American Journal of Chemical Engineering

2015; 3(2-2): 6-12

Published online January 27, 2015 (http://www.sciencepublishinggroup.com/j/ajche)
doi: 10.11648/j.ajche.s.2015030202.12
ISSN: 2330-8605 (Print); ISSN: 2330-8613 (Online)

Study of relationship between flotation rate and bubble

surface area flux using bubble-particle attachment
efficiency
Behzad Shahbazi
Mining Engineering Department, Science and Research Branch, Islamic Azad University, Tehran, Iran

Email address:
[email protected]

To cite this article:

Behzad Shahbazi. Study of Relationship between Flotation Rate and Bubble Surface Area Flux using Bubble-Particle Attachment Efficiency.
American Journal of Chemical Engineering. Special Issue: Flotation Technology. Vol. 3, No. 2-2, 2015, pp. 6-12.
doi: 10.11648/j.ajche.s.2015030202.12

Abstract: Understanding the attachment micro process is a fundamental step toward predicting the rate constant of flotation
kinetics. In this research, the effect of bubble-particle attachment efficiency on k-Sb relationship was investigated under Yoon,
Stokes and Potential conditions. Maximum Stokes attachment efficiency obtained was 55.9% with particle size of -37 µm, Sbof
34.2 1/s and flotation rate of 1.65 1/min. Stokes attachment efficiency was less than Yoon efficiency and it seems to be a
suitable equation for predicting attachment efficiency. Furthermore, three different models were obtained for estimating
attachment efficiency usingk-Sb relationship.

Keywords: Flotation, Kinetics, Bubble, Attachment

active particles. These factors can have a considerable impact

1. Introduction on grades and recoveries, depending on the dominant effects
The bubble–particle collision micro process, attachment in operation [9, 10]. Efficiency of the bubble–particle
and detachment strongly depend on the chemistry and stability depends on the particle size, particle hydrophobicity
physical chemistry of the surface of solid particle and bubble. and external detaching forces. Even in the flotation of fine
At present these two micro processes are the least quantified, particles, the bubble–particle detachment can significantly
because there are many complex phenomena involved which influence the kinetics of flotation taking place in mechanical
are not well understood yet [1]. cells by intensive turbulent agitation [11]. For these fine
Froth flotation is widely used for separating different particles, the bubble-particle detachment is often neglected.
minerals from each other. However, its influence is limited to So far the effect of hydrodynamic parameters on flotation
a relatively narrow particle size range of 10–100 µm [2, 3]. response of coarse and fine particles has been widely
Although the effect of particle size on flotation performance investigated [12-18].
has been widely studied to date [2, 4-6], and many important The surface area flux work has been published in a series
of papers [19-24] and Gorain et al (1997) investigated the
physico-chemical factors related to particle size have been
effect of gas dispersion properties on the flotation rate
identified, the net effect of these factors are very difficult to
constant in plant and pilot scale mechanical cells over a range
predict. For example, in bubble-particle interaction, particle
of operating conditions for four impeller types. They found
size is known to play a critical role in the probability of that the rate constant was not readily related to the bubble
particles colliding with bubbles, attachment of particles to size, gas holdup or superficial gas rate individually, but it was
bubbles after collision, as well as remaining attached in the related to bubble surface area flux ( S b ). For example, for
pulp [7, 8].
Fine particles typically show slow recovery rates, owing to shallow froths the relationship was linear [23]. These authors
decreased particle-bubble collisions, and are prone to found [23, 24] that S b (1/s) was strongly related with k
entrainment. Moreover, very small particles tend to have (1/min) and that the relationship was linear, as represented by
large specific areas, which can lead to excessive adsorption following equation:
of reagents, and other effects associated with chemically
 American Journal of Chemical Engineering 2015; 3(2-2): 6-12 7

