Minerals Engineering 81 (2015) 109–115

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

Effect of frothers and dodecylamine on bubble size and gas holdup

in a downflow column
M.A. Corona-Arroyo a, A. López-Valdivieso a,⇑, J.S. Laskowski b, A. Encinas-Oropesa c
a
Area de Ingeniería de Minerales, Instituto de Metalurgia, Universidad Autónoma de San Luís Potosí, Av. Sierra Leona 550, San Luis Potosí, SLP 78210, Mexico
b
NB Keevil Institute of Mining Engineering, University of British Columbia, 2329 West Mall, Vancouver, BC V6T 1ZA, Canada
c
Instituto de Física, Universidad Autónoma de San Luis Potosí, Av. Salvador Nava 6, San Luis Potosí, SLP 78260, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: Bubble size and gas holdup were characterized in a two phase gas–water system in a laboratory down-
Received 27 March 2015 flow column. The effect of the cationic surfactant dodecyl amine (MW 185, HLB 10.7) and the frothers
Revised 23 June 2015 MIBC (MW 102, HLB 6.05) and polyglycol F507 (MW 425, HLB 8.63) on the bubble size and gas holdup
Accepted 28 July 2015
were investigated. In addition, the effect of blends of MIBC-dodecyl amine (DDA) and F507-DDA on these
Available online 3 August 2015
parameters was assessed. The bubble Critical Coalescence Concentration (CCC) followed the order
MIBC > DDA > F507. When blending the frothers with DDA at a concentration below its CCC, the frother
Keywords:
CCC decreased and bubbles of finer size were obtained below and above the frother CCC. Static surface
Frothers
Bubble size
tension measurements of aqueous solutions with frothers and DDA as well as with frothers-DDA blends
Amine collector show coadsorption of DDA at the air/aqueous solution interface. The surface tension of aqueous solutions
Bubble columns prepared with the blends decreased with the addition of DDA and varied linearly with the frother con-
Flotation centration within the concentration range studied. The gas holdup in the downflow column was deter-
mined by the bubble size and decreased with the bubble size. It is shown that frother-DDA blends
gave the lowest gas holdup in the downflow column. This work is relevant for the reverse flotation of
quartz from iron ores using amine collectors in cells with downflow systems.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction system wherein small bubbles are formed in a high-shear region

Bubble columns are widely used as gas–liquid contactor to
carry out various types of gas–liquid chemical reactions and mass
transfer operations such as stripping, adsorption and mineral flota-
tion. The efficiency of these columns is highly dependent on the
dispersion of the gas in the liquid (Mandal et al., 2005). In particu-
lar, mineral flotation is a material transport process where miner-
als are selectively separated using air bubbles (Wills and
Napier-Munn, 2006). This process is largely determined by the
bubble size and the gas holdup in the flotation cell. Since the con-
ception of the process, various types of flotation cells have been
developed to improve the efficiency of particle collection. For the
sake of discussion we classify them as follows: (a) mechanical
flotation cells, (b) long counter-current flotation columns, (c) short
co-current columns and (d) miscellaneous cells. The Jameson cell
(short co-current column) has been found to be very effective for
both the dispersion of gas in water and bubble-mineral contacting
(Evans et al., 1995). A main component in this cell is a downflow
⇑ Corresponding author.
E-mail address: alopez@uaslp.mx (A. López-Valdivieso).
surrounding a plunging jet, leading to a high interfacial area per
unit volume of gas.
Extensive research has been carried out on the effect of frothers
on bubble size and gas holdup in flotation cells (Cho and
Laskowski, 2002; Laskowski, 2003; Azgomi et al., 2007; Elmahdy
and Finch, 2013; Leiva et al., 2010). This research has been mostly
focused on upflow systems such as mechanical flotation cells with
a rotor–stator mechanism to generate the bubbles and flotation
column cells where bubbles are produced by spargers or perforated
plates. The effect of frothers on bubble size and gas holdup in flota-
tion cells with a downflow system like the Jameson cell has been
scarcely studied. In this system, air and pulp are brought together
into a vertical pipe (downcomer), into which the pulp is injected
through a nozzle creating a high speed jet. Air is dispersed into
the pulp jet, which enters the downcomer in such conditions that
it is broken up into microbubbles. These bubbles are carried down-
ward by the pulp, co-currently with mineral particles, but the pulp
velocity is sufficiently slow that allows the bubbles to rise against
the flow, creating a mixture of high gas content in the downcomer.
