2316 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
and high density and at high radial particle velocity in
the agitator region of the cell. The collision contact time
is the time at which a bubble and particle move towards
each other in a straight line such that the bubble surface
becomes deformed. The sliding contact time theory how-
ever can be applied to particles with low inertia and density
or fine particles. The sliding contact time is the time it takes
for a particle to slide along a bubble surface such that there
is no deformation of the bubble (Albijanic et al., 2010).
Ralston et al. (1999) further stated that during collision
interactions particles usually rebound from the surface of
the bubble which results in multiple collisions and subse-
quently the collision interaction becomes a sliding interac-
tion. The sliding contact time is overall longer than that of
the collision contact time and as a result, the sliding con-
tact time is used as tcontact in Equation 1 for modelling pur-
poses. The contact time available for attachment increases
proportionally with bubble size. Good induction times are
usually around the order of 10 ms. Ye et al. (1989) used the
attachment time measurements to describe the hydropho-
bicity of mineral particles, both natural and induced with
surfactant.
There are a number of instruments designed to mea-
sure attachment and/or induction time to provide a mea-
sure of bubble-particle attachment. This review aims to
consider these instruments and methodologies and provide
an account of their benefits and drawbacks.
FACTORS AFFECTING BUBBLE-
PARTICLE ATTACHMENT
Bubble Characteristics
Gu et al. (2004) studied the effect of bubble diameter on
bubble-particle attachment. The bubble diameter was var-
ied, and attachment time equipment was used with a bed
of silica particles. The results revealed that the time for par-
ticles to attach to the bubble increases with increasing bub-
ble diameter. In a study by Wang et al. (2005), methylated
beads were used to determine the effect of bubble size on
induction time these authors showed that induction time
became longer with bubble size. Yoon and Yordan (1991)
explained the effect of bubble size on induction time and
consequently the rate of film thinning by the Reynolds
lubrication theory. This theory states that the drainage
velocity of the thin film between the sphere (bubble) and an
infinitely large plate (particle bed) is directly proportional
to the force at which the sphere is approaching. The film
thickness is inversely proportional to both the viscosity of
the film and the squared radius. Thus, larger bubbles will
have a lesser rate of film thinning resulting in an increase in
induction time. Additionally, it will take longer to displace
the film between the bubble and particle with larger bub-
bles (Yoon and Yordan, 1991 Ye et al., 1989). The bubbles’
gas type has also been shown to affect how particles attach
to bubbles. Gu et al. (2004) studied the effect of the type
of gas used to form bubbles on the attachment time. These
authors found that hydrogen bubbles resulted in shorter
attachment times compared to oxygen bubbles they attrib-
uted this to the fact that oxygen molecules are more polar
and will result in strong hydrogen bonding when it encoun-
ters water which in turn stabilises the liquid film, resulting
in longer attachment times.
Particle Characteristics
The particle size distribution is considered a crucial param-
eter in designing and optimising flotation processes. Thus,
its impact on bubble-particle attachment becomes impor-
tant to study. A study by Bradshaw and O’Connor (1996)
investigated the effect of particle size class on bubble-par-
ticle attachment by means of bubble loading experiments.
The results from this study showed that the number of
particles per bubble decreased as the size fraction increased
and the mass of particles per bubble increased at increased
size fractions. Yoon and Yordan (1991) used induction time
Figure 1. The process of bubble-particle attachment
(Albijanic et al., 2010)
Previous Page Next Page