Journal of Minerals & Materials Characterization & Engineering, Vol. 8, No.5, pp 367-378, 2009
jmmce.org Printed in the USA. All rights reserved
Characterisation and Pre-concentration of Chromite Values from Plant
Tailings Using Floatex Density Separator
C. Raghu Kumar1*, Sunil Tripathy1 and D.S. Rao2
1R&D Department, TATA Steel Limited, Jamshedpur, India
2Mineralogy Dept., IMMT, Bhubaneswar, India
*Corresponding Author, email: craghu@tatasteel.com, phone: +919204058854
ABSTRACT
Classification is a method of separation of fines from coarse particles and also lighter particles
from heavier particles. The conventional classifiers, such as, hydrocyclone or mechanical
classifiers, decreases the efficiency of the grinding and concentration circuits due to their
imperfect separation. In the process of improving the efficiency of classification, a device that
has been gaining popularity in recent years is the teeter-bed or hindered-bed separator such as
Floatex density separator. Generally for processing chromite ores, different types of gravity
methods are employed after crushing, grinding followed by classification. The Tata Steel
Chrome Ore Beneficiation (COB) plant is generating 50 tph of tailings assaying 17% Cr2O3. A
critical review on practice of the plant operating personnel is concerned in the grade-recovery
characteristics of unit operations. But separation insight and influence of different operating and
process parameters are essential to understand and control the process. The objective of the
present investigation was to study the effect of the important operating variables on floatex
density separator and preconcentration of COB plant tailings for the further beneficiation
process and found that significant removal of iron bearing mineral such as goethite and silica is
possible using FDS in a single stage operation. The maximum of 83% recovery of chromite is
possible with 22 to 23% Cr2O3 content and thus obtained FDS underflow is suitable for flotation
circuit. A low teeter water flow rate with a high bed pressure removes iron bearing mineral like
goethite efficiently in an FDS.
Key Words: Hindered settling, Floatex density separator, Classification and Chromite plant
tailings.
368 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
1. INTRODUCTION
Classification is a method of separation of fines from coarse particles and also lighter particles
from heavier particles. This is performed on the basis of the velocity with which the grains fall
through a fluid medium generally water or air [1]. In view of the fact, that the velocity of
particles in a fluid medium is dependent not only on the size, but also on the specific gravity and
shape of the particles. The conventional classifiers, such as, hydro cyclones or mechanical
classifiers, decreases the efficiency of the grinding and concentration circuits due to their
imperfect separation. Several attempts have been made to improve the efficiency of
classification. They include the use of screens instead of classifiers [2], the use of cone classifiers
to process hydro cyclone under flow [3,4] and two stage classification by hydro cyclones [5,6,7].
In the process of improving the efficiency of classification, a device that has been gaining
popularity in recent years is the teeter-bed or hindered-bed separator such as Floatex density
separator. The upward flow of elutriation water creates a fluidized “teeter-bed” of suspended
particles. The small interstices within the bed create high interstitial liquid velocities that resist
the penetration of the slow settling particles. As a result, small/light particles accumulate in the
upper section of the separator and are eventually carried over the top of the device into a
collection launder. Large/heavy particles, which settle at a rate faster than the upward current of
rising water, finally pass through the fluidized bed and are discharged out through the bottom of
the separator as underflow.
The major Indian chromite deposits are located at Sukinda region of Orissa state. The depletion
of high grade ore resources and having a variety of gangue minerals such as goethite, serpentine,
olivine and talc, etc., has lead to the utilisation of lean ore after beneficiation. Further, ample
amount of plant tailing generation have gained importance from the economics, conservation and
ecology point of view [8].
Generally for processing chromite ores, different types of gravity methods are employed after
crushing, grinding followed by classification. The Tata Steel Chrome Ore Beneficiation (COB)
plant, designed to produce concentrate of +46% Cr2O3 with a recovery of 70% from a feed of 30-
35% average Cr2O3. Presently the COB plant produces 50 tph of tailings analysing 17% Cr2O3,
which is high [9]. This suggested for the incorporation of additional circuit comprising of hydro
cyclone for pre-concentration by de-sliming and multigravity separator and wet high intensity
magnetic separator for the upgradation up to the required quality. For recovering chrome values
from the Karagedik Concentrate tailings a circuit comprising of wet high intensity magnetic
separator and column flotation for producing a concentrate assaying 46 to 48% Cr2O3 was
studied by Guney et. al [10].
Vol.8, No.5 Characterisation and Pre-concentration of Chromite 369
A critical review on practice of the plant operating personnel is concerned in the grade-recovery
characteristics of unit operations. But separation insight and influence of different operating and
process parameters are essential to understand and control the process. The objective of the
present investigation was to study the effect of the important operating variables on floatex
density separator an d pre-concentration for the further beneficiation process.