In the mixing region, particle–bubble collision is very favorable
and rapid collection of hydrophobic particles occurs. The bubbly

http://dx.doi.org/10.1016/j.mineng.2015.07.023
0892-6875/Ó 2015 Elsevier Ltd. All rights reserved.
110 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115
mixture discharges into a large diameter chamber (riser) where
bubbles with attached hydrophobic particles separate from the
pulp. Recently Araya et al. (2014) characterized the effect of
MIBC on the gas dispersion in a two phase gas–water system using
a Jameson pilot unit. They found a strong correlation between the
bubble size and the air-to-liquid ratio, also between the bulk den-
sity of the aerated solution in the riser and the vacuum inside the
downcomer.
Bubble size and gas holdup in a flotation cell are largely affected
by surfactants, which are adsorbed at the gas/aqueous solution
interface (Leja, 1982). In flotation, surfactants called frothers are
used to disperse injected air into fine bubbles and to stabilize the
flotation froth. The frothers are amphiphatic molecules containing
non-polar and polar parts. The non-polar part is a hydrocarbon
chain and the polar part can be only the hydroxyl (OH) group as
in aliphatic alcohol type frothers or OH group bonded to alkoxy
(O–CnHn+1) groups as in polyglycol type frothers. Adsorption of
frothers at the gas/aqueous solution interface is such that the polar
group is oriented toward the aqueous solution phase and the
non-polar group towards the air phase. Frothers dramatically affect
the bubble size and prevent bubble coalescence once their concen-
tration is above of what is known as critical coalescence concentra-
tion, CCC (Cho and Laskowski, 2002; Laskowski, 2003; Grau et al.,
2005). Frothers also have a direct role in the breakup of the air
stream through a mechanism still obscure (Grau and Laskowski,
2006; Finch et al., 2008). Finch et al. (2008) have proposed that
the breakup may arise from local stress or tearing action around
the frother molecule adsorbed at the gas/aqueous solution
interface.
The effect of frother blends on foaming, bubble stability and
bubble size has been scarcely studied. Laskowski et al. (2003)
reported that blends prepared with a very small addition of strong
frothers (polyglycol) to a selective frother like MIBC behave simi-
larly to the powerful power. There is a decrease of the bubble size
below and above the CCC of MIBC. The same behavior was found by
Tan et al. (2005). Elmahdy and Finch (2013) reported that this
decrease in bubble size occurs only below the CCC of the MIBC.
Above the CCC of the MIBC, they reported an increase in the bubble
size when the powerful frother is added at concentrations below
its CCC.
This work aimed at studying the effect of the cationic surfactant
dodecyl amine [DDA, CH3(CH2)11NH2] and frothers of two distinct
hydrophile-lipophile balance (HLB), molecular structure and
molecular weight (MW) on the bubble size and gas holdup in a
downflow system. The effect of blends of frother with DDA was
also investigated. The frothers studied were those widely used in
ore flotation, namely the aliphatic alcohol type frother MIBC
[(CH3)2CH2COHCH3], and a polyglycol-type frother
[CnH2n1(OC3H6)mOH]. This work is relevant to the reverse flotation
of quartz (SiO2) with amines in concentration of iron ores in flota-
tion cells with downflow systems like the Jameson Cell.
2. Experimental
The MIBC used in this work was from Cytec-Mexico. This com-
pany also supplied the polyglycol type frother, Oreprep F507.
Analytical grade DDA from Aldrich was used. The MW of DDA,
MIBC and F507 are 185, 102 and 425, respectively. The HLB of
MIBC and F-507 are 6.05 and 8.625, respectively. Using a value of
9.4 for N (tertiary amine) group and Davies scale (Davies, 1957),
the HLB value for dodecylamine is calculated as 10.7. However,
the HLB group number of a charged group is not strictly constant
and also depends on the electrical charge. According to Ho
(1998) the group number for ammonium cation is 20.0. The bubble
size measurements in this work were carried out in water at pH 7;
at this pH amine is ionized. Thus, assuming the HLB group number
for this group to be 20 the calculated HLB value for dodecylamine
is 21.3. This HLB value holds as long as the amine is ionized, which
for DDA is true below pH 10. Nesset et al. (2012) showed that CCC
values nicely correlate when plotted versus HLB/MW. These values
are: 0.06 for MIBC, 0.02 for F-507 and 0.115 for DDA.
Fig. 1 shows a schematic diagram of the experimental set-up
used in this work. The main components are: (1) a riser of constant
water level, (2) a vertically positioned plexiglass column (down-
comer) with its base extending 25 cm below the surface of the con-
stant water level riser, (3) a storage tank for the underflow and
overflow water of the riser, (4) a persitaltic pump to feed water
to the downcomer, (5) a sampling pipe for collection of the bubbly
water coming out of the downcomer, and (6) a viewing system for
collecting images of bubbles.
The downcomer consisted of a plexiglass tube of 13 mm inner
diameter and 600 mm length. The top of the downcomer was
sealed from the atmosphere allowing air and water to pass into
the column through inlets at the top. The downcomer discharged
into the riser of 100 mm inner diameter and 1L total volume. Tap
water with DDA, frother or frother-DDA blends were pumped into
the downcomer through a conical-type nozzle of 1 mm inner diam-
eter. The nozzle was fixed at the top of the downcomer so that the
water stream was directed along the vertical axis of the down-
comer and plunged into a constant water level riser. A calibrated
peristaltic pump was used to control the water flow fed to the
downcomer. A water superficial velocity (Ja) of 11 cm/s was used
in all tests. This water stream is fed to the downcomer as a high
velocity jet that passes through a headspace at the top of the col-
umn and entrains gas as it plunges into the liquid.
Air was fed at the top of the downcomer at a fixed flow rate of
100 cm3/min, which was measured by a calibrated rotameter. This
air flow corresponded to an air superficial velocity of (Jg) of
1.32 cm/s. In this work, Jg and Ja are referred to the volumetric
air flow rate and volumetric water flow rate per unit cross section
of the dowcomer, respectively (Majumder et al., 2005). Jg was
maintained constant in all the tests.
In the downcomer three zones have been distinguished
(Atkinson et al., 2003): (1) a free jet zone, which contains the water
jet and air, (2) a mixing zone, which is a region of high shear, recir-
culation and energy dissipation; it is in the zone that the gas
entrained by the water jet is broken into fine bubbles before being
transported downward by the water, and (3) a pipe flow zone,
which is below the mixing zone. Bubbles are formed in the mixing
zone from the entrained gas entering at the plunge point of the
Fig. 1. Schematic diagram of downflow column system for the measurement of gas
holdup and bubble size.
 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115
water jet. The high energy dissipation in this mixing zone results in
the generation of fine bubbles, which are carried downward by
water and leave the zone. However, large bubbles are also pro-
duced, which recirculate within the mixing zone where they are
eventually broken up and carried out by the water flow. Evans
et al. (1992) have reported that the residence time of the large bub-
bles inside the mixing zone is of the order of one second. They also
alluded that this time would allow a sufficient number of large
bubble breakages to occur so that an equilibrium maximum stable
bubble size is reached. Their measurements showed a maximum
stable bubble size in the range of 200–600 lm in the mixing zone.
To collect the bubbly mixture coming out of the downcomer, a
vertical pipe with 5 mm inner diameter was placed in the riser
next to the downcomer discharge. Through this pipe the mixture
was transported to a plexiglass viewing chamber with a diffused
backlighting to get bubbles shadows, which were photographed
by a Nikon 1 J1 high speed videocamera (60 frames per second).
Images of the bubble shadows including that of a little steel ball
of known size were transferred to a computer for data processing
using the ImageJ 1.48b free software. Bubbles were collected once
the system reached steady state. Under this condition, equilibrium
bubble size would have been reached. To have a representative
population of the small and large bubbles carried by the water
out of the downcomer, at least 4000 bubbles were processed.
