Journal of Minerals & Materials Characterization & Engineering, Vol. 11, No.6, pp.587-595, 2012 Printed in the USA. All rights reserved
Secondary Recovery of Colum bite from Tailing Dump in
Nigerian Jos Mines Field
1Ayeni*, F.A, 1Madugu, I. A, 1Sukop, P, 2Ibitoye, S. A, 2Adeleke, A. A, 3Abdulwahab, M
1National Metallurgical Development Centre (NMDC), P. M. B. 2116, Jos, Nigeria.
2Department of Materials Science and Engineering, Obafemi Awolowo University,
Ile-Ife, Nigeria
3Department of Metallurgical Engineering, Ahmadu Bello University, Zaria, Nigeria
*Corresponding Author:
Millions of tons of tailing dump at Rayfield mine in Jos in North Central Plateau state of Nigeria
have been found to contain large quantity of columbite. Initial attempts to recover columbite
concentrates by local miners and mineral speculators from the columbite rich tailing dump
failed due to the ineffective processing route employed. Using cone and quartering sampling
method, 0.5 kg of the columbite tailing was obtained for sieve and chemical analyses. 50 kg of
<1mm fraction of the sample was subjected to a first stage magnetic concentration in a three
poles Dry High Intensity Magnetic Separator (DHMS) that separated the columbite in the third
pole. The re-grind of the + 0.355 mm rougher concentrate fraction (containing interlocking
columbite) to pass the sieve aperture was treated on the DHMS in the second stage. The rougher
concentrate undersize and columbite pre-concentrate of the first stage magnetic separation were
then gravity concentrated on the air float machine. The Rayfield tailings and final concentrate
were assayed using ED-XRFS to obtain 12.5% and 69.6% Nb2O5, respectively. The recovery and
separation efficiency were 77.95% and 77.88% in that order. The magnetic and gravity
concentrations were found effective at 77.95% recovery for columbite from the Rayfield tailing
dump. This study also provided database for optimum recovery of columbite from tailings of
mining sites of similar composition.
Key Words: Tailing dump, columbite, recovery, concentration, characterization, Magnetic,
Gravity, Recovery, Assay, Flowsheet
588 Ayeni, F.A, Madugu, I. A, Sukop, P Vol.11, No.6
Recent economic and technological changes in mineral processing techniques have enabled old
dumped tailings of columbite to become profitable to process again. Columbite contains oxide of
niobium (Nb2O5) and oxide of tantalum (Ta2O5) in different proportions. When the niobium
oxide is much more than the tantalum oxide, the mineral is called columbite {(Fe, Mn)Nb2O6},
while it is called tantalite {(Fe, Mn)Ta2O6}, when the tantalum oxide content is much higher.
Niobium metal is best known in connection with HSLA (High Strength Low Alloy) steels, heat
resistance alloys in aerospace, vehicle engines and supersonic air-crafts [1]. Columbite is one of
the most important solid minerals traded in the world market. Hence, many miners and
processing engineers find one way or the other to acquire mining area rich in this mineral.
Millions of tons of the tailings of columbites and cassiterites mined at the Rayfield minesfield by
the British Miners in the late sixties were still very rich in the mineral [2]. In the past, when
mining and processing of columbite and cassiterite were done at this mines field, less attention
was given to the tailings either because their uses were not well defined then or there was no
appropriate method of beneficiating them.
Columbite is the main important mineral form of niobium. The columbite ore dressing usually
involves pre-concentrate and concentrate clean-up. In order to recover columbite, a flowsheet
which combines magnetic with gravity processing has been found suitable [3]. The choice of any
part of or the entire processing route would depend on the nature of the ore, particularly the
content of Nb2O5 and Ta2O5 in the ore relative to its associated minerals and impurities and
difference in their physical properties. The concentration processes may be carried out by wet or
dry gravity, magnetic or electrostatic methods to produce concentrates containing up to 70%
combined pentoxide (Nb2O5 and Ta2O5) to meet extraction requirements. However, the
universally employed method for the concentration of columbite ores are magnetic and gravity
[4]. Solid minerals are so important in the production of materials for different engineering
applications, that today’s mineral processing engineering encourages the optimum method of
beneficiating tailings for maximum secondary recovery, leaving little or no values in the final
The aim of this study is to characterize and assay the columbite tailings of the Rayfield mine
waste, and establish an effective and appropriate secondary recovery processing route for this
2.1 Materials
Vol.11, No6 Secondary Recovery of Columbite from Tailing 589
The material used for this study was collected from columbite tailing dump located in Rayfield
village, about 15 km from Jos, Plateau State, Nigeria.
