XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1559
perspective, the present article investigates the implementa-
tion of a geometallurgical program at the Vargem Grande
Iron Ore Complex in Brazil, with the objective of refining
process reconciliation accuracy and enhancing overall plant
performance.
The beginning of this program can be traced back to
a production scenario characterized by a with a hematite
concentrate containing 1.5% SiO2. This challenge was
aggravated by the inclusion of material from Horizontes
and Abóboras mines not initially considered in the project,
introducing ores with lower quartz liberation. Concurrently,
variability in metallurgical performance among mines sup-
plying Vargem Grande 2 Plant (VGR2) led to suboptimal
recoveries, elevating both project risk and potential revenue
losses, aligning with findings from prior studies (Ashley and
Callow, 2000 Bye, 2011 Vann et al., 2012 Rendu, 2017
Dominy et al., 2018). The Abóboras and Horizontes mines
are situated on the eastern flank of the Moeda Syncline in
the Iron Quadrangle region in Brazil, which is an impor-
tant iron mineral province. The deposits consist of banded
iron formations of the Lake Superior-type and their hydro-
thermal and/or supergene alteration products. These iron
formations are embedded in a metasedimentary sequence
composed of clastic and chemical platform sediments.
The principal ore-hosting rocks are friable and compact
itabirites. Moreover, the ores exhibited low predictability
in mineralogical composition and energy consumption,
impeding efficient processing. Challenges extended to ore
tracking from mine to plant, coupled with a significant
information gap in short-term processing, further compli-
cating decision-making.
The need to adapt and refine existing methodologies
in response to these challenges drove the development and
implementation of the geometallurgical program. The pro-
gram proves beneficial in mitigating risks and improving
the efficiency of processes, encompassing both technical
and operational aspects (Michaux and O’Connor, 2020).
Within this article, a detailed exploration of this initiative
is undertaken, emphasizing its successful resolution of the
unique geological and metallurgical challenges at Vargem
Grande. Ultimately, the program contributes significantly
to optimizing mineral processing, refining production
planning strategies, and ensuring the sustainable extraction
of iron ore resources.
MATERIALS AND METHODS
The Geometallurgy Program implemented at the Vargem
Grande Iron Ore Complex in Brazil adopted a systematic
approach, comprising several crucial stages. Initially, sam-
pling procedures were conducted to ensure the collection of
representative ore samples. Subsequently, mineralogical and
chemical characterization analyses were performed on these
samples to examine their mineral composition and chemi-
cal properties. Concurrently, bench scale tests were con-
ducted. Following this, test development aimed at model
generation was initiated to establish a robust framework for
predicting ore behavior during processing. This phase inte-
grated data from both the characterization and test devel-
opment stages to construct an integrated predictive model.
Finally, industrial data evaluation was conducted to validate
the effectiveness of the developed model under real-world
operating conditions.
Sampling
The sampling process for the Geometallurgy Program
was meticulously designed to capture the intricate geo-
logical attributes essential for understanding metallurgical
responses within ore bodies. Information available in the
project database, such as distribution of Fe, SiO2, Al2O3
contents, particle size distribution of samples, and litho-
logical classification constituted the primary characteristics
considered for sample selection. Samples were collected
from core drillings at Abóboras (ABO) and Horizontes
(HZT) mines, focusing on ore designated for VGR2 (‘wet
ore’) as outlined in the mine production plans for 2021
and 2022. A total of 58 samples from Abóboras and 66
samples from Horizontes were chosen, representing con-
tinuous intervals of core drillings approximately 10 meters
in length. This meticulous sampling approach ensured
the representation of diverse geological characteristics
and spatial variability within the ore body, providing cru-
cial insights for the development and optimization of the
Geometallurgy Program.
Characterization and Bench Scale Tests
The comprehensive understanding of ore samples through
mineralogical and chemical characterization, as well as
bench-scale tests, as illustrated in Figure 1, is crucial for
gaining valuable insights into their composition, proper-
ties, and optimizing processing efficiency and performance.
These methodologies enable the thorough assessment of
diverse processing parameters, thereby facilitating informed
decision-making in mineral processing operations.
In the initial phase of the study, the ore samples under-
went a series of preparatory steps. They were first crushed to
a size of 2 mm and subsequently ground to achieve a P95 of
0.150 mm. Grinding operations were conducted in jar mills
with dimensions of 20,32cm in diameter and 30,48cm in
length, utilizing balls as grinding media. Following grind-
ing, the PRED test (Donda Energy Requirement Prediction
perspective, the present article investigates the implementa-
tion of a geometallurgical program at the Vargem Grande
Iron Ore Complex in Brazil, with the objective of refining
process reconciliation accuracy and enhancing overall plant
performance.