Nevertheless, the bubble sampling tube used in this work may
not collect some large size bubbles. This could introduce some
uncertainties in the determination of the bubble size distribution,
in particular for the tests which were carried out using frother con-
centrations below the CCC. Once the bubble size was computed
with the software, the bubble size was stored automatically in an
Excel workbook for the determination of the bubble size distribu-
tion (BSD) and the mean Sauter diameter (d32). A stock aqueous
solution of DDA at a given concentration was prepared by dissolv-
ing the DDA in tap water at pH 3, which was adjusted using HCl.
Using NaOH, the pH of the water plus DDA and/or frothers injected
to the downcomer was adjusted to 7.
Gas holdup (e) in the downcomer was measured following
Tasßdemir et al. (2007). Once the system reached steady state (with
the proper set up of water feed rate and the air flow), it was shut
down and a rubber stopper was quickly inserted into the bottom
exit of the downcomer. As a result, the froth inside the downcomer
collapsed and the resulting height of the liquid was measured. The
gas holdup was calculated using the following expression:
L1
¼1
Lc  Lj
where Lc is the downcomer length, Lj is the jet length (distance from
the nozzle exit to liquid surface) and L1 is the liquid height after the
froth collapse. This procedure was repeated threefold and the e val-
ues reported below are the average values.
A ST 9002 NIMA Surface Tensionmeter with a platinum ring was
used to measure the surface tension of aqueous solutions of DDA,
MIBC and F507 as well as of mixtures of frothers and DDA. The sur-
face tension measurements were carried out at a constant pH of 7
and at 23 °C. Three measurements were performed for each frother
or DDA concentration and the average value is reported in this
work. All aqueous solutions were prepared with deionized water
with a surface tension of 72.6 mN/m.
3. Results and discussion
3.1. Effect of DDA, MIBC and F507 on bubble size and gas holdup
Fig. 2 shows the effect of the concentration of DDA, MIBC and
F507 on the bubble size (given by d32). It is seen that the bubble
111
size decreases rapidly with the frother concentration up to a given
concentration above which the bubble size is constant. This gen-
eral behavior of frothers has been amply explained by Laskowski
(2003). In the low frother concentration region bubble size decays
with concentration very rapidly and bubble coalescence occurs.
While in the frother concentration range over which the bubble
size is not concentration dependent bubbles are stable. This transi-
tion concentration for bubble coalescence-stability is known as
Critical Coalescence Concentration (CCC) (Cho and Laskowski,
2002). The arrows in Fig. 2 indicate the CCC for DDA, MIBC and
F507; CCC was determined as the intersection of the horizontal
asymptote to the bubble size-concentration curves at higher con-
centration with the slope line approximating the curve at lower
concentration (Cho and Laskowski, 2002). The CCC value of each
reagent is given in Table 1 and it follows the order
CCCMIBC > CCCDDA > CCCF507. Such trend agrees well with that
reported elsewhere in that the CCC decreases with the frother
MW (Cho and Laskowski, 2002; Laskowski et al., 2003; Tan et al.,
2013). For MIBC and F-507, our CCC values were found to be smal-
ler than those reported by Laskowski (2003) and Zhang et al.
(2012) employing upflow systems; i.e., mechanical flotation cells.
With a Jg of 0.5 cm/s, they found a CCC of 11.2 mg/L for MIBC and
6 mg/L for a polyglycol frother with a similar molecular weight
to F507. It is known that CCC is not unique for a given frother
and it is also affected by Jg (Finch et al. 2008). The CCC may also
be affected by the system of bubble generation in the flotation cell,
i.e., upflow or downfow system. Hence, these two variables can
account for the differences between the CCC values reported else-
where and our data.
As Fig. 2 shows in the bubble-stable concentration region, bub-
bles have the same Sauter mean diameter for DDA and MIBC. There
is a clear difference in the BSD, however, as shown in the insert in
Fig. 2. A finer BSD is produced by DDA in comparison to MIBC. As
shown below, adsorption of DDA at the aqueous solution/air inter-
face is higher than that of MIBC, due to its longer hydrocarbon
chain. Therefore, DDA is more surface active than MIBC. F507 pro-
duces finer bubbles than DDA and MIBC.
