916 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
The sharp annual decreases are linked to the thawing
of the snow cover over the tailings ponds, and the resulting
dilution of the process water. During the summer months,
conductivity rises again due to the continuous dissolution
of ore in the process water, which occurs in the flotation
circuit and then in the tailings ponds over longer term. The
figure also illustrates the overall trend of increasing conduc-
tivity over years.
In order to address the issue of water management
in its entirety, it therefore appears necessary to take into
account ore dissolution in the modeling, which accounts
for these variations of water quality. This would be manda-
tory to simulate long-loop circuits, for which the evolution
of water composition mainly occurs in the tailings ponds,
as well as for short-loop recirculation scenarios, for which
dissolution occurs in the flotation circuit.
The latter case is discussed in this paper, which pres-
ents the extension of the previously developed plant model
(Braak et al., 2022) by integrating the ore dissolution aspect
based on an experiment developed during the project (Le
2020, 2021, Le, Schreitofer and Dahl 2020). The aim of
this experiment was to predict the long-term composition
of the process water obtained by reproducing, at lab scale,
the dissolution of ore that may occur into the grinding/
flotation circuit and the effect of water recycling by redo-
ing the same dissolution test using recycled water. For the
first cycle, distilled water (DW), onsite river water (RW) or
onsite process water (PW) have been used as starting water
(SW). The water obtained by slurry filtration at the end
of each cycle was used for the next cycle, and so on until
a stable composition is achieved (obtained after approxi-
mately 8 cycles). Water composition was monitored, and
flotation tests were carried out with the dissolution loop
water (DLW) obtained at the end of the experiment, and
compared with similar tests using SW and PW. As simi-
lar results have been observed for RW and DW, only DW
results are considered. Figure 2 shows the variations in sul-
fate concentration over the test cycles and dissolution mod-
elling results as described in the next section.
Points correspond to the measured concentrations at
the end of cycles 1,3, 6 and 8. The dot-line curve is given by
the dissolution model and its specific shape illustrates the
various stages of the experimental protocol: dilution at the
start of the repeated cycles caused by addition of SW, which
is necessary to compensate for water lost in the filtration
cake and for analytical samples, and quasi-linear increase in
concentration during the dissolution sequence.
The present paper first details the dissolution model
adapted to these experimental results, its calibration, and
the faced issues. This model is then integrated into the plant
model, which considers short-loop scenarios and predicts
the composition that could be reached in steady state with-
out any water treatment operation. In a further step, the
flotation laws obtained experimentally are also integrated
Figure 1. Conductivity variation in Kevitsa process water (Muzinda and Schreitofer 2018)
The sharp annual decreases are linked to the thawing
of the snow cover over the tailings ponds, and the resulting
dilution of the process water. During the summer months,
conductivity rises again due to the continuous dissolution
of ore in the process water, which occurs in the flotation
circuit and then in the tailings ponds over longer term. The
figure also illustrates the overall trend of increasing conduc-
tivity over years.
In order to address the issue of water management
in its entirety, it therefore appears necessary to take into
account ore dissolution in the modeling, which accounts
for these variations of water quality. This would be manda-
tory to simulate long-loop circuits, for which the evolution
of water composition mainly occurs in the tailings ponds,
as well as for short-loop recirculation scenarios, for which
dissolution occurs in the flotation circuit.
The latter case is discussed in this paper, which pres-
ents the extension of the previously developed plant model
(Braak et al., 2022) by integrating the ore dissolution aspect
based on an experiment developed during the project (Le
2020, 2021, Le, Schreitofer and Dahl 2020). The aim of
this experiment was to predict the long-term composition
of the process water obtained by reproducing, at lab scale,
the dissolution of ore that may occur into the grinding/
flotation circuit and the effect of water recycling by redo-
ing the same dissolution test using recycled water. For the
first cycle, distilled water (DW), onsite river water (RW) or
onsite process water (PW) have been used as starting water
(SW). The water obtained by slurry filtration at the end
of each cycle was used for the next cycle, and so on until
a stable composition is achieved (obtained after approxi-
mately 8 cycles). Water composition was monitored, and
flotation tests were carried out with the dissolution loop
water (DLW) obtained at the end of the experiment, and
compared with similar tests using SW and PW. As simi-
lar results have been observed for RW and DW, only DW
results are considered. Figure 2 shows the variations in sul-
fate concentration over the test cycles and dissolution mod-
elling results as described in the next section.
Points correspond to the measured concentrations at
the end of cycles 1,3, 6 and 8. The dot-line curve is given by
the dissolution model and its specific shape illustrates the
various stages of the experimental protocol: dilution at the
start of the repeated cycles caused by addition of SW, which
is necessary to compensate for water lost in the filtration
cake and for analytical samples, and quasi-linear increase in
concentration during the dissolution sequence.
The present paper first details the dissolution model
adapted to these experimental results, its calibration, and
the faced issues. This model is then integrated into the plant
model, which considers short-loop scenarios and predicts
the composition that could be reached in steady state with-
out any water treatment operation. In a further step, the
flotation laws obtained experimentally are also integrated
Figure 1. Conductivity variation in Kevitsa process water (Muzinda and Schreitofer 2018)