2.2 Methods
2.2.1 Sample collection
The sample of the Rayfield tailings dump was collected from the site using grab sampling
method. Fifty five kilogram (55 kg) of the columbite tailings was picked randomly within a
short period of time as the representative from the tailing dump.
2.2.2 Sample preparation
Using cone and quartering method, fifty five kilogram sample (55kg) of the tailing was poured
into a conical heap, flattened and divided into four identical parts using a metal cutter. Two
opposite corners were taken as sample; the other two corners were kept aside. The portion
chosen as the sample was further coned and quartered and this continued until a sample of 0.50
kg of the columbite tailing was obtained for sieve and chemical analyses.
2.2.3 Particle size distribution analysis
The columbite tailing was subjected to screen distribution analysis on a set of sieve arranged
using geometric progression based on 2. The 500 g prepared sample of the tailing was placed
on the topmost screen and the nest of sieve was automatically vibrated for 20 minutes. The
weight of ore retained on each sieve was then determined [5].
2.2.4 Energy dispersion-X Ray fluorescence spectrometry (ED-XRF) analysis
Twenty grams of the sample was finely ground to pass through a 200-250 mesh sieve. It was
dried in a Genlab oven, model MINO/50 and type Y6D144 at 105ºC for at least 1 hour and
cooled. Thereafter, the sample was intimately mixed with a binder in the ratio of 5:1 sample(s) to
cellulose flakes binder and pelletized at a pressure of about 19.4 kg/m2 in a pelletizing machine.
At this stage, the pelletized sample(s) were stored in a dessicator for analysis. The machine, ED-
XRFS, was switched on and allowed to stabilize for 2 hours. It was set at the default mode to
analyze the compositions of the Rayfield columbite tailings and products from each processing
unit, in form of oxides of the elements present [6]. The results of analysis was reported in
percentage (%).
2.2.5 Concentration method
590 Ayeni, F.A, Madugu, I. A, Sukop, P Vol.11, No.6
Fifty kilograms of the columbite tailing was screened using the 1 mm sieve. The oversize portion
was fed into rod milling machine to reduce the size and the mill product was recycled to the 1
mm sieve. The milling was repeated until the 50 kg lot passed through the sieve. The 50 kg
undersize of 1 mm was thereafter fed into an optimum efficiency calibrated Dry High-intensity
Magnetic Separator (Rapid) Model 4-3-15 OG, operating at a feed rate of 250 kg/hr and 0.5, 0.2
and 1.0 A for magnetite, hematite and columbite; respectively. The equipment was switched on
and the sample was processed for about 15 minutes. 50 kg of the Rayfield tailing (-1.00 mm) was
fed through the hopper of the rapid and the shutter was opened slightly to allow even and gradual
spread of the feed on the magnetic belt. The magnetic belt conveyed the feed through the three
discs of the rapid. The ferromagnetic material (magnetite) in the feed was separated from the
feed at the first disc, the second disc separated rougher (hematite, escaped magnetite and some
interlocking columbite), the third disc removed columbite and the non-magnets are collected in
front of the belt. For maximum recovery, the rougher was processed further by sieving using
0.355 mm sieve size. The use of this sieve size was based on the results of the sieve analysis
which showed that 0.355 mm size fraction had the highest number of grains reported. The
oversize particles of 0.355 mm was crushed using the rod mill until all passed through the sieve.
The undersize was recycled to the Rapid for further separation.
The undersize of the 0.335 mm sieve was further reprocessed using pneumatic (air) floating
table, Kipp Kelly model MY. The products of the air floating table were rougher concentrate,
middling, and tailings. The middling was recycled to the table while the tailing made up the final
tailing. The pre-concentrate from the rapid which consisted mainly of columbite, small quantity
of silica and hematite was further processed using air floating machine operated at a tilted angle
of 60°, speed of 2.75 stroke/s. motor speed 1425 rpm and feed rate of 250 kg/hr. The products of
the air flotation were final concentrate (columbite), middling and tailing. The air floatation
procedure was repeated for the middling and tailing while the resulting tailing was added to the
tailing of the rougher process to form the final tailing of the process. Finally, samples were taken
from products of each processing unit for assaying using ED-XRFS machine (Energy
Dispersion-X Ray Fluorescence spectrometry) Minipal 4 Model 20.
3.1 Results
The results obtained on the particle size and chemical composition analyses are presented in
Tables 1 and 2 respectively, while that of mass and metallurgical balances, recovery and
separation efficiency for the process are presented in Tables 3 to 5. The secondary recovery
processing flowsheet and the particle size distributions curve are presented in Figures 1 and 2
Vol.11, No6 Secondary Recovery of Columbite from Tailing 591
Table 1: Grain size distribution of Rayfield columbite tailing
Sizes (m) Weight
weight (%)
weight (%)
+1000 6.04 1.2 1000 98.8 1.2 0.3 0.0
1000+850 17.87 3.6 850 95.2 4.8 2.5 0.4
850+710 49.79 10.0 710 85.2 14.8 3.8 0.7
710+500 75.41 15.1 500 70.1 29.9 8.2 1.1
500+355 101.55 20.3 355 49.8 50.2 11.4 1.5
355+250 137.47 27.5 250 22.3 77.7 16.8 1.9
250+180 78.22 15.7 180 6.6 93.4 17.2 2.1
180+125 23.29 4.7 125 1.9 98.1 16.1 2.1
125+90 7.46 1.5 90 0.4 99.6 8.4 1.7
90 2.06 0.4
Total 499.16
Table 2: Chemical analysis of Rayfield columbite tailing and concentrate using ED – XRF
Rayfield tailing Pre – conc. Rougher Concentrate Tailing
Al2O3 4.70 1.30 0.22 0.43 4.84
SiO2 40.3 7.00 9.30 4.51 54.60
TiO2 0.74 0.90 0.14 0.56 0.54
Fe2O3 12.90 12.36 1.17 11.18 3.48
ZrO2 7.80 1.31 76.26 1.13 18.90
Nb2O5 12.50 63.50 2.16 69.60 2.91
Ag2O 1.90 ND ND ND 0.62
Ta2O5 1.90 4.59 0.22 4.88 0.31
ThO2 2.78 0.27 0.33 0.19 1.42
Table 3: Mass balance for the process
Item Quantity(kg)
Feed 50
Concentrate 7
Rougher concentrate 19
Tailing 23.7
Loss 0.3
592 Ayeni, F.A, Madugu, I. A, Sukop, P Vol.11, No.6
Table 4: Metallurgical balance for the process
OF Nb2O5
OF Nb2O5 (%)
Feed 50 12.5 6.25 100
Concentrate 7 69.6 4.87 77.9
Rougher concentrate 19 2.16 0.41 8.3
Total Concentrate 26 20.7 5.28 86.2
Unaccounted Loss 0.28 4.5
Tailing 23.7 2.91 0.69 11.0
Table 5: Recovery and Separation efficiency for the process
Item Percentage (%)
Retrospective recovery 80.1
Separation efficiency 77.88
Grade 69.9
Fig. 1: Particle size distribution curve
Vol.11, No6 Secondary Recovery of Columbite from Tailing 593
Co lumbitetailingsfrom
Ra yfie ldMines
Undersize (-1mm)Oversize (+1mm)
Tailing Rougher P re-co ncent r ate
Screen(0.355mm)Gravity separation
-0.355mm +0.355mm
Gravity separation
Middling Rougher
con centr at e
Rougher tailing
End UserDu mp
Product Dump
Fig.2: Secondary Recovery Process Flowsheet for Processing of Rayfield Tailing
3.2 Discussion of Results
The particle size analysis gave 98.8% and 49.8% for fractions of Rayfield columbite tailing
passing 1 mm and 0.355 mm sieves. The chemical analysis of the 98.8% cumulative undersize
594 Ayeni, F.A, Madugu, I. A, Sukop, P Vol.11, No.6
fraction gave 12.5% Nb2O5. These results indicate that the tailing is rich and can be
economically processed at this size fraction assay which is far higher than 4 to 6% Nb2O5
required for columbite tailing [7]. The size distribution also indicates that the tailing is alluvial
and the form may lead to reduction in energy for comminution and hence cost of processing.