k = P × Sb (1) flotation cell. Impeller diameter was 0.07 meter for a cell
with square section of 0.13 and height of 0.12 meter. Impeller
Where P summarized the operational and chemical factors. rotating speed was 900, 1000, 1100 and 1200 rpm and air
This k-Sb relationship appears to be independent of both the flow rate was 60, 120, 180 and 240 l/hr. Quartz particles
type and size of mechanical cell and, consequently, has (specific gravity=2.65 g cm 3 ) of four size classes contain -
important implications with regard to the optimization, scale-
37, -53+37, -75+53 and -106+75 µm were used for flotation
up and design of mechanical flotation cells.
experiments (all size fractions were combined together
In this research, the effect of bubble surface area flux on before flotation). The frother was MIBC (methyl iso-butyl
bubble-particle attachment efficiency was investigated and carbinol) with concentration of 22.4 ppm and CCC200
bubble-particle attachment efficiency was calculated under (critical coalescence concentration). The flotation
Yoon, Stokes and Potential conditions. Furthermore, three
experiments were carried out by dodecylamine collector (50
models were obtained for estimating bubble-particle
g/ton) at the natural pH of 8.5 using local tap water. When
attachment efficiency with k-Sb relationship in different
Sbvalues were set, all the size fractions floated together under
conditions. According to author's knowledge, these models those exact conditions (Table 1).
are the first equations for predicting bubble-particle
The air flow rate and impeller speed were set and float
attachment efficiency with k-Sb relationship.
product collected at time intervals 1, 2, 3 and 5 minutes. The
recovery R determined as a function of time t and flotation
2. Materials and Methods rate constant was calculated. The batch wise flotation of
mineral particles may be described by a first order rate
Flotation tests were carried out in a mechanical laboratory equation where the rate of removal of particles is given by:
Table 1. Flotation tests conditions
Gas aflow Rate
60 120 180 240
(l/hr)
Impeller Speed
900 1000 1100 1200 900 1000 1100 1200 900 1000 1100 1200 900 1000 1100 1200
(rpm)
d b ( µm ) 829 752 686 628 826 749 683 625 824 747 681 624 823 746 680 623
ε g (%) 2.86 3.75 4.75 5.87 3.62 4.71 5.93 7.30 4.08 5.29 6.65 8.17 4.48 5.79 7.27 8.92

J g (cm/s) 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.2 0.3 0.3 0.3 0.3 0.4 0.4 0.4 0.4
S b (1/s), Eq. (14) 10.21 10.88 11.53 12.16 16.77 17.94 19.08 20.19 22.69 24.31 25.89 27.43 28.23 30.28 32.27 34.2
Vb (m/s) 0.37 0.31 0.25 0.21 0.37 0.30 0.25 0.21 0.37 0.30 0.25 0.21 0.37 0.30 0.25 0.21
Re b 307 229 174 133 303 226 171 131 301 224 170 130 299 223 169 130
ε (W/kg), Energy 2.15 2.95 4.18 5.98 2.15 2.95 4.18 5.98 2.15 2.95 4.18 5.98 2.15 2.95 4.18 5.98
dissipation

dC using LAUDA tensiometer (TE3) after flotation tests. The

= − kC (2)
dt LAUDA tensiometer simplifies the characterization of
wetting behavior of the whole surface. The contact angle was
Where C is the particle concentration in mass per unit measured by means of the precise adjustment of the
volume and k is a ‘rate constant’. The flotation rate constant immersion/ receding rate and micrometer-accurate
was calculated assuming the first order rate equation for a measurement of the immersion dept. Contact angle of the
batch cell, R = R ∞ (1 − exp(−kt )) , and plotting ln(1 − R R ∞ ) quartz samples after flotation was obtained 89.7o using this
versus t. method.
The bubble size distribution was measured in a device
similar to the McGill bubble viewer. It consisted of a
sampling tube attached to a viewing chamber with a window
3. Theory
inclined 15o from vertical. The closed assembly was filled The flotation rate constant is proportional to the collection
with water of a similar nature to that in the flotation cell (to efficiency [26], this equation can be seen as the embryonic
limit changes in bubble environment during sampling) and form of the below equation:
the tube was immersed in the desired location below the froth.
Bubbles rose into the viewing chamber and were imaged by a E col = E c E a (3)
digital camera as they slid up the inclined window, which
was illuminated from behind. In this research, at first, frother where Ec is ‘the ratio of the number of the particles
was added to the water of the cell and then the viewing encountering a bubble per unit time to the number of the
particles approaching the bubble at a great distance in a flow
chamber was filled with water of the cell to prevent bubble
tube with a cross-sectional area equal to the projected area of
coalescence [25].
the bubble’ [27]. Eais the attachment efficiency. When the
The Washburn method was used to measure contact angle
detachment of the attached particles from the bubble surface
on powder. Contact angle of quartz particles was measured is considered, Eq. (3) becomes
8 Behzad Shahbazi: Study of Relationship between Flotation Rate and Bubble Surface Area Flux
using Bubble-Particle Attachment Efficiency