The beginning of this program can be traced back to
a production scenario characterized by a with a hematite
concentrate containing 1.5% SiO2. This challenge was
aggravated by the inclusion of material from Horizontes
and Abóboras mines not initially considered in the project,
introducing ores with lower quartz liberation. Concurrently,
variability in metallurgical performance among mines sup-
plying Vargem Grande 2 Plant (VGR2) led to suboptimal
recoveries, elevating both project risk and potential revenue
losses, aligning with findings from prior studies (Ashley and
Callow, 2000 Bye, 2011 Vann et al., 2012 Rendu, 2017
Dominy et al., 2018). The Abóboras and Horizontes mines
are situated on the eastern flank of the Moeda Syncline in
the Iron Quadrangle region in Brazil, which is an impor-
tant iron mineral province. The deposits consist of banded
iron formations of the Lake Superior-type and their hydro-
thermal and/or supergene alteration products. These iron
formations are embedded in a metasedimentary sequence
composed of clastic and chemical platform sediments.
The principal ore-hosting rocks are friable and compact
itabirites. Moreover, the ores exhibited low predictability
in mineralogical composition and energy consumption,
impeding efficient processing. Challenges extended to ore
tracking from mine to plant, coupled with a significant
information gap in short-term processing, further compli-
cating decision-making.
The need to adapt and refine existing methodologies
in response to these challenges drove the development and
implementation of the geometallurgical program. The pro-
gram proves beneficial in mitigating risks and improving
the efficiency of processes, encompassing both technical
and operational aspects (Michaux and O’Connor, 2020).
Within this article, a detailed exploration of this initiative
is undertaken, emphasizing its successful resolution of the
unique geological and metallurgical challenges at Vargem
Grande. Ultimately, the program contributes significantly
to optimizing mineral processing, refining production
planning strategies, and ensuring the sustainable extraction
of iron ore resources.
MATERIALS AND METHODS
The Geometallurgy Program implemented at the Vargem
Grande Iron Ore Complex in Brazil adopted a systematic
approach, comprising several crucial stages. Initially, sam-
pling procedures were conducted to ensure the collection of
representative ore samples. Subsequently, mineralogical and
chemical characterization analyses were performed on these
samples to examine their mineral composition and chemi-
cal properties. Concurrently, bench scale tests were con-
ducted. Following this, test development aimed at model
generation was initiated to establish a robust framework for
predicting ore behavior during processing. This phase inte-
grated data from both the characterization and test devel-
opment stages to construct an integrated predictive model.
Finally, industrial data evaluation was conducted to validate
the effectiveness of the developed model under real-world
operating conditions.
Sampling
The sampling process for the Geometallurgy Program
was meticulously designed to capture the intricate geo-
logical attributes essential for understanding metallurgical
responses within ore bodies. Information available in the
project database, such as distribution of Fe, SiO2, Al2O3
contents, particle size distribution of samples, and litho-
logical classification constituted the primary characteristics
considered for sample selection. Samples were collected
from core drillings at Abóboras (ABO) and Horizontes
(HZT) mines, focusing on ore designated for VGR2 (‘wet
ore’) as outlined in the mine production plans for 2021
and 2022. A total of 58 samples from Abóboras and 66
samples from Horizontes were chosen, representing con-
tinuous intervals of core drillings approximately 10 meters
in length. This meticulous sampling approach ensured
the representation of diverse geological characteristics
and spatial variability within the ore body, providing cru-
cial insights for the development and optimization of the
Geometallurgy Program.
Characterization and Bench Scale Tests
The comprehensive understanding of ore samples through
mineralogical and chemical characterization, as well as
bench-scale tests, as illustrated in Figure 1, is crucial for
gaining valuable insights into their composition, proper-
ties, and optimizing processing efficiency and performance.
These methodologies enable the thorough assessment of
diverse processing parameters, thereby facilitating informed
decision-making in mineral processing operations.
In the initial phase of the study, the ore samples under-
went a series of preparatory steps. They were first crushed to
a size of 2 mm and subsequently ground to achieve a P95 of
0.150 mm. Grinding operations were conducted in jar mills
with dimensions of 20,32cm in diameter and 30,48cm in
length, utilizing balls as grinding media. Following grind-
ing, the PRED test (Donda Energy Requirement Prediction