The implications of sulphides in GCC feed and the potential for their removal during alkaline amine flotation
Abstract
Ground calcium carbonate (GCC) slurries are commonly used in paper to improve
properties such as whiteness, printability, opacity and gloss. The brightness R457 (i.e.
the reflectance at a wavelength of 457 nm) and colour specifications constitute the main
quality parameters for GCC slurries, and must be within certain tolerances set by the
paper industry. Hence, the removal of any dark or colouring minerals from the raw
material is essential to obtain high quality.
Norsk Mineral is one of the world’s biggest suppliers of raw material for GCC
production. The raw material is a coarse grained calcite marble extracted from the
Akselberg deposit situated in Brønnøysund, 250 km north of Trondheim, Norway. Here,
some 2 million tonnes of GCC raw material are produced annually by the subsidiary
Brønnøy Kalk. The beneficiation of this raw material involves purification through
reverse flotation followed by micronisation of the concentrates. The resulting fine
grained slurries (d80 ≈ 1 μm) comprise the GCC product manufactured for the paper
industry.
Graphite and silicates represent the main contaminant phases in the Akselberg (Watne
2001). These contaminants are easily separated from the calcite marble by froth
flotation, thus resulting in a white, high-quality product. However, substantial volumes
of the deposit consist of a sulphide bearing marble, where the main contaminants are
pyrite and pyrrhotite, the latter being the most abundant.
The first part of the PhD study included an attempt to identify and quantify the different
crystallographic phases of pyrrhotite since they display different magnetic, oxidative
and flotation properties. The magnetic colloid method, SEM (scanning electron
microscope) and EMP (electron micro probe) confirmed the presence of both pyrrhotite
phases. Although the accurate quantification was not achieved, the experience gained is
still documented in the thesis. Mineralogical and chemical investigations of the sulphide
bearing marble revealed that it consists of >97% calcite with ≤1.3% pyrrhotite and
pyrite grains (1 μm – 3 mm) evenly distributed in the calcite. A mineral liberation
analysis (MLA) showed that approximately 95% of the sulphides were liberated in the
<250 μm flotation feed.
The next step of the project was to test the colouring effect that different contaminants,
and especially pyrrhotite and pyrite, display on actual GCC products. Of the 14
contaminants tested, pyrrhotite, graphite and chalcopyrite displayed the most
detrimental reduction in brightness, where a contamination level as low as 0.01 wt%
compromised product quality in the finest slurry (90% <2 μm). By evaluating the colour
information in the CIEL*a*b* parameters, pyrrhotite, graphite and chalcopyrite were
shown to introduce a more bluish colour than the other contaminants tested. The contaminant particle size produced a significant effect on brightness and properties such
as inherent optical differences, grindability and smearing effects of soft mineral were
suggested as possible explanations for the variation in colouring properties of the
different contaminant phases. Furthermore, the brightness as a function of contaminant
concentration displayed a non-linear behaviour and it was shown that the system could
be linearized by using the Kubelka-Munk model. The applicability of the Kubelka-
Munk model in order to predict the GCC brightness was tested and the model was found
to be usable.
Since it was shown that sulphides significantly colour GCC products, the potential for
removing pyrrhotite and pyrite during typical amine GCC flotation in saturated CaCO3
solution was tested. Microflotation results showed that both sulphides float to a certain
degree, but that recoveries are highly dependent of pH. The overall highest recoveries
were obtained at a pH close to 8. In addition, pyrrhotite recoveries were shown to be
highly dependent on the extent of surface oxidation. The overall highest recovery of
pyrrhotite was found at pH 8 for a sample that, prior to flotation, had been exposed to
air for no more than 5 min. The flotation results showed strong correlation with zeta
potential measurements, EDTA extraction results and redox (Eh) measurements.
Finally, the flotation solution chemistry was shown to significantly influence the
sulphide recoveries. The overall highest microflotation recoveries were produced in
saturated CaCO3 solution. Significantly lower recoveries were obtained when using
CaCl2 solution or deionised water.
Bench scale flotation experiments showed that by using an amine collector
concentration ≥300 g/t, recoveries became independent of pH and oxidation level. With
lower collector concentrations, pH must be kept at 8.3-8.4 (the natural CaCO3 solution
pH) and surface oxidation must be limited.
The main findings in this PhD study have shown that sulphides undoubtedly reduce the
brightness of GCC slurries. However, sulphides can to a certain degree be floated in a
typical amine GCC flotation by controlling the operating parameters pH, surface
oxidation or collector dosage. Bench scale flotation experiments with mixtures of the
main raw material (i.e. the graphite speckled marble) and the sulphide bearing marble
SBM in the flotation feed showed that substantial amounts of sulphide bearing marble
can be blended into the raw material without compromising the brightness cut-off value.
Brønnøy Kalk would benefit from improved subdivision of production blocks based on
geological parameters as opposed to today’s regular pattern. Several recommendations
as to how the sulphide bearing marble can be detected in the mine are given. In addition,
it is highly recommended that the company performs a more detailed mineral liberation
analysis of the sulphide bearing marble in order to evaluate the potential for increased
removal of sulphides within the existing beneficiation process or in a separate
processing line.