E col = E c E a (1 − E d ) (4) 0.6, and parameter A is inversely proportional to the particle

contact angle θ. Based on these findings, the following
Where Ed is the particle-bubble detachment efficiency [28]. equation was used [34]:
The availability of stream function also makes it possible to
predict Ea. For predicting attachment probability under Yoon ti =
75
d p0.6 (13)
conditions, can use bellow equation [29]: θ

 ( )
 − 45 + 8 Re 0b.72 vt i
E a = sin 2 2 arctan exp


Where: tiis given in second, θ in degrees and dpin meter.
 15d (d d + 1)  (5) The probability of adhesion can now be calculated for given
  b b p 
values of bubble size, particle size and contact angle.
Where: dp is the diameter of particle, db is the diameter of
bubble,ti is induction time, V is bubble raise velocity. There 4. Results and Discussions
is another generalized equation for calculation Ea according
to [1]: 4.1. Bubble Diameter
 2UAt i  Bubble diameter was measured. According to Table 1,
E a = sec h 2   (6)
d +d  bubble diameter was 623 to 829 micron. Maximum bubble
 p b 
diameter obtained was 829 micron when impeller speed, air
Where: dp is the diameter of particle, db is the diameter of flow rate and bubble surface area flux were 900 rpm, 60 l/hr
bubble, ti is induction time, U is particle settling velocity and and 10.21 1/s, respectively. Minimum bubble diameter
A is a dimensionless parameter under Stokes flow (A1) or obtained was 623 micron with impeller speed of 1200 rpm,
Potential flow (A2) according below equations: air flow rate of 240 l/hr and bubble surface area flux of 34.2
−1 −3 1/s. Flotation rate increased with decreasing bubble diameter
V 3 d  1 d 
A1 = +1− 1 + p  − 1 + p  (7) due to increasing bubble-particle collision efficiency.
U 4  db  4 db 

−3
4.2. Bubble Surface Area Flux
V 1 dp 
A2 = + 1 + 1 +  (8)
U 2 d b  CalculatingSbin flotation cells is difficult and expensive
due to difficulty in measuring bubble diameter. In this
Sincedb, V, U and ti are four necessary parameters for research, an empirical model was obtained for estimating Sb
calculating attachment efficiency, they were obtained using in different operating conditions (under conditions of Table 1
following equations. The mean bubble diameter adopted was with impeller diameter of 0.07 and 0.09 meter and pulp
the Sauter diameter, calculated by below equation [30]: density up to 20%). Impeller peripheral speed (Ns) instead of
impeller speed (N) was used to determine the contemporary
d b or d 32 =
∑n d i
3
i
(9) influence of impeller speed and impeller diameter in models.
∑n d i
2
i Also, superficial gas velocity (Jg) instead of air flow rate per
unit cell cross-sectional area (Q/A) was used for modeling.Pd
Where n is the number of bubbles, and d is the bubble is pulp density.
diameter. If the surface of a drop or bubble is immobile for
any reason, the floating velocity is the same as that of a solid Sb = 25 N s0.699 J 0g.805 Pd0.005 R2=0.92 (14)
sphere and the bubble raise velocity can be described by
Stokes' equation [31]: 4.3. k-Sb Relationship

d 2b According to Fig. 1, flotation rate increased with

V= g (10)
18ν increasing bubble surface area flux and decreasing particle
size. When particle size, impeller speed and bubble surface
Where ν is the kinematic viscosity. Also, Particle settling area flux were -37 µm, 1100 rpm and 32.27 1/s, maximum
velocity can be described by below equation [32]: flotation rate was obtained 1.89 1/min. coarser quartz
3g (ρ p − ρ )d p particles showed a pronounced lower flotation rate than finer
U= (11) ones. It seems that bigger particles demand much more
ρ
turbulence to become suspended and collide with air bubbles
Where g, ρ p and ρ are gravitational acceleration, particle than smaller ones.
A deviation is observed from linear k-Sbrelationship in Fig.
density and fluid density respectively. The induction time is a 1. This could be due to the inapplicability of the hypothesis
function of the particle size and contact angle which can be by Gorain et al. (1997) of linear k-Sbrelationship. So, a
determined by experiment and correlated in the form of [33]: k = P × Sαb relationship was assumed for flotation response of
ti = Ad pB (12) quartz particles and a model was fitted to actual values for
obtaining influence of particle size on k-Sbrelationship. Eq.
Where parameters A and B are independent of particle (15) explains the effect of particle size in P and αconstants.
size.It was found that parameter B is constant with a value of
 American Journal of Chemical Engineering 2015; 3(2-2): 6-12 9

