Powder Technology 191 (2009) 240246

Contents lists available at ScienceDirect

Powder Technology
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p o w t e c

Inuences of Jameson otation operation variables on the kinetics and recovery of

unburned carbon
M. Uurum
Nide University, Mining Engineering Department, 51100 Nide, Turkey

a r t i c l e

i n f o

Article history:
Received 8 February 2008
Received in revised form 15 October 2008
Accepted 19 October 2008
Available online 5 November 2008
Keywords:
Waste lter powder
Unburned carbon
Jameson otation
Flotation kinetics

a b s t r a c t
The purpose of this study was to investigate the inuences of Jameson otation operation variables on the
recovery and kinetics of unburned carbon (UC). The waste sample of petroleum coke, lter powder or y ash,
used in the experiments was collected from lime calcination plant tailings. The effect of Jameson otation
parameters on the recovery and kinetics efciencies of UC was systematically studied. The feasibility of
separating unburned carbon and refuse was determined from the combustible recovery (CR) and ash
reduction (AR) (%) curves. Within the range studied, the optimum diesel oil dosage was 3500 g/tonne, pine
oil dosage was 2500 g/tonne, pulp density was 15%, wash water rate was 0.17 cm/s and downcomer
immersion depth was 50 cm. The results indicate that the Jameson otation technique is effective in
removing the UC from waste lter powder. Furthermore, the classical rst-order kinetic otation model
(R = R [1 exp (k t)]) was applied to data from the tests. The model was evaluated by statistical technique,
after non-linear regression on the model parameters. It is found that the classical rst order otation kinetic
model, most extensively used among otation models, ts the tests data very well.
2008 Elsevier B.V. All rights reserved.

1. Introduction
Flotation is one of the most complex mineral processing operations
and is affected by a large number of variables. Many of these are
beyond the control of the mineral engineer, and some cannot even be
measured quantitatively with the available instruments. The relations
between measured and controlled variables are intricately related.
Sometimes simultaneously changing various component settings will
reinforce a particular attribute. In addition, various component
settings can cancel or counteract each other if changes are not chosen
wisely. For coal nes (0.5 mm), froth otation is the most effective
method of separating ash forming mineral matter from the carbonaceous material. Currently, froth otation is the only effective and
economical means of recovering 0.15 mm coal on an industrial scale
[1].
Flotation is increasingly used in waste treatment, especially in the
mining and metallurgical industry. Furthermore, the introduction of
new, superior, otation devices should lead to better applications for
remediation of mineral industry contaminated waters and solids [2].
The industrial use of otation in the minerals processing industry has
experienced a remarkable growth in recent years. Many new otation
devices have been developed. They can be grouped into three
categories: otation machines, otation columns, and otation cells.
Flotation machines have been widely applied in the minerals
processing industry all over the world. Flotation columns also have
Tel.: +90 388 225 23 50; fax: +90 388 225 01 12.
E-mail address: cevher@nigde.edu.tr.
0032-5910/$ see front matter 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.powtec.2008.10.014

many new areas of application, such as de-inking recycled paper pulp

and de-oiling water. Flotation cells are novel otation developments.
They are the Jameson cell, pneumatic cell, contact cell, centrioat cell,
and LM Flotation Cell [3].
The Jameson cell is a patented technology which uses a plunging
jet to generate air bubbles and results in a more effective otation
process [4]. It has major advantages over conventional otation
technologies, including a more compact design and lower capital cost,
a minimum amount of maintenance and the ability to operate at
elevated temperatures [5].
The Jameson cell can be divided into three main zones [6] as
described in reference Fig. 1. The downcomer is where primary
contacting of bubbles and particles occurs. Feed pulp is pumped into
the downcomer through an orice plate, creating a high-pressure jet.
The plunging jet of liquid shears and then entrains air, which has been
naturally aspirated. Due to a high mixing velocity and a large
interfacial area there is rapid contact and collection of particles. The
tank pulp zone is where secondary contacting of bubbles and particles
occurs and bubbles disengage from the pulp. The aerated mixture
exits the downcomer and enters the pulp zone of the otation tank.
The velocity of the mixture and large density differential between it
and the remainder of pulp in the tank results in recirculating uid
patterns, keeping particles in suspension without the need for
mechanical agitation. The tank froth zone is where entrained
materials are removed from the froth by froth drainage and/or froth
washing.
Compared to mechanical cells, the functions of producing bubbles
and particlebubble collision/attachment in a Jameson cell are done

