468
Sensor-Based Sorting Applied to the Beneficiation of a
Siliceous Limestone
E. Tomey, S. Seiler, I. de Amores, G. Sánchez
Departamento Ingeniería de Materiales y Minas, Facultad de Ingeniería, Universidad de la República, Uruguay
ABSTRACT: Limestone, a versatile material with applications spanning diverse industries, finds its most
prominent use in cement production. However, its required properties vary greatly depending on the specific
application. Limestone from the Queguay Formation in Uruguay has quartz as main impurity, which is not
detrimental for the cement industry, yet it does not meet the specifications of more demanding industries, such
as the lime and pulp industries. In this way, improving the calcite grade of the local limestone came out as a
challenge to place this material in other markets. Sensor based sorting, is a dry preconcentration technology
which does not require grinding. Before implementing this technology, a study needs to be conducted to
select a suitable sensor capable of properly sorting the material of interest. Thus, in the present work four
different sensors, x-ray transmission (XRT), x-ray fluorescence (XRF), laser, and colorimeter, were used to scan a
limestone sample and determine their sorting performance. The information collected was process through two
approaches, receiver operating characteristic and simple linear regression. Results showed that XRT and XRF
were the sensing technologies whose performances stood out from the rest.
INTRODUCTION
Limestone, predominantly constituted of calcite, is an
extensively employed material within the manufactur-
ing sector. Beyond serving as the primary raw material in
lime and cement production, it finds application in diverse
industries, including sugar, glass, steelmaking, and pulp
production. Within the context of Uruguay, limestone is
a relevant non-metallic mineral resource, primarily used in
the cement industry (Cernuschi, 2014).
Uruguayan limestone from the Queguay Formation is
characterized by the presence of siliceous impurities, pri-
marily in the form of quartz, a composition well-suited for
the cement industry (de Amores, Selier, Tomey, &Sánchez,
2023). However, it does not meet the stringent require-
ments demanded by other industries, such as the rapidly
developing pulp industry in the country (UPM Biofore,
2023).
To enhance the calcite grade of the local limestone, the
implementation of particle sorting technology has emerged
as a promising approach. Sensor-based sorting (SBS) is a
preconcentration process that stands out because water
usage is needless to operate (Li, Klein, Sun, &Kou, 2020
Robben, 2019). Its operation is based on comparing a mea-
sured property against a predetermined threshold. This
threshold serves as a criterion to determine whether the
material should be accepted or rejected.
The successful execution of four interconnected pro-
cesses is essential for effective particle sorting (Wotruba,
2015). Firstly, the material is conveyed on a belt, arranging
rocks in a monolayer to facilitate the measurement process.
Subsequently, a sensor measures rock by rock a property
Sensor-Based Sorting Applied to the Beneficiation of a
Siliceous Limestone
E. Tomey, S. Seiler, I. de Amores, G. Sánchez
Departamento Ingeniería de Materiales y Minas, Facultad de Ingeniería, Universidad de la República, Uruguay
ABSTRACT: Limestone, a versatile material with applications spanning diverse industries, finds its most
prominent use in cement production. However, its required properties vary greatly depending on the specific
application. Limestone from the Queguay Formation in Uruguay has quartz as main impurity, which is not
detrimental for the cement industry, yet it does not meet the specifications of more demanding industries, such
as the lime and pulp industries. In this way, improving the calcite grade of the local limestone came out as a
challenge to place this material in other markets. Sensor based sorting, is a dry preconcentration technology
which does not require grinding. Before implementing this technology, a study needs to be conducted to
select a suitable sensor capable of properly sorting the material of interest. Thus, in the present work four
different sensors, x-ray transmission (XRT), x-ray fluorescence (XRF), laser, and colorimeter, were used to scan a
limestone sample and determine their sorting performance. The information collected was process through two
approaches, receiver operating characteristic and simple linear regression. Results showed that XRT and XRF
were the sensing technologies whose performances stood out from the rest.
INTRODUCTION
Limestone, predominantly constituted of calcite, is an
extensively employed material within the manufactur-
ing sector. Beyond serving as the primary raw material in
lime and cement production, it finds application in diverse
industries, including sugar, glass, steelmaking, and pulp
production. Within the context of Uruguay, limestone is
a relevant non-metallic mineral resource, primarily used in
the cement industry (Cernuschi, 2014).
Uruguayan limestone from the Queguay Formation is
characterized by the presence of siliceous impurities, pri-
marily in the form of quartz, a composition well-suited for
the cement industry (de Amores, Selier, Tomey, &Sánchez,
2023). However, it does not meet the stringent require-
ments demanded by other industries, such as the rapidly
developing pulp industry in the country (UPM Biofore,
2023).
To enhance the calcite grade of the local limestone, the
implementation of particle sorting technology has emerged
as a promising approach. Sensor-based sorting (SBS) is a
preconcentration process that stands out because water
usage is needless to operate (Li, Klein, Sun, &Kou, 2020
Robben, 2019). Its operation is based on comparing a mea-
sured property against a predetermined threshold. This
threshold serves as a criterion to determine whether the
material should be accepted or rejected.
The successful execution of four interconnected pro-
cesses is essential for effective particle sorting (Wotruba,
2015). Firstly, the material is conveyed on a belt, arranging
rocks in a monolayer to facilitate the measurement process.
Subsequently, a sensor measures rock by rock a property