.94% for 90µm and
63µm sieves, respectively. The ball mill discharge was then treated in three stages of roughing,
cleaning and re-cleaning in a pilot plant froth flotation process line and the products obtained
gave 66.66%, 66.51%, 65.44% (Fe) and total acid (silica and alumina) contents of 4.22%,
4.39%, 5.15%, respectively, for the three stages. The super-concentrate obtained also gave
81.13%, 55.19% and 30.57% cumulative undersize for 90µm, 63µm and 45µm sieves,
respectively. The results obtained showed that the roughing stage of the flotation process
produced a super-concentrate that fall in the range of 66% to 68% for use in midrex direct
reduction plant at the Delta Steel Plant, Nigeria. The lowest total acid gangue content of 4.22%
was also obtained for the roughing stage, indicating that the two subsequent processes of
cleaning and re-cleaning were not needed. Although, the total acid content of 4.22% for the
roughing stage product slightly exceeds the upper limit of 3.50% for the Midrex process, Itakpe
sinter grade concentrates with up to 5% acid gangue has been successfully used at the Delta
Steel Plant Midrex plant.
Key Words: sinter, middrex, concentrate, froth flotation, super-concentrate

406 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5
1. INTRODUCTION
Itakpe iron ore deposit, with an estimated reserve of about 200 million tons was found in 1977.
The deposit is embedded in the Itakpe hill, near Okene, in the north central Kogi state of Nigeria.
The deposit extends approximately 3,000m in length and includes about 25 layers of ferruginous
quartzite. From a tectonic point of view, the Itakpe deposit is confined to the southern limb of a
large Itakpe-Ajabanoko anticline with enclosing rocks and conformable ore layers striking sub-
latitudinally slightly bending to the North and dipping southwards at angles ranging from 40° to
80° with local complications like minor folds. The deposit contains a mixture of magnetite and
hematite whose ratio varies throughout the deposit. The ore consists of coarse, medium and fine
grained particles. The fine ores are mainly in the eastern part of the deposit and in the thin layers,
while the coarse and medium ores are relatively mixed. However, the coarse ore particles
predominate in the north and west of the central layers and the medium sizes in the centre of the
central layers. The average content of iron in the ore is approximately 35%.[1]
The Nigerian National Iron Mining Company Ltd. (NIOMCO), Itakpe, was established to
upgrade Itakpe iron ore to sinter grade of 63% to 64% Fe for the blast furnace based Ajaokuta
integrated steel plant, Ajaokuta, Kogi state. The Itakpe sinter grade concentrate is however not
suitable for direct reduction operations at the Delta Steel Company (DSC), Aladja, Delta state,
Nigeria, and needs to be up-graded to a superconcentrate grade with 66% to 66.88% Fe. The
world production of direct reduced iron (DRI) has increased from 1 million tons in 1970 to 40
million tons currently. The Midrex process has accounted for over 60% of the annual worldwide
DRI production.[2] Steelmaking slag contains calcium oxide, magnesium oxide, silica, alumina
and other compounds in smaller concentrations. Pure silica has a very high viscosity, but the
addition of other metal oxides, except alumina, reduces the viscosity. The preferred characteristic
for DRI grade pellets is typically 67% Fe (minimum) and 3.0% (maximum) silica + alumina+
Titanium oxide.[3]
Froth flotation is a physico-chemical method of concentrating ground ores. The process involves
chemical treatment of an ore pulp to create conditions favourable for the attachment of pre-
determined mineral particle to air bubbles carrying the selected minerals to the surface of the
pulp, there forming a stabilized froth which is skimmed off and from which the pre-determined
mineral particles are recovered. Other minerals remain sub-merged in the pulp.[4]
The market requirements for higher grade concentrates of iron to improve the productivity of the
blast furnace has increased the importance of flotation process with respect to the conventional
pre-concentration of iron ore by gravity or magnetic methods. The flotation method commonly
applied is the one based on cationic flotation of silica and silicates, that is, reverse flotation [5].
The reagents required for froth flotation are collectors, frothers and regulators, such as activators
and depressants. Collectors adsorb on mineral surfaces, rendering them hydrophobic and