M. Uurum / Powder Technology 191 (2009) 240246

241

Fig. 1. The schematic view of the Jameson otation cell.

separately inside the downcomer. A schematic diagram of the

downcomer is shown in Fig. 2. The following steps occur within the
downcomer [7]:
1. the jet created by the slurry passing through the orice promotes
the inducement of air into the downcomer;
2. shearing action of the jet generates ne bubbles and transports
them through the mixing zone;
3. particles and the bubbles collide and attach to each other and
subsequently travel down the downcomer; through the pipe ow
zone;
4. bubbles are removed by hydrostatic pressure from the downcomer
creating a vacuum for further air entrainment.
In froth otation, air bubbles are injected into a moving stream of
aqueous slurry containing a mixture of particles, so that only
hydrophobic ores collected on the bubble surface exit the stream.
Owing to its simplicity, the process is widely used for separating a
great variety of solid particles. However, a number of complex
chemical and physical interaction aspects are involved in the literature
on the study of this process. Of these, the kinetic approach has been
highly instrumental in a better understanding leading to reasonably
accurate predictions [8]. Kinetic models are often used to analyze
batch otation data and to evaluate the effect of various parameters
such as otation chemicals and equipment operating conditions for
otation processes [9]. Agar et al. [10] applied the well-known
classical rst-order kinetic model to optimize the design of otation
circuits. They introduced the concept of optimum otation time by
maximizing the recovery difference between the valuable and gangue
minerals. Dowling et al. [11] presented a comprehensive review of
otation kinetic models. More recently, Yuan et al. [12] studied several
kinetic models for a complex sulphide ore otation. Lynch et al. [13]
provide an extensive discussion of otation models. One important
aspect of the kinetic models is that the model parameters should in
some way be characteristic of a otation process [14]. Among the
many otation models, the classical rst-order otation model is
widely used and can be utilized to optimize the design of otation
circuits [9,10,15].
R = R 1 expkt 

where R is the recovery at time t, R is the ultimate recovery and k is

the rst-order rate constant. Two parameters which are R (ultimate

Fig. 2. Schematic representation of the Jameson cell downcomer.

recovery) and k (rst-order rate constant) are obtained from the

model t to an experimental recovery-time curve [9].
2. Materials and methods
The waste sample of lter powder used in the experiments was
collected from Kaksan Lime Stet. Plant tailings basin in Adana, Turkey.
The chemical analysis of the sample is given in Table 1. The caloric
value of the measured sample was 2082 kcal/kg. The volatile matter
of sample was determined as 19.9%, and the ash content and the
moisture content of sample were analyzed 65.89% and 1.49%,

Table 1
Chemical composition of the sample
Element

CaO

SiO2

Na2O

MgO

K2O

Al2O3

TiO2

LOI

37.17

0.76

0.06

0.12

0.01

0.32

0.001

3.18

0.03

53.24

242

M. Uurum / Powder Technology 191 (2009) 240246

Fig. 3. Particle size distribution of the sample.

Fig. 4. The effect of collector dosages on the combustible recovery (%) and ash reduction
(%).

respectively. The size distribution of the sample is given Fig. 3. The

reagents used were diesel oil as collector and pine oil as frother. All
experiments were conducted at pH 6.57.0, where no occulation of
gangue mineral was observed. The solution of H2SO4 prepared
stability was observed to be higher and froth volume was larger,
whereas with coarse as 1 wt.% solution was used as pH modier.
Laboratory otation tests were carried out in a Jameson otation cell,
constructed in stainless steel, 10 cm in diameter and 75 cm in length.
The downcomer was 2.5 cm in diameter and 100 cm in length. The
collectors were added in solution and conditioned for 5 min. The pulp
was then oated for 1, 2, 3, 5 and 8 min with wash water. The
hydrophobic particles (mainly consisting of UC particles) and tailings
were ltered and dried in an oven at 90 5 C to constant weight and
assayed. Tap water was used throughout the experiments. Chemical
analysis was undertaken at the laboratories of Kaksan Co., Adana,
Turkey. Experimental details of the otation tests are summarized in
Table 2. In the evaluation of experimental results, combustible
recovery (%) and ash reduction (%) of the concentrates were
considered. The combustible recovery (CR) and ash reduction (AR)
in the concentrates were calculated for 8 minute cumulative concentrate as follows:
CRk = C t 41XC t 4100=F41XF 