The surface tension of aqueous solutions as a function of MIBC,
F507 and DDA concentration was determined in order to calculate
the adsorption density of MIBC, DDA and F507 at the CCC. Fig. 3
shows the surface tension as a function of frother concentration.
As noted, the surface tension linearly decreased with frother
concentration, very slightly with MIBC and significatively with
DDA and F507. Then the surface activity follows the order
F507 > DDA > MIBC. From the line slope, the adsorption density
ð1Þ
(Ui) of the frother and DDA was calculated using the Gibbs
adsorption isotherm:
C i dc
Ci ¼  ð2Þ
RT dC i
where Ci is the concentration of the surfactant at the CCC, c is the
dc
surface tension and dC is the slope of the c versus C line. Table 2 pre-
i
sents the values of Ui and the area occupied per molecule of surfac-
tant at the air/aqueous solution interface at the CCC.
At the CCC, the adsorption density of MIBC is very low and that
for DDA and F-507 are one order of magnitude higher. Accordingly,
the area occupied by a molecule of MIBC is very large when com-
pared to that of a molecule of DDA and F507.
Grau et al. (2005) compared the surface tension isotherms
obtained for three frothers (DF-200, DF-250, DF-1012) with the
bubble size versus concentration curves determined for these
frothers. They found that while for DF-200 the surface tension at
its CCC (8.4  105 mol/L) was about that of pure water
(72 mN/m), the surface tension for DF-250 at its CCC
(3.3  105 mol/L) was about 66 mN/m and for DF-1012
112 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115

Fig. 2. Bubble Sauter mean diameter as a function of MIBC, DDA or F507 concentration in a downflow system for a Jg = 1.32 cm/s and Ja = 11 cm/s. The insert is the bubble size
distribution at a 10 mg/L concentration of the reagent.

Table 1 Table 2
CCC of MIBC, DDA and F507 in downflow system measured, Jg = 1.32 cm/s and MIBC, F507 and DDA adsorption density and area occupied by the surfactant molecule
Ja = 11 cm/s. at the CCC.

Reagent MW, g/mol CCC Reagent CCC Adsorption density

mg/L mmol/L mg/L mmol/L lmol/m2 nm2/molecule
MIBC 102 8.1 0.0794 MIBC 8.1 0.0794 0.053 31.10
DDA 185 6.2 0.0335 DDA 6.2 0.0335 0.560 2.96
F507 425 4.2 0.0097 F507 4.1 0.0097 0.530 3.22

Fig. 3. Surface tension of aqueous solutions of MIBC, F507 and DDA as a function of
the surfactant concentration at pH 7 and 23 °C.
5
(CCC = 1.5  10 mol/L) was about 61 mN/m. This led to the con-
clusion that the adsorption of DF-1012 at its CCC is much larger
than the adsorption of DF-250 at the corresponding CCC value for
DF-250; in turn, the adsorption of DF-250 was much higher than
the adsorption of DF-200 at their corresponding CCC values. Thus,
the adsorption for larger and more surface active molecules of
Fig. 4. Gas holdup as a function of MIBC, DDA or F507 concentration in a downflow
system for a Jg = 1.32 cm/s and Ja = 11 cm/s.
propylene glycols must be larger to prevent bubble coalescence. It
is worth noting that DDA adsorbs almost equally to F507 but its
CCC is about three times larger than that for F507 (Table 2). The
HLB/MW ratio for DDA is larger too. After Nesset et al. (2012) the
 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115 113

Fig. 5. Bubble Sauter mean diameter as a function of MIBC in the absence and presence of various DDA concentrations in a downflow system for a Jg = 1.32 cm/s and
Ja = 11 cm/s. The insert is the bubble size distribution at a concentration of MIBC of 10 mg/L MIBC in the absence and presence of various amounts of DDA.