Vol.8, No.5 Pilot Scale Froth Flotation Studies 407
facilitating bubble attachment. The frothers help maintain a reasonably stable froth. Regulators
are used to control the flotation process to either activate or depress mineral attachment to air
bubbles and are also used to control the pH of the slurry.[6]
The aim of this study is to obtain pilot scale froth flotation process parameters to upgrade the
Itakpe iron ore sinter grade to a super-concentrate that meet the physical characteristics of less
than 30% passing 45µm that will not pose transportation and handling problems at DSC. The
study involved physical and chemical characterization of Itakpe iron ore sinter concentrate and
pilot plant upgrade of the concentrate to a super-concentrate by froth flotation for use in the
direct reduction plant at DSC.
2. MATERIALS AND METHODS
2.1 Materials
The following reagents were used for the flotation aspect of the work:
Flotigam EDA and Flotanol M both manufactured by CLARIANT were used as collector and
frother respectively. 500g/t Starch, causticized with 25% sodium hydroxide, was used as
depressant for iron minerals.500g/t. The sodium hydroxide was also used as pH regulator.
2.2 Methods
2.2.1 Particle size distribution analysis
The sinter grade iron ore, the flotation feed and the super-concentrate were subjected to screen
distribution analysis on a set of sieve. The stack of sieve with the ore charged on the topmost
sieve was clamped on the sieve shaker and shaken for 20 minutes. The weight retained on each
sieve was then determined.
2.2.2 Chemical analysis
The iron, alumina and silica contents of the ore were determined by classical wet analysis as
follows:
2.2.2.1 Determination of total iron
0.2g of the iron ore was weighed into a 25ml conical flask and 1.5ml distilled water was added to
moisten the sample. 20ml of concentrated hydrochloric acid was then added and the flask was
covered. After adding two drops of hydrofluoric acid, the mixture was placed on the hot plate
and 50% stannous chloride (SnCl
2
)was added until the solution turns colourless or green. The
mixture was covered with silica crucibles and allowed to boil on the hot plate until complete
dissolution. - drops of ferroin indicator was added to the solution. The solution was removed

408 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5
from the hot plate, the cover and the wall of the conical flask were rinsed and 5% KMnO
4
added
until yellow colour of ferroin appeared.
The solution was after heated to boiling and ferric iron was reduced by adding 15% stannous
chloride solution until the yellow colour faded away. 10 to 15mls of 1:3 sulphuric acid was
added, the solution was mixed and the wall of the flask was rinsed with distilled water until the
total volume equalsled100mls. The solution was now cooled to about 25°C and 10mls of 2%
mercuric chloride was added. The solution was allowed to stand for 5 minutes and 10mls of
0.02% Fe indicator in H
3
PO
4
(1:4) was added. The solution was then titrated with 0.1N
Potassium di-chromate (K
2
Cr
2
O
7
). The total Fe % was finally determined with the formula:
1 mls of 0.1 N K
2
Cr
2
O
7
= 0.005585g Fe
sampleofweight
XOCrKofvolumeXOCrKoftitre
Fe 100 %
722722
=
2.2.2.2 Determination of silica content
One gram of the sample was weighed into a platinum crucible and mixed thoroughly with the
fusion mixture and then covered up with some of the fusion mixture. It was fused at a
temperature of 950
o
C in a muffle furnace and allowed to cool. The fused mass was extracted
into a 400ml beaker using 50ml of 1:1 hydrochloric acid (HCI).
The extract was dehydrated on the hot plate using asbestos pad, after which it was cooled and
10ml concentrated HCI was added. Distilled water was used to make up the volume to the 50ml
mark in the beaker. It was allowed to boil and filtered through No 541 filter paper. The residue
was washed with 1:4 HCl 3 times and with hot water several times until the residue was free
from HCl. The filter paper was folded into the platinum crucible and ignited at a temperature of
950
o
C to constant weight in 20 minutes.
The crucible was then cooled in the desiccators and weighed as W1. Eight drops of 1:1 H
2
SO
4
and 15ml of hydrofluoric acid (HF) were added to the residue in the crucible and it was heated in
a sand bath placed on a hot plate to ensure slow evaporation. At the end, a copious fume was
observed leaving the dried mass. The residue was now ignited in a muffle furnace at 950
o
C for
20 minutes. It was cooled in a desiccators and the crucible weighed as W2. The percent weight
of silica was calculated with the formula:
s
W
xWW
SiO 100)21( %
2
−
=
Where: W1 = Weight of the platinum crucible with the material before hydrofluorisation.
W
s
=Weight of the sample.
2.2.2.3 Determination of alumina content