ARk = 1XC t =XF 4100

program (Curve Expert 1.3, shareware from http://www.ebicom.net/

~dhyams/cvxpt.htm) was used to determine the otation rate constant
(k), the ultimate recovery (R) and correlation coefcients (R2).
3. Results and discussion
3.1. Inuence of collector dosage
The collector dosage is a critical otation parameter in the otation
experiments. Adjusting the collector dosage can change the surface
character of the particles and inuence the recovery efciency. Diesel

where F is the feed mass, C(t) is the concentrate mass and XF, XC (t) are
the gravimetric mass fractions of ash in the feed and concentrate,
respectively.
In this study, the fractional recoveries after 1, 2, 3, 5 and 8 min of
otation time were tted to the models. A proprietary curve-tting
Table 2
Jameson otation column design and operating specications
Column variable
Diameter (mm)
Height (mm)
Downcomer diameter (mm)
Downcomer height (mm)
Froth height (mm)
d80 (mm)
pH
Collector dosage (diesel oil) (g/t)
Frother dosage (pine oil) (g/t)
Pulp density (%)
Wash water rate (L/min)
Downcomer immersion depth (cm)
Flotation time, (min)

100
750
20
1000
Variable
0.300
6.57.0
2000, 2500, 3000, 3500, 4000
1000, 1500, 2000, 2500, 3000
5, 7.5, 10, 15, 20
0, 0.200, 0.500, 0.800, 1.200
20, 30, 40, 50, 60
1, 2, 3, 5, 8

Fig. 5. Fitted to data set for a) 2000 and 2500. b) 3000 g/t, 3500 and 4000 g/t. Collector
dosages on the UC recovery (%) (Exp. = Experimental).

M. Uurum / Powder Technology 191 (2009) 240246

243

Table 3
Parameters obtained from model (R = R[1 exp(kt)]) t to data sets of collector
dosages
Collector
dosage (g/t)

2000

2500

3000

3500

4000

R
k
R2

0.915
0.612
0.999

0.883
0.607
0.999

0.915
0.621
0.999

0.972
0.512
0.998

0.941
0.542
0.997

oil was chosen as collector in the trials. In this study, ve different

collector dosages (2000, 2500, 3000, 3500 and 4000 g/tonne) were
used. The inuence of collector dosage on the unburned carbon
recovery (CR) and ash reduction (AR) efciency is shown in Fig. 4.
From Fig. 4, it is seen that 3500 g/tonne collector dosage gave the best
results for CR and AR as 94.8%, 92.9% respectively. Therefore, 3500 g/
tonne was determined as the optimum collector dosage. Test results of
collector dosages were tted to a rst-order kinetic otation model
(Fig. 5) and the ultimate recovery (R), otation rate constant (k) and
correlation coefcients (R2) were obtained and given in Table 3. The
correlation coefcients (0.999, 0.999, 0.999, 0.998, 0.997) showed that
the model ts the experimental data very well.
3.2. Inuence of frother dosage
A bubble produced in water is unstable. One of the prerequisites for
a successful otation operation is the stability of the bubbleparticle
aggregate. A frother has a number of functions in otation. First, it
reduces the surface tension of the airliquid interface in order that a
stable bubble is produced in the system. Secondly, it inuences the
kinetics of bubbleparticle adhesion. Thirdly, it thins the liquid layer
by interacting with collector molecules. Finally, it stabilizes the
bubbleparticle aggregate [16,17]. In order to determine the effect of
the frother concentration on the Jameson otation performance, pine
oil was used as frother. The frother concentration directly affects the
bubble size (by controlling bubble coalescence), selectivity and
recovery. The main objective of this study is therefore to determine
the optimum frother dosage. The Jameson otation experiments were
carried out with frother dosages, in the range 10003000 g/tonne. The
results are presented in Fig. 6. Among the studied frother dosages,
2500 g/tonne gave the best values for CR (%) and AR (%) with 96.5%,
91.2% respectively. Test results of frother dosages were tted to a rst
order kinetic otation model (Fig. 7) and ultimate recovery (R),
otation rate constant (k) and correlation coefcients (R2) were given
in Table 4. The high correlation coefcients (R2); 0.999, 0.996, 0.997,
0.999 and 0.998 show that a rst-order kinetic otation model (R = R
[1 exp (kt)]) provides a good t to the data.