2. EXPERIMENTAL
The Sukinda COB Plant tailing sample of Tata Steel, India was the feed material in the present
studies. The plant tailing as received sample was subjected to characterization in terms of size
and chemical assay, size wise microscopic liberation studies. The mineralogical studies were
carried out with the help of stereomicroscope and reflected light microscope.
The experimental campaign was undertaken in a lab scale Floatex density separator (Model No.
LPF-0230), supplied by Outokumpu of 230 mm X 230 mm cross section and 530 mm high
(square tank height) followed by a 200mm high conical section. The Floatex density separator
(FDS) can be divided into three main zones,
the upper zone (zone A ) above the feed inlet,
the intermediate zone (zone B) between the feed inlet, and teeter water addition
point, and
the lower section (z o n e C ) below the teeter water addition point.
Feed slurry is introduced to the FDS tangentially through a centralized feed well that extends to
approximately one third of the main tank length. Fluidizing (teeter water) is introduced over the
entire cross-sectional area at the base of the teeter chamber through evenly spaced water
distribution pipes. As the feed enters the main separation zone it expands into a teetered or
fluidized bed as a result of the rising current of water. The teeter water flow rate is dependent
upon a) feed particle size distribution, b) density and c) the desired cut-point for the separation.
The separation takes place in zone B and the separated lighter/finer particles and the
coarser/heavier particles leave the separator through zone A and zone C respectively. This
separator is equipped with a pressure sensor mounted in zone B above the teeter water pipes and
an underflow discharge control valve. The pressure, sensed by a level sensor, is transmitted to
the underflow control valve using a specific-gravity set-point controller. The instrumentation
helps in maintaining a constant height of the teeter bed and a steady discharge of the underflow.
FDS is an efficient hydraulic classifier for classify the material based on their slip velocity. The
slip velocity is the relative velocity between the particles and the water velocity and is the
function of size and density of the minerals [11]. A schematic diagram of Floatex density
separator is presented in Figure 1.
370 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
About 20 tests were conducted with different combination of operating variables which is shown
in table 1. Both the underflow and overflow products for each experiment were collected, dried,
weighed and subjected to granulometry and chemical analysis. The experimental data were
scrutinized and the performance of the FDS was quantified in terms of cut size (D50),
Imperfection (I) and the Cr2O3 percentage recovery of the underflow in each condition. The
effect of teeter water flow rate and bed pressure was evaluated.
Table: 1 Design of tests with floatex density sep a r a tor
Variables Level
1 2 3 4
Teeter water flow rate (in lpm) 6 8 10 12
Bed pressure (in bar) 0.06 0.0650.07 0.075
3. RESULT AND DISCUSSION
3.1. Characterisation Studies
As received sample contains 17.76% of Cr2O3 and the major impurities are Fe(T) 22.28%, Al2O3
22.40%, SiO2 5.71%, MgO 3.39%, CaO 0.17% and LOI of 12.55%. The chemical analysis of
each size fraction was carried out in an ICP analyser and the analysis data is shown in Table 2. It
Overflow Launder
Fluidization
water
Discharge valve
Feed
B
A
C
Figure 1. Floatex density separator
Vol.8, No.5 Characterisation and Pre-concentration of Chromite 371
may be seen that the 250 micron size fraction contained 32.5% by weight and assayed 10.3%
Cr2O3. Whereas less than 25 micron size fraction contains 11.67% Cr2O3 with maximum Fe(T)
i.e 33.16%. It can also be seen from the table that the Cr2O3 content increasing as size decreases.
Where as in the case of iron there is no much variation upto plus 25 microns size. From the
Figure 2 it is evident that 80% of the particle size below 410 microns whereas 50% of the sample
is below 195 microns. It has been observed that particles below 25 microns size are 19.52% by
weight.
Table: 2 Size analysis and size wise chemical analysis of the as received sample.
Mesh size
(micron) Wt%
Retained Assay Value (%)
Cr2O3 Fe(T) Al2O3 SiO2 CaO MgO LOI
+500 11.43 8.81 22.46 30.22 6.46 0.17 1.55 18.51
+250 21.10 11.46 20.61 27.76 5. 21 0.10 2.06 15.64
+150 22.26 24.95 19.01 23.20 4. 64 0.18 4.93 11.28
+104 9.48 28.75 17 .73 21.66 4.76 0.15 5.69 1 0 .07
+74 5.26 20.68 16 .96 21.87 5.00 0.11 4.62 1 0 .99
+53 4.33 22.34 17 .46 20.85 5.22 0.11 4.49 1 1 .07
+37 4.47 24.85 19 .80 20.40 5.48 0.16 4.85 9.19
+25 2.15 28.24 20 .67 19.29 5.38 0.20 5.29 9.14
-25 19.52 11.67 33.16 12.76 7. 89 0.26 1.89 10.25
3.2. Mineralogical Characterisation Studies
Mineralogical Characterization studies were carried out for different size fractions using
stereomicroscope and reflected light microscope. XRD studies of some of the sieve fractions
372 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
were also carried out to confirm the mineralogical results. Liberation studies were carried out for
the sieve classified samples viz. +500µ; +300µ; +210µ; +150µ; +100µ and -100µ samples.