Vol.8, No.5 Pilot Scale Froth Flotation Studies 409
One gram of the sample was weighed into a nickel crucible and 10 pellets of sodium hydroxide
and a pinch of sodium carbonate was added. The mixture was melted over hot plate and finally
fused over a low flame at 700
o
C for 15 minutes.
The crucible was set aside and allowed to cool in air. When cooled, the crucible along with the
content was dropped in a 250ml beaker and 100ml of hot distilled water was slowly added and
boiled for 10 minutes. The crucible was removed with a glass rod and washed thoroughly with
water into the beaker. The hot solution was filtered with No 40 filter paper into a 500ml conical
flask.
The precipitate was then washed with hot water for 6 times to ensure that no sodium aluminate
remains trapped in the precipitate. The filtrate was acidified with hydrochloric acid and 25ml
standard EDTA was also added. The pH was adjusted to 5.5 with ammonium hydroxide and
hydrochloric acid and the solution boiled for 15 minutes.
The solution was allowed to cool and 25ml acetic acid ammonium acetate buffer solution was
added. The pH was checked and adjusted to 5.5 and drops of xylenol orange indicator were
added and a lemon colour developed. Standard zinc acetate was used to tiltrate the solution. The
colour changed from lemon yellow to purple indicating the end point. From the titre value the
Al
2
O
3
percentage was calculated with the formula:
s
W
xXxTA
OAl 100)( %
32
−
=
Where;
A = Volume of zinc acetate equivalent to 25ml EDTA solution.
B= Volume of zinc acetate solution required for titration after adding 25 ml EDTA
solution to the 5 ml standard
32
OAl solution
T = The titre value of zinc acetate with the sample under Experiment
W
s
= Weight of sample
The results of the chemical analysis are presented in Table 7.
2.3 Pilot Plant Beneficiation Method
The sinter grade concentrate was fed into the ball mill (ø 100 x 1500mm) at 500kg/ hr. The pilot
plant flotation operation was carried out sequentially in a bank of flotation cells of 0.13m
3
capacity each as shown in Fig. 1. The sinter grade iron ore was treated with 110g/ton of Flotigam
EDA as collector and 10g/ton of flotanol M frother. The slurry pH was adjusted with sodium
hydroxide to be 10.5 and treated with maize starch causticised with 25% NaOH as the depressant
for iron oxide. The slurry at 50% solids was then conditioned in the bank of 4 roughing flotation
cells for 2 minutes. The charged slurry having 40% solid was deslimed at 5 µ m size using 2mm
hydrocyclones operating at 50 Psi pressure. The cyclone underflow having 33% solids was
delivered to the 4 cleaner cells and cleaned for two minutes. The cleaned slurry at 33% solids

410 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5
was after re-cleaned in the bank of 2 re-cleaning cells for another 2 minutes. The mass balance
was done by taking samples with pulp densities cans at some points in the process lines and
working out the corresponding weight % at each point.
3. RESULTS AND DISCUSSION
3.1 Results
The results obtained on the chemical composition and particle size analyses are presented in
Tables 1 to 7, while the flotation flowsheet, mass balance for the process and curves for the
particle size distributions are presented in Figures 1 to 5.
Table 1: Grain size distribution of Itakpe iron ore sinter grade concentrate.
S/N Sieve
sizes ( µ m)
Weight%
retained
on sieve
Cumulative%
undersize
Cumulative%
oversize
1.
2.
-710
-500+355
11.13
11.81
88.87
77.06
11.13
22.94
3. -355+250 14.39 62.67 37.33
4. -250+180 18.50 44.17 55.83
5. -180+125 16.79 27.38 72.62
6. -125+90 13.00 14.38 85.62
7. -90+63 8.04 6.34 93.66
8. -63+45 4.22 2.12 97.88
9. -45 2.00 0.65 99.35
Table 2: Gates-Gaudin-Schuhmann and Rosin Rammler data for Grain size distribution of Itakpe
iron ore sinter grade concentrate.
S/N Cumulative
% undersize
Log
Cumulative %
undersize
(x 10
-1
)
Log
Cumulative
Rosin
Rammler.
Sieve
sizes
( µ m)
Log
sieve
aperture.
1. 88.87 19.5
0.34
-710 2.85
2. 77.06 18.90 0.16 -500+355 2.70
3. 62.67 18.00 -0.006 -355+250 2.55
4. 44.17 16.50 -0.23 -250+180 2.40
5. 27.38 14.40 -0.49 -180+125 2.26
6. 14.38 11.60 -0.81 -125+90 2.10
7. 6.34 8.00 -1.17 -90+63 1.95
8. 2.12 3.30 -1.66 -63+45 1.80
9. 0.65 -1.90 -2.19 -45 1.65