Fig. 7. Fitted to data set for a) 1000, 1500 and 2000 g/t. b) 2500 and 3000 g/t. Frother
dosages on the UC recovery (%) (Exp. = Experimental).

3.3. Inuence of pulp density

Pulp densities used in otation columns are generally lower than
those used in mechanical otation cells. It is well known that pulp
density is an important parameter, so column otation experiments
were carried out at different pulp densities (7.520%) in the Jameson
otation cell in order to determine its effect on the recovery of UC.
Results of otation tests are plotted in Fig. 8. When the pulp density
increased from 5% to 10%, the CR (%) increased from 87.7% to 90.6%,
respectively. Increasing the pulp density however brought about a
signicant decrease in the AR (%) up to 15%. When the pulp density
was 15%, the combustible recovery and ash reduction were 87.7% and
78.5%, respectively. A pulp density of 15% gave essentially the same CR
as a density of 5%. In the mean time, 5% and 15% pulp densities for AR
(%) gave 85.5% and 78.5% values. Obviously, a pulp density of 15% is to
be preferred for economic reasons. Test results with various pulp
densities were tted to a rst-order kinetic otation model and are
shown in Fig. 9. In Table 5, the kinetic parameter values (R, k) and
correlation coefcients (R2) are given for various pulp densities. The
correlation coefcients obtained (0.998, 0.997, 0.997, 0.998, 0.997)
show that a rst order otation model (R = R [1 exp (kt)]) is a very
good t to the data.

Table 4
Parameters obtained from model (R = R[1exp(kt)]) t to data sets of frother dosages

Fig. 6. The effect of frother dosages on the combustible recovery (%) and ash reduction (%).

Frother
dosage (g/t)

1000

1500

2000

2500

3000

R
k
R2

0.886
0.665
0.999

0.935
0.514
0.996

0.935
0.514
0.997

1.004
0.427
0.999

0.952
0.620
0.998

244

M. Uurum / Powder Technology 191 (2009) 240246

Table 5
Parameters obtained from model (R = R[1 exp(kt)]) t to data sets of pulp densities
Pulp density
(%)

7.5

10

15

20

R
k
R2

0.881
0.619
0.998

0.952
0.508
0.997

0.919
0.609
0.997

0.970
0.294
0.998

0.900
0.345
0.997

10 cm above the top of the column) was used and wash water was
added uniformly over the entire froth surface.
Experiments were run at different wash water rates (0 cm/s = 0 L/
min, 0.043 cm/s = 0.200 L/min, 0.11 cm/s = 0.500 L/min, 0.17 cm/
s = 0.800 L/min, 0.26 cm/s = 1.200 L/min) in the Jameson otation cell
and the results are shown in Fig. 10. It can be seen that in all cases, a

Fig. 8. The effect of pulp densities on the combustible recovery (%) and ash reduction (%).

3.4. Inuence of wash water rate

Wash water is introduced into otation froths to reduce the amount
of entrained material reporting to the product. However, the behaviour
of wash water in a froth is not well understood at present, and detailed
information is sparse [18]. The froth phase is extremely important in
the operation of a otation cell, because, it is critical in determining the
amount of unwanted gangue collected in the concentrate which, in
turn, affects the purity of the product. In conventional otation cells,
the wash water is generally either added uniformly by sprays over the
entire froth surface, or by a single row of sprays near the weir over-ow
[19]. In this study, a sprays-type wash water system (which was located

Fig. 10. The effect of wash water rates on the combustible recovery (%) and ash
reduction (%).