The chemical analysis result of the Rayfield tailing passing 1 mm shows 40.3% silica, 12.9%
iron oxide, 7.8% zirconia and 12.5% columbite as its major constituents. However, after
magnetic separation the pre-concentrate has a significant reduction of silica and zirconia to 7%
and 1.31% respectively and increase of columbite to 63.5%, while the iron remains almost the
same at 12.36%, which may be due to interlocking with the columbite. The result of the cleaning
operation of pre–concentrate using pneumatic table indicates the presence of 4.51% silica,
11.18% iron oxide, 1.13% Zirconia and 69.6% columbite in the final concentrate. These results
showed that the columbite content was increased by about 82.04%, while reduction percents of
88.81, 13.33 and 85.51 were obtained for silica, iron oxide and zirconia, respectively. The results
suggest that the combined magnetic/gravity concentration method produced a substantial
upgrade of the tailings.
The mass and metallurgical balance of the two final products from cleaning of pre – concentrate
shows 7 kg and 77.9% distribution of columbite in the final concentrate assaying 69.6%
respectively with a loss of 0.3 kg. Combining the final concentrate with the rougher concentrate,
mass and metallurgical balance of 26 kg concentrate, 23.7 kg final tailing and 86.2% distribution
of columbite assaying 20.7% respectively were obtained. This indicates that the proportion of
metals in the feed that reports to concentrate may have an inverse relationship with the grade [7].
Recovery and separation efficiency of 77.95% and 77.88% respectively was obtained using this
processing method. The appreciable recovery and separation efficiency achieved with very little
loss is an evidence of the effectiveness of the applied processing route for the Rayfield columbite
tailing dump.
Rayfield tailing has been characterized into particle size and composition using physical and
chemical methods respectively. ED-XRF spectrometry analysis indicates that 12.5 % columbite,
40.3% silica, 7.8% zirconia, 12.9% iron, alumina and other associated minerals form the
constituents of the tailing. The tailing has been successfully beneficiated to a higher columbite
content of 69.6% and lower silica, zirconia, iron contents of 4.51%, 1.13% and 11.18% in that
order using magnetic and gravity concentration methods. Recovery and separation efficiency
using these concentration methods or process route are established to be 77.95% and 77.88%
respectively. This is an evidence of the effectiveness of the processing route applied for the
tailing dump.
Vol.11, No6 Secondary Recovery of Columbite from Tailing 595
[1] Ogbonna, A. I., Nwakaudu S.M., Onyemobi O.O (1999), “Strategic Mineral Deposits in
Nigeria (The neglected Case of Niobotantalites)”, proceedings of 16th Nigerian Metallurgical
Society Annual Conference, Abuja; pp59.
[2] Ayeni, F. A., Alabi, O. O, Ola, S. O (1999), “Beneficiation of Tailings”; Proceedings of 16th
Nigeria Metallurgical Society Annual Conference, Abuja; pp67-75.
[3] Zlagnean, M; Sarpong, R.K (2000), “Processing of Valuable Vein Minerals, AS
Molybdenite, Monazite, Magnetite, Pyrite and Ilmenite”, Proceedings of the XXI
International Mineral Processing Congress, Vol. 4, pp C10-5
[4] Schumann, W (1985), “Collins Photo Guide for Rocks, Minerals and Gemstone”; Harper
Collins Publisher, Glasgow, pp 12.
[5] Ola, S. A., Usman, G. A., Odunaike, A. A., Kollere, S. M., Ajiboye, P. O., Adeleke, A. O.
(2009), “Pilot Scale Froth Flotation Studies To Upgrade Nigerian Itakpe Sinter Grade Iron
Ore To A Midrex-Grade Super-Concentrate”; Journal of Minerals & Materials
Characterization & Engineering, Vol. 8, No. 5, pp 405-416.
[6] MiniPAL 4 Model, (2005), “Energy Dispersive X-Ray Fluorescence Spectrometry (ED-
XRFS) PANalytical B.V”, Lelyeg1, 7602 EA, Almelo, The Netherlands, pp 3-9.
[7] Wills, B. A (1988), “Mineral Processing Technology”, 5th Edition; Pergamon Press, Oxford
pp 8-45.
[8] Gupta, C. K (2003), Chemical Metallurgy Principles and Practice; Willey – Verlag GmbH &
Co. KGaA, Weinheim.