Vol.8, No.5 Pilot Scale Froth Flotation Studies 411
Table 3: Grain size distribution of sinter grade pilot plant flotation feed.
S/N Sieve sizes .
(µm)
Flotation feed,
cumulative (%)
undersize.
Flotation
feed,
cumulative(%)
oversize.
1.
2.
-500
-355+250
99.83
99.20
0.17
0.80
3. -250+180 99.58 0.42
4. -180+125 97.58 2.42
5. -125+90 90.33 9.67
6. -90+63 74.37 25.63
7. -63+53 42.94 57.06
8. -53 8.22 91.78
Table 4: Gates-Gaudin-Schuhmann and Rosin Rammler data for Grain size distribution of Itakpe
iron ore sinter grade pilot plant feed.
S/N Cumulative %
undersize
Log
Cumulative%
undersize
Log
Cumulative Rosin
Rammler (x 10
-1
)
Sieve
sizes
( µ m)
Log
sieve
aperture
1.
2.
99.83
99.20
1.999
1.996
8.10
6.80
-500
-355+250
2.70
2.55
3. 99.58 1.998 7.40 -250+180 2.40
4. 97.58 1.989 5.70 -180+125 2.26
5. 90.33 1.956 3.70 -125+90 2.10
6. 74.37 1.871 1.30 -90+63 1.95
7. 42.94 1.633 -2.50 -63+53 1.80
8. 8.22 0.915 -0.107 -53 1.72
Table 5: Grain size distribution of pilot plant super-concentrate.
S/N Sieve
sizes
( µ m)
Super-concentrate,
cumulative(%)
undersize
Super-concentrate,
cumulative(%)
oversize
1.
2.
-500
-355+250
99.76
99.70
0.24
0.30
3. -250+180 9.22 0.78
4. -180+90 97.34 2.66
5. -90+63 81.13 18.87
6. -63+45 55.19 44.81
7. -45 30.57 69.43

412 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5
Table 6: Gates-Gaudin-Schuhmann and Rosin Rammler data for Grain size distribution of Itakpe
iron ore super-concentrate.
S/N Cumulative%
undersize
Log
Cumulative%
undersize
Log
Cumulative
Rosin Rammler
(x 10
-1
)
Sieve
sizes
( µ m)
Log
sieve
aperture
1.
2.
99.76
99.70
1.999
1.999
7.80
7.60
-500
-355+250
2.70
2.55
3. 99.22 1.997 6.90 -250+180 2.40
4. 97.34 1.988 5.60 -180+125 2.26
5.
81.13
1.909 2.20 -90+63 1.95
6.
55.19
1.74 -0.95 -63+45 1.80
7.
30.57
1.49 -4.40 -45 1.65
Table 7: Chemical analyses of Itakpe sinter concentrate and super-concentrate grades.
Parameters
(%)
Itakpe sinter
grade
Stages in pilot plant flotation
Roughing Cleaning Re-cleaning
Fe
SiO
2
Al
2
O
3
SiO
2
+Al
2
O
3
63.63
5.90
0.72
6.62
66.66
3.71
0.51
4.22
66.51
3.88
0.51
4.39
65.44
4.55
0.60
5.15

Vol.8, No.5 Pilot Scale Froth Flotation Studies 413
Fig.1: Pilot Plant Flotation Scheme.
Itakpe Sinter Grade
Concentrate (as received).
Pilot Ball Mill
- 180 µ m
Flotation Cells
100 63.63 100
Superconcentrate
96.10 65.21 98.50
Fig. 2 Overall Mass Balance Flowsheet for the Production of Itakpe
Superconcentrate by Froth Flotation.
Ore Bin
Ball Mill
Roughing Cells
Cleaning Re-cleaning
Super
Concentrate
Tailings
Wt. %
Recovery
Fe %
Recovery
Fe %
3.90 24.47 1.50

