1438 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
by process mineralogists, but rather by flotation specialists.
It is envisaged to incorporate flotation test works into the
process mineralogical workflow in the future.
The pre-concentrate from the dry magnetic separation
process is subject to grinding tests. These tests follow a spe-
cific method developed for magnetite ore at LKAB, called
the Malmberget method. It is described by 10 minutes grind-
ing in the rod mill followed by 25 and 35 minutes in the
ball mill, leading to three comminution products (Drugge
2009). The laboratory mills have a diameter of 200 mm
and a length of 250 mm. They were designed based on
empirical studies in the beneficiation plant KA1 in Kiruna.
To match the test quality criteria, rods are counted (n=53)
and measured (±14.200 g ± 50 g) and steel balls (diameter
15 mm) measured to a total mass of 13.100 g ± 50 g. The
feed material characterised by a sample of 2 kg ± 100 g is
first ground with 1 liter of water for 10 min in the rod
mill (A). After drying at ca. 100°C for 12 h and subsequent
homogenization, the sample is further ground in the ball
mill for 25 min (B) or 35 min (C), respectively. Commonly,
particle size analyses for the material after each grinding
stage is obtained by wet sieving at LKAB’s laboratory for
physical testing. This information is subsequently used to
generate particle size distribution (PSD) curves and energy
demand calculations based on observations and results from
Bergström and Anttila (1973). The calculations relate to
energy consumption and conversion ratios from laboratory
scale test works, pilot scale test works up to industrial scale
in beneficiation plants in Kiruna. They are used to compare
the energy demands needed to grind to a certain P80 for
more heterogenous ores of currently investigated depos-
its. Furthermore, chemical analysis by X-ray fluorescence
(XRF) of each size fraction is carried out at LKAB’s chemi-
cal laboratory. If necessary, SEM-QEMSCAN/TIMA work
is carried out to identify the degree of liberation of par-
ticles and to infer on the modal mineralogy of the different
size fractions of the three grinding products. Such commi-
nution test work at the laboratory scale has been shown
effective in characterising heterogenous ores and to predict
their behavior during comminution at the industrial scale
(Krolop et al. 2022).
Separation test works by DT or Blue Ribbon are used
to generate magnetic concentrates consisting mainly of
magnetite. Depending on the achieved degree of libera-
tion in the grinding process, gangue and other ore min-
erals account for the waste material. The selected samples
originate either from grinding step B or C, depending on
the achieved PSD and P80 value. The amount of material
used in the separation runs depends on the used magnetic
separator. In any way, concentrate (magnetic) fractions
and waste (nonmagnetic) fractions for each sample are
achieved and subsequently analysed for chemical composi-
tion, elemental and mass recovery as well as their mineral-
ogical (modal mineralogy and liberation) characteristics. If
Figure 3. Illustration of the common process mineralogical workflow including sampling procedure from drill core, dry
low‑magnetic separation followed by subsequent grinding and wet low-magnetic separation for magnetite ore analysed by
SEM‑TIMA
by process mineralogists, but rather by flotation specialists.
It is envisaged to incorporate flotation test works into the
process mineralogical workflow in the future.
The pre-concentrate from the dry magnetic separation
process is subject to grinding tests. These tests follow a spe-
cific method developed for magnetite ore at LKAB, called
the Malmberget method. It is described by 10 minutes grind-
ing in the rod mill followed by 25 and 35 minutes in the
ball mill, leading to three comminution products (Drugge
2009). The laboratory mills have a diameter of 200 mm
and a length of 250 mm. They were designed based on
empirical studies in the beneficiation plant KA1 in Kiruna.
To match the test quality criteria, rods are counted (n=53)
and measured (±14.200 g ± 50 g) and steel balls (diameter
15 mm) measured to a total mass of 13.100 g ± 50 g. The
feed material characterised by a sample of 2 kg ± 100 g is
first ground with 1 liter of water for 10 min in the rod
mill (A). After drying at ca. 100°C for 12 h and subsequent
homogenization, the sample is further ground in the ball
mill for 25 min (B) or 35 min (C), respectively. Commonly,
particle size analyses for the material after each grinding
stage is obtained by wet sieving at LKAB’s laboratory for
physical testing. This information is subsequently used to
generate particle size distribution (PSD) curves and energy
demand calculations based on observations and results from
Bergström and Anttila (1973). The calculations relate to
energy consumption and conversion ratios from laboratory
scale test works, pilot scale test works up to industrial scale
in beneficiation plants in Kiruna. They are used to compare
the energy demands needed to grind to a certain P80 for
more heterogenous ores of currently investigated depos-
its. Furthermore, chemical analysis by X-ray fluorescence
(XRF) of each size fraction is carried out at LKAB’s chemi-
cal laboratory. If necessary, SEM-QEMSCAN/TIMA work
is carried out to identify the degree of liberation of par-
ticles and to infer on the modal mineralogy of the different
size fractions of the three grinding products. Such commi-
nution test work at the laboratory scale has been shown
effective in characterising heterogenous ores and to predict
their behavior during comminution at the industrial scale
(Krolop et al. 2022).
Separation test works by DT or Blue Ribbon are used
to generate magnetic concentrates consisting mainly of
magnetite. Depending on the achieved degree of libera-
tion in the grinding process, gangue and other ore min-
erals account for the waste material. The selected samples
originate either from grinding step B or C, depending on
the achieved PSD and P80 value. The amount of material
used in the separation runs depends on the used magnetic
separator. In any way, concentrate (magnetic) fractions
and waste (nonmagnetic) fractions for each sample are
achieved and subsequently analysed for chemical composi-
tion, elemental and mass recovery as well as their mineral-
ogical (modal mineralogy and liberation) characteristics. If
Figure 3. Illustration of the common process mineralogical workflow including sampling procedure from drill core, dry
low‑magnetic separation followed by subsequent grinding and wet low-magnetic separation for magnetite ore analysed by
SEM‑TIMA