−0 .059
exp( )
R2=0.71 (15) efficiency obtained was 55.9% with particle size of -37 µm,
k = 0.057 d p−0.075 × S b
dp

Sb of 34.2 1/s and flotation rate of 1.65 1/min. Stokes

attachment efficiency was less than Yoon efficiency.
-37 Micron
k (1/min)

Actual
Predict

Sb (1/s)

-53+37 Micron
k (1/min)

Actual

Sb (1/s)

-75+53 Micron
k (1/min)

Actual

Sb (1/s)

-106+75 Micron
k (1/min)

Actual Fig. 2. Attachment efficiency under Yoon, Stokes and Potential flow
conditions (particle density of 2.65 g / cm 3 and contact angle of 89.7 0 )
Sb (1/s)
According to Fig. 2c, Potential attachment efficiency was
Fig. 1. k-Sbrelationship for different particle sizes calculated for different particle sizes using Eqs. (6,8).
Maximum attachment efficiency obtained was 15.5% with
4.4. Calculating Attachment Efficiency particle size of -37 µm, Sb of 34.2 1/s and flotation rate of
1.65 1/min. Attachment efficiency calculated by Potential
According to Fig. 2a, Yoon attachment efficiency was flow conditions was very lower than both Yoon and Stokes
calculated for different particle sizes using attachment efficiencies. So, Potential efficiency was not a
Eq. (5). In this research, with decreasing particle size, suitable equation for predicting attachment efficiency under
attachment efficiency increased to maximum magnitude of conditions of this research (laboratory mechanical flotation
94.8%. With increasing Sb, flotation rate increased but cells). Stokes attachment efficiency was between Yoon and
attachment efficiency decreased. So, increasing flotation rate Potential attachment efficiency.
can be attributed to increasing collision efficiency.
According to Fig. 2b, Stokes attachment efficiency was 4.5. Primary Predicting of Bubble- Particle Attachment
calculated for different particle sizes using Efficiency Using k-Sb Relationship
Eqs. (6,7). Minimum attachment efficiency obtained was 1.9% Three different models were obtained for estimating
with particle size of -106+75 µm, Sb of 10.21 1/sec and attachment efficiency using k-Sbrelationship. All of four
flotation rate of 0.41 1/min and maximum attachment particle size classes (-37, -53+37, -75+53 and -106+75 µm)
10 Behzad Shahbazi: Study of Relationship between Flotation Rate and Bubble Surface Area Flux
using Bubble-Particle Attachment Efficiency

were used for this purpose. Different forms of multiple equations can be used only for small particles (<53µm).
regression models (exponential, linear, polynomial and
k −0.137
power) were examined by comparing their statistical Ea−Y = 7.76 ×10−4 t i−1.632 ( ) (16)
significance using coefficient of multiple determinations Sb
(R2)and following models were obtained for
k −0.145
10.21 < S b < 34.2 1 / s and 0.32 < k < 1.89 1 / min . So, following Ea−S = 10.59 × 10−4 t i−1.488 ( ) (17)
Sb
equations explain influence of Yoon ( E a − Y ), Stokes ( E a −S )
and Potential ( E a −P ) attachment efficiency onk-Sbrelationship k −0.026
Ea−P = 0.013×10−4 t i−2.336 ( ) (18)
for different particle size fractions. N is data numbers of the Sb
each model. The effect of k-Sbrelationship on attachment
efficiency has been illustrated in Fig. 3.