Fig. 6. Bubble Sauter mean diameter as a function of F507 concentration in the absence and presence of various DDA concentrations in a downflow system for a Jg = 1.32 cm/s
and Ja = 11 cm/s. The insert is the bubble size distribution at a concentration of F507 of 7.5 mg/L F507 in the absence and presence of various amounts of DDA.

CCC for DDA would be around 60–85 mg/L, one magnitude higher
than the value found in this work. Then the empirical relationship
between CCC and HLB/MW ratio given by Nesset et al. (2012) does
not hold for surfactants of the DDA type. The CCC for F507 is lower
than that of DDA despite having identical adsorption densities. This
indicates that the hydrophilic groups in the non-polar chain of F507
impart a greater surface elasticity to the gas/water interface than
the DDA alkyl chain with only non-polar groups. Accordingly,
F507 stabilized the bubbles more efficiently than DDA.
Fig. 4 shows e as a function of the frothers and DDA concentra-
tion at Jg = 1.32 cm/s and Ja = 11 cm/s. The arrows in the figure indi-
cate the CCC. As noted, e decreases with the frother concentration
down to the CCC. Above CCC, e is about constant. Comparing Figs. 2
and 4, e decreases with the bubble size in the bubble coalescence
concentration region and reaches a constant value above CCC
where the bubble size is constant. The e values are lower
for F507, which gave smaller bubbles in comparison to MIBC and
DDA.
114 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115
Fig. 7. Gas holdup as a function of MIBC concentration in the absence and presence
of various DDA concentrations in a downflow system for a Jg = 1.32 cm/s and
Ja = 11 cm/s.
The value of e is determined by the bubble velocity, which in
turn depends on the bubble size. In the downcomer, the fluid flows
downward and bubbles of small size have a downward velocity
higher than that of large size bubbles ought to their lower buoyancy
force. This buoyancy force increases with the radius of the bubble to
the cube (Wills and Napier-Munn, 2006). Hence, e decreases with
concentration of DDA and frothers because the bubble size
decreases. This is opposite to upflow systems where the bubble size
decreases with the gas holdup (Tan et al., 2013; Elmahdy and Finch,
2013). In these systems, the uprising velocity of small bubbles is
lower than that of large bubbles (Leiva et al., 2010).
3.2. Effect of MIBC–DDA and F507-DDA blends on bubble size and gas
holdup
In the reverse flotation of quartz for iron ore concentration,
amines are used as collectors. Frothers are also added to the pulp
to have a stable froth (Batisteli and Peres, 2008). Hence, the effect
of frother-DDA blends on the bubble size and gas holdup in the
downcomer was assessed. Fig. 5 shows the d32 value as a function
of MIBC concentration in the absence and presence of various addi-
tions of DDA, namely 1, 2 and 4 mg/L. These concentrations are
below the CCC for DDA. As noted, the CCC for MIBC decreases with
Fig. 8. Gas holdup as a function of F507 concentration in the absence and presence
of various DDA concentrations in a downflow system for a Jg = 1.32 cm/s and
Ja = 11 cm/s.
Fig. 9. Surface tension of MIBC aqueous solutions as a function of MIBC concen-
tration in the absence and presence of various constant concentrations of DDA at pH
7 and 23 °C.
the addition of DDA. d32 is smaller below the CCC, in agreement
with Laskowski et al. (2003) and Elmahdy and Finch (2013).
Above the CCC, d32 was not apparently affected with 1 and
2 mg/L DDA, but it decreased with 4 mg/L DDA. The BSD was deter-
mined at 10 mg/L MIBC, which is above the CCC in the absence and
presence of DDA. It is noted that bubbles are of finer size with DDA
addition. This agrees with Laskowski et al. (2003). Elmahdy and
Finch (2013) reported that powerful frothers increased the bubble
size above the CCC of MIBC for upflow columns.
Fig. 6 presents the d32 as a function of the concentration of F507
in the absence and presence of 1 and 2 mg/L DDA. As in the case of
the MIBC–DDA mixture, the CCC for F507 decreased with DDA. d32
also decreased below and above the CCC. The BSD was determined
at a concentration of 10 mg/L F507 (above CCC) in the absence and
with the two DDA concentrations. This BSD is shown in the insert
of Fig. 6. As in the case of the MIBC–DDA blends bubbles are finer
above the CCC for F507. This is in agreement with the gas holdup
values attained for the blends and reported further below.