Vol.8, No.5 Characterisation and Pre-concentration of Chromite 373
3.3. Microscopic Studies
Chromite: Chromite occurs as euhedral to subhedral crystal grains, rarely angular and elongated
grains are also noticed. Occasionally chromite grains were fractured. Chromite is more enriched
in Fe than normal as alternation product from chromite. Rarely the alteration of chromite leads to
box-work type texture (Fig.4e). Many times chromite grains are locked either within the iron ore
minerals (goethite/hematite) or within silicates or the chromite with inclusions of silicate
(Figs.4a and b).
Iron minerals: Goethite and hematite are the two main iron ore minerals. Goethite occurs as
massive, colloform bands, botryoids and also as highly friable forming fine matrix. Because of
this fineness it gives a lateritic coating on the sample (Fig.3 +100µ). Hematite occurs as irregular
masses, streaks, laths, and very intimately and intricately associated with goethite. Hematite
contains inclusions of chromite and vice versa (Figs.3c and d).
Silicate: Generally serpentine and quartz form the silicate matrix in the sample. Quartz is coarse
grained and liberated at a coarse size (Fig.3, +300 µ). Many a times it is observed that the
serpentine (Fig.3; +500µ) and locked with in the chromite grains (Fig.3; +210 µ and +150µ).
3.4. Effect of Teeter Water on Cut Size (D50)
The relation of cut size (D50) with teeter water flow rate and bed pressure is shown in Figure 5.
From the figure it is evident that there is an increase in the bed pressure and teeter water flow
rate will increase the separation size or cut size from 35 to 125 microns, which has a large impact
on the maximum quantity of particles transport to the overflow. This can be explained that when
bed pressure increases, the teeter bed height increases, which is a function of teeter water flow
rate and bed pressure, there by pushing the coarse particles to overflow launder will increase i.e
the distribution of coarse particle is more in overflow. As a result the cut size increases with
increase in the bed pressure and teeter water flow rate.
3.5. Effect of Teeter Water on Recovery of Cr2O3
The effect of teeter water flow rate on the percentage recovery of Cr2O3 at different bed pressure
is presented in Figure 6. It can be observed from the Figure 6 that with an increase in the teeter
water flow rate from 6lpm to 12lpm there is a decrease in the recovery of Cr2O3 in underflow.
For example at constant bed pressure (0.06 bar) the recovery to underflow decreases from
93.86% to 86.33%. Similarly at higher bed pressure (0.075 bar) the recovery to underflow is very
low i.e 65.92%. This decrease in the recovery of Cr2O3 at underflow is attributed due to the
increase in the teeter water flow rate through which the fluidized column will become more
374 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
loosened and the upward water current may force the fine heavies and coarse lights report into
the over flow fraction along with the fine light.
Vol.8, No.5 Characterisation and Pre-concentration of Chromite 375
3.6. Effect of Bed Pressure
Floatex density separator works based on the principle of hindered settling. Richardson and Zaki
proposed the particle slip velocity, which is the relative velocity between the particle and water
velocity, is a function of the terminal settling velocity and liquid fraction of the suspension.
When the slip velocity of the particle is equal to the interstitial teeter water velocity, the particle
will have a zero velocity with respect to stationary viewer and will have an equal chance to
report either to overflow or to the underflow stream. Moreover, the interstitial teeter water
velocity is related to the voidage (liquid fraction) of the bed, decreases with increase in bed
pressure. The pressure setting determines the accumulation of the deposited material inside the
unit and hence the bed height inside the FDS. As the bed height increases the average density
and viscosity of the suspension goes up increasing the resistance for the particle to settle. Thus
the cut size increase which fa vo r ably reject the gangue minerals.
3.7. Effect of Teeter Bed Pressure on Recovery of Cr2O3
The effect of teeter bed pressure on percentage recovery of Cr2O3 to underflow is shown in
Figure 7. It can be illustrated from the figure that there is a decrease in the recovery of Cr2O3 to
underflow as teeter bed pressure increases. That shows an increase in the bed height owing to
which more loosed fluidized column, the upward water current may force the coarse particles
along with the fine chromite particles reporting into the over flow fraction. In other words the
coarse lighter and fine heavier particles, having the same slip velocity segregate in the teetered
column and push them to overflow fraction. Further this may be elucidated that the bed pressure
controls the performance of the classifier.