Fig. 9. Fitted to data set for a) 5% and 7.5%. b) 10%, 15% and 20%. Pulp densities on the UC
recovery (%) (Exp. = Experimental).

Fig. 11. Fitted to data set for a) 0, 0.200 and 0.500 L/min. b) 0.800 and 1.200 L/min. Wash
water rates on the UC recovery (%) (Exp. = Experimental).

M. Uurum / Powder Technology 191 (2009) 240246

245

Table 6
Parameters obtained from model (R = R[1 exp(kt)]) t to data sets of wash water
rates
Wash water
Rate (L/min)

0.200

0.500

0.800

1.200

R
k
R2

0.825
0.360
0.996

0.886
0.455
0.999

0.971
0.339
0.999

100.16
0.429
0.999

0.869
0.199
0.991

wash water rate of 0.17 cm/s. gave best results with a CR of 87.7% and
an AR of 85.5%. A rst-order kinetic model was tted to the data as
shown in Fig. 11. In Table 6, the kinetic parameters (R, k) and the
correlation coefcients (R2) are given. The correlation coefcients (R2)
(0.996, 0.999, 0.999, 0.999, 0.991) show that the model ts the
experimental data quite well.
3.5. Inuence of downcomer immersion depth
The downcomer is the heart of the Jameson cell and is where
primary contacting of air bubbles and particles occurs. Feed pulp is
pumped into the downcomer through a slurry lens, creating a highpressure jet. The jet of liquid shears and then entrains air, which has
been naturally aspirated. Due to a high mixing velocity and a large
interfacial area, there is rapid contact and collection of air bubbles/
particles. The downcomer residence time varies from 10 to 30 s [20]. It
is well known that downcomer immersion depth is an important
variable in Jameson cell otation. Fig. 12 shows the results obtained at
the end of the tests performed at various downcomer immersion
depths. When the downcomer immersion depth was 20, 30 and
40 cm, the combustible recovery (CR) was 93.0%, 91.1% and 92.3%,
respectively. When the downcomer immersion depth was 50 cm, the
combustible recovery decreased to 90.0%. At this immersion depth,
the ash reduction (AR) reached its highest value (93.7%). The test
results suggest that the optimum immersion depth is 50 cm based on
the AR data alone, because the CR at the various immersion depths has
essentially the same value. Test results at the various downcomer
immersion depths were tted to a rst-order kinetic otation model,
as shown in Fig. 13. In Table 7, the ultimate recovery (R), the rate
constant (k) and the correlation coefcient (R2) values are given. High

Fig. 13. Fitted to data set for a) 20 and 30 cm. b) 40, 50 and 60 cm. Downcomer
immersion depth (cm) on the UC recovery (%) (Exp. = Experimental).

correlation coefcients (R2; 0.997, 0.998, 0.998, 0.998, 0.998) show

that model ts to unburned carbon recovery was exceptionally good.
4. Conclusion
A study was carried out to determine the main trends on the
impact of various operating and design parameters on Jameson
otation recovery of unburned carbon (UC). The conclusions obtained
from this study are as follows:
i) By optimizing the otation parameters at collector dosage
(diesel oil) of 3500 g/t, frother dosage (pine oil) of 2500 g/tonne,
pulp density of 15%, wash water rate of 0.17 cm/s and downcomer immersion depth of 50 cm, UC has been successfully
recovered from waste lter powder.
ii) For this study, rst-order kinetic otation model described as
R =R [1 exp (kt)]) was tested for its applicability to batch
otation time-recovery proles for the unburned carbon.
Modelling indicated that the rst-order kinetic otation model
gave a very good t for experimental data.
iii) The main advantages of the Jameson otation cell are the rapid
collection of particles in the downcomer, and the relatively small
Table 7
Parameters obtained from model (R = R[1 exp(kt)]) t to data sets of downcomer
immersion depth

Fig. 12. The effect of downcomer immersion depth on the combustible recovery (%) and
ash reduction (%).