Since bubbles are finer for the frother-DDA blends below and
above the frother CCC, the gas hold up decreased for the
frother-DDA blends. This is observed in Figs. 7 and 8. The decrease
in gas holdup has been explained in the last paragraph of Section 3.1.
The surface tension of MIBC–DDA and F507-DDA blends are
shown in Figs. 9 and 10, respectively, as a function of MIBC and
F507 concentration in the absence and presence of the various con-
centrations of DDA used for assessing the effect of the frother-DDA
blends on the bubble size and gas holdup in the downflow column.
In the figures, it is noted that the surface tension decreased with
DDA addition and has a linear behavior with the frother concentra-
tion. This decrease in surface tension is an indication of coadsorp-
tion of DDA at the air/aqueous solution interface causing a
synergetic effect on the bubble size. Tan et al. (2005) have sug-
gested that blends of MIBC with strong frothers gave a better foam-
ing performance. This resulted from a greater surface elasticity
produced by a closed packed molecular cohesive films at the
air/aqueous solution interface. Matsubara et al. (2000) have
reported that there is a strong interaction between the cation polar
group of DDA with the hydrophilic groups of tetraethylen glycol at
the air/aqueous solution interface. Such amphiphilic compound
has hydrophilic groups similar to those of F507.
4. Conclusions
The effect of MIBC, the polyglycol frother F507 and the cationic
surfactant dodecyl amine (DDA) on the bubble size in a downflow
 M.A. Corona-Arroyo et al. / Minerals Engineering 81 (2015) 109–115
Fig. 10. Surface tension of F507 aqueous solutions as a function of F507 concen-
tration in the absence and presence of various concentrations of DDA at pH 7 and
23 °C.
system has the typical behavior of a significant decrease in bubble
size at low concentration up to the Critical Coalescence
Concentration (CCC). Above this CCC the bubble size is essentially
constant. The CCC and the bubble size distribution decreased with
the reagent molecular weight following the order
MIBC > DDA > F507. Blends of MIBC and F507 with DDA lowered
the CCC of the frother and gave bubbles of finer size below and above
the frother CCC in comparison to individual frothers. The gas holdup
in the downflow column decreased with the bubble size and was
lowest for the F507-DDA blends. It should be expected that the flota-
tion kinetics for fine particles will improve using DDA-frother blends
in flotation systems with downflow columns because of smaller size
of the bubbles.
Acknowledgements
M.A. Corona-Arroyo gratefully acknowledges the Consejo
Nacional de Ciencia y Tecnología of Mexico (CONACyT) for the
Fellowship No. 230133 to pursue Ph.D. studies on Materials
Science and Engineering at the State University of San Luis
Potosi, México.
References
Atkinson, B.W., Jameson, G.J., Nguyen, A.V., Evans, G.M., Machniewski, P.M., 2003.
Bubble breakup and coaslescence in a plunging jet bubble column. Can. J. Chem.
Eng. 81, 519–527.
115
Araya, R., Gomez, C., Finch, J., Huynh, L., Young, M., 2014. Characterization of frother
effects on gas dispersion in a Jameson cell. In: XXVII international mineral
processing congress, pp. 145–154.
Azgomi, F., Gomez, C.O., Finch, J.A., 2007. Characterizing frothers using gas hold-up.
Can. Metall. Q. 46, 237–242.
Batisteli, G.M., Peres, A.E., 2008. Residual amine in iron ore flotation. Miner. Eng. 21,
873–876.
Cho, Y.S., Laskowski, J.S., 2002. Effect of flotation frothers on bubble size and foam
stability. Int. J. Miner. Process. 64, 69–80.
Davies, J.T., 1957. A quantitative kinetic theory of emulsion type, I. Physical
chemistry of the emulsifying agent. In: Gas/Liquid and Liquid/Liquid Interface,
Proceedings of the International Congress of Surface Activity, pp. 426–438.