Fig. 3. Predicting bubble-particle attachment using k-Sb relationship

(particle density of 2.65 g / cm 3 and particle size of <106 micro meters)

Fig. 4. Compares bubble-particle attachment calculated

using different equations. According to Figs. (4a, 4b), for
particle size of -37 and -53+37 µm, below equations can be
useful for estimating Yoon, Stokes and Potential attachment
efficiency and equivalent plots are compatible together.
According to Figs. (4c, 4d), for particle size of -75+53 and
-106+75 µm, following equations are not suitable for
predicting attachment efficiency because the equivalent plots
are not compatible together. It can be attributed to high
efficiency of bubble-particle collision and detachment.
Bubble-particle collision and detachment efficiency of coarse Fig. 4. Comparing bubble-particle attachment calculated using different
particles are effective on the flotation rate too. So, below equations (particle density of 2.65 g / cm 3 and contact angle of 89.7o)
 American Journal of Chemical Engineering 2015; 3(2-2): 6-12 11

5. Conclusions Metallurgical Engineering, Tech Publ, 414 (1931) 3-23.

In this research, the effect ofk-Sbrelationship on bubble- [5] R.M. Anthony, D.F. Kelsall, W.J. Trahar, The effect of particle
size on the activation and flotation of sphalerite, Proceedings
particle attachment efficiency was investigated under Yoon, of the Australian Institute of Mining and Metallurgy, 254
Stokes and Potential conditions and following conclusions (1975) 47-58.
were illustrated:
Attachment efficiency calculated under Yoon conditions [6] W.J. Trahar, A rational interpretation of the role of particle size
in flotation, Int J Miner Process, 8 (1981) 289-327.
was more than Stokes and Potential efficiency. Maximum
Stokes attachment efficiency obtained was 55.9% with [7] H.R. Spedden, W.S. Hannan, Attachment of Mineral Particles
particle size of -37 µm, Sb of 34.2 1/s and flotation rate of to Air Bubbles in Flotation, Min Tech, 12 (1984) 2354.
1.65 1/min. Maximum Potential attachment efficiency
[8] P.F. Whelan, D.S. Brown, Particle-Bubble Attachment in Froth
obtained was 15.5% with particle size of -37 µm, Sb of 34.2 Flotation, Transactions of the Institute of Mining and
1/s and flotation rate of 1.65 1/min. Attachment efficiency Metallurgy, 65 (1956) 181-192.
calculated by Potential flow conditions was very lower than
both Yoon and Stokes attachment efficiency. Also, three [9] T.M. Morris, Measurement of Equilibrium Forces between an
Air Bubble and an Attached Solid in Water, Trans. AIME 187
different models were obtained for estimating attachment (1950) 91-95.
efficiency using k-Sb relationship.
[10] H.J. Schulze, New Theoretical and experimental
investigations on stability of bubble-particle aggregates in
Nomenclature flotation: A theory on the upper particle size of floatability, Int
J Miner Process, 4 (1977) 241-259.
A Unit cell cross N Impeller speed
sectional area [11] D.A. Deglon, F. Sawyerr, C.T. O’Connor, A model to relate
C Ns Impeller peripheral the flotation rate constant and the bubble surface area flux in
Concentration mechanical flotation cells, Miner Eng, 12 (1999) 599-608.
speed
CCC Critical Pd [12] Shahbazi, B., Rezai, B. and Koleini, S.M. Javad, Noaparast M.
coalescence Pulp density (2013) Iran. J. Chem. Chem. Eng., 32: 109-118.
concentration
db Q [13] Shahbazi, B., Rezai, B. and Koleini, S.M. Javad, Noaparast M.
Bubble diameter Air flow rate (2014) Geosciences ScientificQuarterlyJournal, in press.
dp Re b Bubble Reynolds
Particle size [14] Shahbazi, B., Rezai, B. (2014) Journal of Dispersion Science
number
Ea Attachment R and Technology, in press.
Recovery
efficiency [15] S. Chehreh Chelgani, B. Shahbazi, B. Rezai, Estimation of
g Acceleration due to R∞ froth flotation recovery and collision probability based on
Infinite recovery operational parameters using an artificial neural network, Int J
gravity
Superficial gas Sb Bubble surface area Min Met Mater, 17 (2010) 526-534.
Jg
velocity flux [16] B. Shahbazi, B. Rezai, S.M. Javad Koleini, Bubble–particle
k Flotation rate ti collision and attachment probability on fine particles flotation,
Induction time
constant ChemEng Process, 49 (2010) 622–627.
Ec Collision efficiency ν Kinematic viscosity [17] B. Shahbazi, B. Rezai, S.M. Javad Koleini, The effect of
E col Collection ε hydrodynamic parameters on probability of bubble-particle
Energy dissipation
efficiency collision and attachment, Miner Eng, 22 (2009) 57–63.
Ed Detachment θ
Contact angle [18] B. Shahbazi, B. Rezai, S.M. Javad Koleini, Effect of
efficiency Dimensionless Hydrodynamic Parameters on Coarse Particles
Flotation, Asian J Chem, 3 (2008) 2180-2188.