Downcomer
immersion
Depth (cm)

20

30

40

50

60

R
k
R2

0.915
0.593
0.997

0.906
0.743
0.998

0.927
0.619
0.998

0.947
0.774
0.998

0.894
0.630
0.998

246

M. Uurum / Powder Technology 191 (2009) 240246

footprint. The cell is simple and easy to operate, and draws air
from the atmosphere, eliminating the need for compressors or
blowers. In conclusion, Jameson otation seems a viable and
effective separation process for recovering the unburned carbon
(UC) from waste lter powder.
Acknowledgements
The author thanks the Kaksan Lime Company and ukurova
University, Mining Engineering Department, for their great help on
laboratory support.
References
[1] F.F. Aplan, The historical development of coal otation in the United States, in: B.K.
arekh, J.D. Miller (Eds.), Advances in Flotation Technology, SME, 1999, pp. 269287.
[2] J. Rubio, M.L. Souza, R.W. Smith, Overview of otation as a wastewater treatment
technique, Minerals Engineering 15 (2002) 139155.
[3] C. Xinghua, Study and test on a new otation device: LM Flotaton Cell, Minerals
Engineering 11 (5) (1998) 457462.
[4] G.J. Jameson, Hydrophobicity and oc density in induced air otation for water
treatment, Colloids Surfaces 151 (1999) 269.
[5] Y. Yan, G.J. Jameson, Application of the Jameson cell technology for algae and
phosphorus removal from maturation ponds, International Journal of Mineral
Processing 73 (2004) 2328.
[6] G. Harbort, S.D. Bono, D. Carr, V. Lawson, Jameson cell fundamentalsa revised
perspective, Minerals Engineering 16 (2003) 10911101.

[7] J. Cowburn, G. Harbort, E. Manlapig, Z. Pokrajcic, Improving the recovery of coarse

coal particles in Jameson cell, Minerals Engineering 19 (2006) 609618.
[8] F. Herninz, M. Calero, Froth otation: kinetic models based on chemical analogy,
Chemical Engineering and Processing 40 (2001) 269275.
[9] M. Xu, Modfed otation rate constant and selectivity index, Minerals Engineering
11 (3) (1998) 271278.
[10] G.E. Agar, J. Chia, L. Requis, Flotation rate measurements to optimize an operating
circuits, Minerals Engineering 11 (4) (1998) 347360.
[11] E.C. Dowling, R.R. Klimpel, F.F. Aplan, Model discrimination in the otation of a
porphyry copper ore, Minerals & Metallurgical Processing 2 (1985) 87101.
[12] X.M. Yuan, B.I. Palsson, K.S.E. Forssberg, Statistical interpretation of otation
kinetics for a complex sulphide ore, Minerals Engineering 9 (4) (1996) 429442.
[13] A.J. Lynch, N.W. Johnson, E.V. Manlapig, C.G. Thorne, Mineral and Coal Flotation
Circuits, Elsevier, Amsterdam, 1981, pp. 5796, Chapter 3.
[14] E.T. Woodburn, Mathematical modelling of otation processes, Minerals Science
and Engineering 2 (2) (1970) 317.
[15] J.F. Oliveira, S.M. Saraiva, J.S. Pimenta, A.P.A. Oliveira, Kinetics of pyrochlore
otation from Araxa mineral deposits, Minerals Engineering 14 (1) (2001) 99105.
[16] J.H. Schulman, J. Leja, Molecular interactions at the solidliquid interface with special reference to otation and solid particle stabilized emulsions, KolloidZeitschrift
136314 (1954) 107120.
[17] J. Leja, Mechanisms of collector adsorption and dynamic attachment of particles to
air bubbles as derived from surface chemical studies, Transactions of the Institution
of Mining and Metallurgy 669 (1956) 425437.
[18] P. Ireland, R. Cunningham, G.J. Jameson, The behaviour of wash water injected into
a froth, International Journal of Mineral Processing 84 (2007) 99107.
[19] S.J. Neethling, J.J. Cilliers, Simulation of the effect of froth washing on otation
performance, Chemical Engineering Science 56 (2001) 63036311.
[20] www.jamesoncell.com.