Elmahdy, A.M., Finch, J.A., 2013. Effect of frother blends on hydrodynamic
properties. Int. J. Miner. Process. 123, 60–63.
Evans, G.M., Atkinson, B.W., Jameson, G.J., 1995. The Jameson cell. Flot. Sci. Eng. 11,
331–363.
Evans, G.M., Jameson, G.J., Atkinson, B.W., 1992. Prediction of the bubble size
generated by a plunging liquid jet bubble column. Chem. Eng. Sci. 47 (13/14),
3265–3272.
Finch, J.A., Nesset, J.E., Acuña, C., 2008. Role of frother on bubble production and
behaviour in flotation. Miner. Eng. 21, 949–957.
Grau, R.A., Laskowski, J.S., 2006. Role of frothers in bubble generation and
coalescence in a mechanical flotation cell. Can. J. Chem. Eng. 84, 170–182.
Grau, R.A., Laskowski, J.S., Heiskanen, K., 2005. Effect of frothers on bubble size. Int.
J. Miner. Process. 76, 225–233.
Laskowski, J.S., 2003. Fundamental properties of flotation frothers. In: Proceedings
of the 22nd International Mineral Processing Congress, vol. 2. The South African
Institute of Mining and Metallurgy, pp. 788–797.
Laskowski, J.S., Tlhone, T., Williams, P., Ding, K., 2003. Fundamental properties of the
polyoxypropylene alkyl ether flotation frothers. Int. J. Miner. Process. 72, 289–
299.
Leiva, J., Vinnett, L., Contreras, F., Yianatos, J., 2010. Estimation of the actual bubble
surface area flux in flotation. Miner. Eng. 23, 888–894.
Leja, J., 1982. Surface Chemistry of Froth Flotation. Plenum Press, New York.
Majumder, S.K., Kundu, G., Mukherjee, D., 2005. Mixing mechanism in a modified
co-current downflow bubble column. Chem. Eng. J. 112, 45–55.
Mandal, A., Kundu, G., Mukherjee, D., 2005. A comparative study of gas holdup,
bubble size distribution and interfacial area in a downflow bubble column.
Chem. Eng. Res. Des. 83, 423–428.
Matsubara, H., Ohta, A., Kameda, M., Ikeda, N., Aratono, M., 2000. Interaction
between ionic and nonionic surfactants in the adsorbed film and micelle.
Dodecylammonium chloride and tetraethylene glycol monooctyl ether.
Langmuir 16, 7589–7596.
Nesset, J.E., Zhang, W., Finch, J.A., 2012. A benchmarking tool for assessing flotation
cell performance. In: Proceedings of 44th Annual Meeting of the Canadian
Mineral Processors (CIM). Ottawa, Canada, pp. 183–209.
Ho, O’Boen, 1998. Electrokinetic studies on emulsions stabilized by ionic
surfactants: the electroacoustophoretic behavior and estimation of Davies’
HLB increments. J. Colloid Interface Sci. 198, 249–260.
Tan, S.N., Pugh, R.J., Fornasiero, D., Sedev, R., Ralston, J., 2005. Foaming of
polypropylene glycols and glycol/MIBC mixtures. Miner. Eng. 18, 179–188.
Tan, Y.H., Rafiei, A.A., Elmahdy, A., Finch, J.A., 2013. Bubble size, gas holdup and
bubble velocity profile of some alcohols and commercial frothers. Int. J. Miner.
Process. 119, 1–5.
Tasßdemir, A., Tasßdemir, T., Öteyaka, B., 2007. The effect of particle size and some
operating parameters in the separation tank and the downcomer on the
Jameson cell recovery. Miner. Eng. 20, 1331–1336.
Wills, B.A., Napier-Munn, T.J., 2006. Mineral Processing Technology. Elsevier Ltd,
Oxford.
Zhang, W., Nesset, J.E., Rao, R., Finch, J.A., 2012. Characterizing frothers through
critical coalescence concentration (CCC) 95-hydrophile–lipophile balance (HLB)
relationship. Minerals 2, 208–227.