3.8. Classification Performance
Classification is a significant unit operation for separating the mixtures of minerals into two ore
more fractions on the basis of the velocity with which the grains fall through a fluid medium.
Floatex density separator is a hydraulic classifier which works on the principle of hindered
settling and fluidisation. Therefore, the performance is evaluated in terms of imperfection (I)
which is defined as
I = (D75 – D25) / 2 D50
Where D75 is the particle size where 75% of the mass reported to the underflow, D25 is the particle
size where 25% of the mass report to the underflow and D50 is the cut size where 50% of the
mass report either to underflow or overflow. The lower imperfection (I) value indicates better
classification efficiency which explains the degree of misplacement of the fine particles to the
underflow or the short-circuiting of the feed to overflow without any classification.
376 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
From the Figure 8 it is clear that, as the cut size (D50) increases the separation efficiency or
imperfection (I) increases. This can be explained as the cut size increases the coarse light
particles will also be pushed to overflow fraction along with the fine heavy particles. A linear
dependence of misplacement of particles i.e imperfection (I) on cut size (D50) is evident from the
Figure 8. The correlation coefficient was found to be reasonable at 0.83. As the key operating
variables i.e. teeter water flow rate and bed pressure which has the direct impact on cut size. So
Vol.8, No.5 Characterisation and Pre-concentration of Chromite 377
at higher bed pressure or teeter water the coarse light to the overflow has a predominant in the
classification.
4. CONCLUSION
It may be concluded from the results that significant removal of iron bearing mineral such as
goethite and silica was possible using FDS. In a single stage operation with FDS, a maximum of
83% recovery of chromite is possible with 22 to 23% Cr2O3 content. Thus obtained FDS
underflow is suitable for further enrichment using flotation or any other techniques. A low teeter
water flow rate with a high bed pressure removes iron bearing mineral like goethite efficiently.
ACKNOWLEDGEMENT
Authors are thankful to Dr. D. Bhattacharjee, Chief R&D and SS and General Manager, Sukinda
for giving an opportunity to work on this project. Special thanks are due to the COB plant
personnel especially Mr.R.N.Behra for helping in sample collection and valuable discussion. The
support and services provided by staff of R & D division are also duly acknowledged.
REFERENCES
[1] Heiskanen, K., 1993, Particle Classification, Chapman and Hall, London.
[2] De Kok, S.K., 1975, “Fine sizing in milling circuits.” Journal of SAIMM, October, pp 83-86.
[3] Heiskanen, K., 1979, “Two stage classification.” World Mining, Vol. 32, pp 44-46.
[4] Hukki, R.T., and Heiskanen, K., 1981, ”Two stage hydrallic classification: A report on
industrial application.” 110th Annual AIME meeting, Chicago.
[5] Lynch, A.J., 1977, Mineral crushing and grinding circuits, Elsevier, New York.
[6] Luckie, P., Hogg, R., and Schaller, R., 1980, “A review of two fine particle processing unit
operations – Classification and Mixing.” Fine particle processing, Somasundaran (Ed),
Vol.1, pp 167-180.
[7] Rogers, R.S.C., Hukki, A.M.., Steiner, G.J., and Arterburn, R.A., 1981, “An evaluation of the
use of two vs. one stage hydro cyclones in a pilot scale ball mill.” 110th Annual AIME
meeting, Chicago.
[8] Guney A., Onal G., and Atmaca T., 2001, New aspect of chromite gravity tailings
reprocessings, Vol. 14, pp 1527-1530.
[9] Rao S.M., Chandrakala K., Kapure G., Nath G., and Rao N.D., 2006, “Recovery of chromite
values from chrome ore beneficiation plant tailings.” Tata Search, pp 59-64.
378 C. Raghu Kumar and Sunil Tripathy Vol.8, No.5
[10] Guney A., Sirkesi A. A., Gurkan V., and Onal G., 1996, “The recovery of chromite fines
from the tailings of Uckopru chromium plant using HIWMS.” Changing scopes in mineral
processing, pp 149-154.
[11] Sarkar. B., Das A, and Mehrotra S.P., 2008, “Study of separation features in floatex density
separator for cleaning fine coal.” Int. J. Miner. Process., Vol. 86, pp. 40–49.
[12] Richardson J. F., and Zaki W. N., 1954, Trans. Inst. Chem. Eng., Vol. 32, pp. 35.