[19] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, Studies on

References impeller type, impeller speed andair flow rate in an industrial
[1] A.V. Nguyen, J. Ralston, H.J. Schulze, On modeling of scale flotation cell. Part 1: Effect on bubble size
bubble–particle attachment probability in flotation, Miner Eng, distribution,Miner Eng, 8 (1995a) 615-635.
53 (1998) 225-249.
[20] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, Studies on
[2] W.J. Trahar, The selective flotation of galena from sphalerite impeller type, impeller speed andair flow rate in an industrial
with special reference to the effects of particle size, Int J scale flotation cell. Part 2: Effect on gas holdup, Miner Eng, 8
Miner Process, 3 (1976) 151-166. (1995b) 1557-1570.

[3] B.A. Wills, Mineral Processing Technology, 4th edn.Pergamon, [21] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, Studies on
New York, (1998). impeller type, impeller speed andair flow rate in an industrial
scale flotation cell. Part 3: Effect on superficial gas
[4] A.M. Gaudin, J.O. Groh, H.B. Henderson, Effect of particle velocity,Miner Eng, 9 (1996a) 639-654.
size on flotation, American Institute of Mining and
12 Behzad Shahbazi: Study of Relationship between Flotation Rate and Bubble Surface Area Flux
using Bubble-Particle Attachment Efficiency

[22] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, The effect of gas [28] Z. Dai, D. Fornasiero, J. Ralston, Particle-bubble collision
dispersion properties on thekinetics of flotation, Column 96, models-a review, Adv Colloid Interfac, 85 (2000) 231-256.
Proceedings of the 35th Annual Conference of
Metallurgists,CIM, Montreal, Canada, (1996b) 299-313. [29] R.H. Yoon, The role of hydrodynamic and surface forces in
bubble-particle interaction,Int J Miner Process, 58 (2000) 129-
[23] B.K. Gorain, J.P. Franzidis, E.V. Manlapig, Studies on 143.
impeller type, impeller speed and airflow rate in an industrial
scale flotation cell. Part 4: Effect of bubble surface area flux [30] R.T. Rodrigues, J. Rubio, new basis for measuring the size
on flotationkinetics, Miner Eng, 10 (1997) 367-379. distribution of bubbles, Miner Eng, 16 (2003) 757-765.

[24] B.K. Gorain, N. Munn, J.P. Franzidis, E.V. Manlapig, Studies [31] S.S. Dukhin, R. Miller, G. Loglio, Physico-chemical
on impeller type,impeller speed and air flow rate in an hydrodynamics of rising bubble, Drops and Bubbles in
industrial scale flotation cell. Part 5: Validation of k- Interfacial Research D. Mobius and R. Miller (Editors),
Sbrelationship and effect of froth depth, Miner Eng, 11 (1998) Elsevier Science B.V. All rights reserved, 1998.
615-626.
[32] H.J. Schulze, Physico-Chemical Elementary Processes in
[25] E.H. Girgin, S. Do, C.O. Gomez, J.A. Finch, Bubble size as a Flotation-An Analysis from the Point of View of Colloid
function of impeller 339 speed in a self-aeration laboratory Science Including Processes Engineering Considerations, Dev.
flotation cell, Miner Eng, 19 (2006) 201–203. Int J Miner Process 4 (Fuerstenau, D.W., Advisory Ed.),
Elsevier, Amsterdam, 1984.
[26] G.J. Jameson, S. Nam, M.M. Young, Physical factors affecting
recovery rates in flotation, Min SciEng, 9 (1977) 103-118. [33] P.T.L. Koh, M.P. Schwarz, CFD modelling of bubble-particle
attachments in flotation cells, Miner Eng, 19 (2006) 619-626.
[27] H.J. Schulze, Hydrodynamics of bubble-mineral particle
collisions, Min Process Extractive Metall, 5 (1989) 43-76. [34] Z. Dai, D. Fornasiero, J. Ralston, Particle-bubble attachment
in mineral flotation, Adv Colloid Interfac, 217 (1999) 70-76.