2889
Studying Particle Entrainment Mechanisms in Froth Flotation
Ali Hassan
Institute for Process Engineering and Environmental Technology, Technische Universität Dresden, Germany
Helmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, Freiberg, Germany
Maryrose Joy Aparece, Martin Rudolph, Lucas Pereira
Helmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, Freiberg, Germany
ABSTRACT: Particle entrainment in froth flotation plays a pivotal role in overall process performance and
separation efficiency, but modelling its mechanisms remain a significant research challenge. This study aims
at the application of those models on a simple model ore comprising a pyrite-quartz system and a complex
porphyry copper ore in batch flotation tests. Moreover, the effects of hydrodynamic conditions (impeller tip
speed and superficial gas velocity) and reagent concentrations (collector and frother) on entrainment were
evaluated, along with the combined effect and interaction of these variables. In the last step, particle-based
separation modelling, i.e., logistic regression-based models were trained on particle datasets obtained from
2D-automated mineralogy from complex copper ore flotation to compute recovery probabilities for all particles.
Entrainment modelling approaches were applied on individual particles and efforts were made to understand
entrainment as a function of process conditions and particulate properties (size, density, shape, overall and
surface mineral composition, etc.).
Keywords: Entrainment modelling, automated mineralogy, particle-based separation modelling
INTRODUCTION
Flotation is a versatile and widely used physiochemical
separation technique that takes advantage of natural and
induced differences in surface properties of minerals. In
flotation, air bubbles are introduced to a pulp containing
the mineral particles. Selective attachment of target mineral
particles to the air bubbles allows their recovery in the froth
phase, while other mineral particles (i.e., gangue) remain
suspended in the pulp. The flotation process involves three
main mechanisms: 1) selective recovery of mineral particles
after their attachment to air bubbles, known as true flota-
tion, 2) entrainment of mineral particles in water passing
through the froth, and 3) physical entrapment of particles
between bubbles in the froth. Of these, true flotation is the
dominant mechanism for the selective recovery of valuable
minerals. Entrainment and entrapment, while not selec-
tive, also contribute to the overall recovery (Wills &Finch,
2015). Entrainment occurs because water inherently trans-
fers between the pulp and froth phases during flotation.
This water carries fine suspended solids into the froth as
well (Yu et al., 2017 Wang &Liu, 2021, Laplante et al.,
1989 Smith &Warren, 1989 Maachar &Dobby, 1992
Neethling &Cilliers, 2002 Zheng et al., 2006 Wang et
al., 2016). The more general form of defining the degree
of entrainment (ENTi) according Savassi et al. (1998) is
given by:
Studying Particle Entrainment Mechanisms in Froth Flotation
Ali Hassan
Institute for Process Engineering and Environmental Technology, Technische Universität Dresden, Germany
Helmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, Freiberg, Germany
Maryrose Joy Aparece, Martin Rudolph, Lucas Pereira
Helmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, Freiberg, Germany
ABSTRACT: Particle entrainment in froth flotation plays a pivotal role in overall process performance and
separation efficiency, but modelling its mechanisms remain a significant research challenge. This study aims
at the application of those models on a simple model ore comprising a pyrite-quartz system and a complex
porphyry copper ore in batch flotation tests. Moreover, the effects of hydrodynamic conditions (impeller tip
speed and superficial gas velocity) and reagent concentrations (collector and frother) on entrainment were
evaluated, along with the combined effect and interaction of these variables. In the last step, particle-based
separation modelling, i.e., logistic regression-based models were trained on particle datasets obtained from
2D-automated mineralogy from complex copper ore flotation to compute recovery probabilities for all particles.
Entrainment modelling approaches were applied on individual particles and efforts were made to understand
entrainment as a function of process conditions and particulate properties (size, density, shape, overall and
surface mineral composition, etc.).
Keywords: Entrainment modelling, automated mineralogy, particle-based separation modelling
INTRODUCTION
Flotation is a versatile and widely used physiochemical
separation technique that takes advantage of natural and
induced differences in surface properties of minerals. In
flotation, air bubbles are introduced to a pulp containing
the mineral particles. Selective attachment of target mineral
particles to the air bubbles allows their recovery in the froth
phase, while other mineral particles (i.e., gangue) remain
suspended in the pulp. The flotation process involves three
main mechanisms: 1) selective recovery of mineral particles
after their attachment to air bubbles, known as true flota-
tion, 2) entrainment of mineral particles in water passing
through the froth, and 3) physical entrapment of particles
between bubbles in the froth. Of these, true flotation is the
dominant mechanism for the selective recovery of valuable
minerals. Entrainment and entrapment, while not selec-
tive, also contribute to the overall recovery (Wills &Finch,
2015). Entrainment occurs because water inherently trans-
fers between the pulp and froth phases during flotation.
This water carries fine suspended solids into the froth as
well (Yu et al., 2017 Wang &Liu, 2021, Laplante et al.,
1989 Smith &Warren, 1989 Maachar &Dobby, 1992
Neethling &Cilliers, 2002 Zheng et al., 2006 Wang et
al., 2016). The more general form of defining the degree
of entrainment (ENTi) according Savassi et al. (1998) is
given by: