2
[8, 47–55], graphite [56–59], calcite [60], bitumen and oil
[61–67], muscovite [68], kaolinite [69], wolframite [70],
magnesite [71], PGM-bearing ore [72], niobium [73], ser-
pentine [74], ions [75], wastewater treatment [76–85], fly
ash [86], and spent batteries [87], etc. Extensive fundamen-
tal research has been done to improve the understanding
of cavitation generated nanobubble/microbubble [4–8, 47,
88–108] on various applications.
Over 1,000 Eriez flotation columns with a variety
of sparging technologies have been installed for floating
quartz from iron ore, various kinds of sulfide minerals, gold,
phosphate, coal, graphite, bauxite, calcite, bitumen, kaolin-
ite, PGM-bearing ore, niobium, potash, and oil/water, etc.
Typically, flotation columns produce higher grade and have
lower operating costs than conventional mechanical cells.
Figure 1 shows Eriez’ CavTube flotation column.
Flotation column cells function as three phase settlers where
particles move downwards in a hindered settling environ-
ment counter-current to a flux of rising air bubbles, which
are generated by spargers located near the bottom of the
cell. The sparger technology is an important design choice
and allows the user to optimize the performance based on
the feed characteristics, such as particle size distribution,
liberation class and floatability, etc. Eriez can recommend
and design the most appropriate equipment for the specific
application.
The mechanism of particle/bubble collision in columns
is different from intensive mixing devices such as mechani-
cally agitated cells. Under the low intensity mixing caused
only by a rising bubble field, particle drifting from the fluid
streamlines is caused by gravity and inertial forces and by
interception. A column can support a deep froth bed and
may use wash water to maintain a downward flow of water
(positive bias) evenly across the cross-section of the ves-
sel. This eliminates the entrainment of hydrophilic particles
in the float product when the vessel is used for solid/solid
separation. This property, along with the absence of bypass
of feed material to the float product from turbulence,
means that column cells are normally superior to impeller
type machines for the selective separation of ultrafine/fine
particles (high grade).
This paper investigates the effects of flotation column
wash water rate, feed solids content, and feed particle size
distribution on the product P2O5 grade and P2O5 recov-
ery of an ultrafine phosphate flotation. A three-level three-
factor central composite experimental design was employed
for investigating these three parameters for improving
ultrafine phosphate flotation selectivity. For comparison,
both benchtop Denver cell and laboratory column flotation
tests were performed at their respective optimal flotation
conditions. Various kinds of phosphate ores with the
gangue minerals composed of quartz/muscovite, calcite/
dolomite, and iron/silica were studied using CavTube gen-
erated fine and ultrafine bubbles.
MATERIALS AND METHODS
Three(3) separate studies were carried out in this investi-
gation, including microscopic observations of ultrafine
phosphate slime flotation froth with and without wash
water, the three-factor three-level experimental design,
and the column vs. mechanical cell comparison flotation.
A Celestron 44348 LCD digital microscope was used for
microscopic observation study. A S3500 MicroTrac particle
size analyzer using laser diffraction was employed to mea-
sure the feed particle size distribution.
Table 1 shows the three-factor three-level central com-
posite experimental design conducted for ultrafine phos-
phate column flotation tests using the Design-Expert
software acquired from Stat-Ease Inc., Minneapolis, MN.
Three process parameters included froth wash water flow
rate, feed slurry solids content, and feed particle size (P80).
Each numeric factor shown in Table 1 and Table 2 var-
ies over three levels: plus 1 and minus 1 (factorial points)
and the center point. The levels of process variables were
coded as “–1” “0” and “+1,” respectively, where “–” repre-
sents the low level, “0” represents the middle level and “+”
represents the high level of the factors. The specific levels of
individual variables are indicated in Table 1. The details of
designed experiments are shown in Table 2.
Figure 1. Illustration of Eriez CavTube flotation column
[8, 47–55], graphite [56–59], calcite [60], bitumen and oil
[61–67], muscovite [68], kaolinite [69], wolframite [70],
magnesite [71], PGM-bearing ore [72], niobium [73], ser-
pentine [74], ions [75], wastewater treatment [76–85], fly
ash [86], and spent batteries [87], etc. Extensive fundamen-
tal research has been done to improve the understanding
of cavitation generated nanobubble/microbubble [4–8, 47,
88–108] on various applications.
Over 1,000 Eriez flotation columns with a variety
of sparging technologies have been installed for floating
quartz from iron ore, various kinds of sulfide minerals, gold,
phosphate, coal, graphite, bauxite, calcite, bitumen, kaolin-
ite, PGM-bearing ore, niobium, potash, and oil/water, etc.
Typically, flotation columns produce higher grade and have
lower operating costs than conventional mechanical cells.
Figure 1 shows Eriez’ CavTube flotation column.
Flotation column cells function as three phase settlers where
particles move downwards in a hindered settling environ-
ment counter-current to a flux of rising air bubbles, which
are generated by spargers located near the bottom of the
cell. The sparger technology is an important design choice
and allows the user to optimize the performance based on
the feed characteristics, such as particle size distribution,
liberation class and floatability, etc. Eriez can recommend
and design the most appropriate equipment for the specific
application.
The mechanism of particle/bubble collision in columns
is different from intensive mixing devices such as mechani-
cally agitated cells. Under the low intensity mixing caused
only by a rising bubble field, particle drifting from the fluid
streamlines is caused by gravity and inertial forces and by
interception. A column can support a deep froth bed and
may use wash water to maintain a downward flow of water
(positive bias) evenly across the cross-section of the ves-
sel. This eliminates the entrainment of hydrophilic particles
in the float product when the vessel is used for solid/solid
separation. This property, along with the absence of bypass
of feed material to the float product from turbulence,
means that column cells are normally superior to impeller
type machines for the selective separation of ultrafine/fine
particles (high grade).
This paper investigates the effects of flotation column
wash water rate, feed solids content, and feed particle size
distribution on the product P2O5 grade and P2O5 recov-
ery of an ultrafine phosphate flotation. A three-level three-
factor central composite experimental design was employed
for investigating these three parameters for improving
ultrafine phosphate flotation selectivity. For comparison,
both benchtop Denver cell and laboratory column flotation
tests were performed at their respective optimal flotation
conditions. Various kinds of phosphate ores with the
gangue minerals composed of quartz/muscovite, calcite/
dolomite, and iron/silica were studied using CavTube gen-
erated fine and ultrafine bubbles.
MATERIALS AND METHODS
Three(3) separate studies were carried out in this investi-
gation, including microscopic observations of ultrafine
phosphate slime flotation froth with and without wash
water, the three-factor three-level experimental design,
and the column vs. mechanical cell comparison flotation.
A Celestron 44348 LCD digital microscope was used for
microscopic observation study. A S3500 MicroTrac particle
size analyzer using laser diffraction was employed to mea-
sure the feed particle size distribution.
Table 1 shows the three-factor three-level central com-
posite experimental design conducted for ultrafine phos-
phate column flotation tests using the Design-Expert
software acquired from Stat-Ease Inc., Minneapolis, MN.
Three process parameters included froth wash water flow
rate, feed slurry solids content, and feed particle size (P80).
Each numeric factor shown in Table 1 and Table 2 var-
ies over three levels: plus 1 and minus 1 (factorial points)
and the center point. The levels of process variables were
coded as “–1” “0” and “+1,” respectively, where “–” repre-
sents the low level, “0” represents the middle level and “+”
represents the high level of the factors. The specific levels of
individual variables are indicated in Table 1. The details of
designed experiments are shown in Table 2.
Figure 1. Illustration of Eriez CavTube flotation column