414 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5

Vol.8, No.5 Pilot Scale Froth Flotation Studies 415
3.2 Discussion of Results
Particle size distribution analysis gave 77.06% and 99.83%, 27.38% and 97.88%, 2.12% and
42.94% for fractions of sinter grade concentrate and pilot plant feed passing 500µm, 180µm and
63µm sieves. These results indicate that ball mill grinding produced a great increase in the
cumulative undersize at 180µm from 27.38% to 97.88% and at 63µm from 2.12% to 42.94%.
The cumulative under sizes obtained at 500µm, 180µm and 45µm for the super-concentrate were
99.76%, 97.34% and 30.51%, respectively. The 30.57% obtained for the super-concentrate at
45µm is far higher than 0.65% for the pilot plant feed. The results obtained showed that the super-
concentrate obtained meet the specification of 30% minimum of grains passing 45 microns to
produce pellets for direct reduction.[7] The cumulative percent plots for the grain size distribution
gave median size of about 178µm, 162µm and 150µm with approximate cumulative undersize
percents of 26.30%, 95.46% and 94.76%, for the sinter grade concentrate, ball mill discharge and
super-concentrate, respectively. These results indicate a progressive decrease in size and rapid
increase in cumulative percent passing the respective sieve apertures.
The Fe content of 63.63% determined for Itakpe sinter grade concentrate to be upgraded to a
super-concentrate is higher than 63.22% in use at the China Anshan iron and steel company and
exceeds the standard specification of 63% for use at the Ajaokuta steel plant blast furnace.
However, the silica and alumina contents of 5.90% and 0.72%, respectively; obtained for the
Itakpe concentrate gives a total acidic oxide content of 6.62% that is far above the upper limit of
3.0% for Midrex direct reduction process. For some other standards, the total acid oxides
include titanium oxide.[8] These analysis results indicate that Itakpe concentrate needs to be
further upgraded for direct reduction by the Midrex process.
For the roughing, cleaning and re-cleaning stages of the flotation process, Fe and total acid
oxides of 66.66% and 4.22%, 66.51% and 4.39% and 65.44% and 5.15%, respectively were
obtained. These results indicate that the intermediate products at roughing and cleaning stages
satisfy the range of 66% to 66.8% Fe content required for the Midrex process, while the final re-
cleaning stage gave a Fe content of 65.44 that fall below this range. Similarly, the total acid
oxides increased from 4.22% at roughing stage to 4.39% and 5.15% at the last two stages. The
Fe content of 66.66% determined for the roughing stage falls within the range of 66-66.8% for
direct reduction at DSC. Thus, the result suggests that the cleaning and re-cleaning stages are not
necessary to upgrade Itakpe sinter grade ore to super-concentrate by flotation. Although the
gangue contents in the three stages are slightly above the required standard of 3.5% maximum
for direct reduction, Itakpe enhanced grade concentrates of up to about 5% gangue have been
successfully processed at DSC.

416 Ola, S. A., Usman G.A., Odunaike, A.A., Kollere, S.M., Ajiboye, P.O., Adeleke, A. O. Vol.8, No.5
4. CONCLUSIONS
The sinter grade of the Nigerian Itakpe iron ore that assayed 63.63% Fe and total acid gangue of
6.62% has been successfully upgraded to a super-concentrate with a higher Fe content of 66.66%
and lower acid gangue of 4.22% at the roughing stage of the pilot plant froth flotation operation.
The Fe content meets the requirement for a Midrex-grade super-concentrate, while the acid
gangue content only slightly exceeds the upper limit of 3.5%. However, Itakpe sinter concentrate
with up to 5% acid gangue has been successfully reduced at DSC Midrex plant.
REFERENCES
1. Umunakwe, P.U.(1985). “Developing a new mine- The Itakpe case”. Proc. Annual Conf. of
Nigeria Mining and Geosciences Society, Jos, Nigeria.
2. (http://www.midrex.com/uploads/documents/Direct, 2008).
3. (http://www.columbia.edu/itc/ldeo/lackner/E4900/Themelis3.pdf, 2008).
4. (http://www.miningbasics.com/html/fundamentals_of_froth_flotation.php, 2008).
5. Houot, R. (1983). “Beneficiation of iron ore by flotation-Review of industrial and potential
applications”. Int. J. of mineral processing, 10, 183-204
6. Wills, B.A.(1992). “Mineral Processing Technology”, 5
th
Edn., Pergamon Press, Oxford.
7. Raw Materials and Products Specifications for Steel Industries, Federal Ministry of Mines,
Power and Steel, Abuja, Nigeria, 1994
8. (http://www.columbia.edu/itc/ldeo/lackner/E4900/Themelis3.pdf, 30
th
October 2008).