XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1581
silicon and aluminum. One of the main methods to differ-
entiate these two minerals is to calculate the Si/Al ratio of
the two minerals, with lepidolite having a higher Si/Al ratio
than muscovite (Guggenheim, 1981 Kosakevitch, 1976).
However, XRF mappings of silicon and aluminum
have limitations for the quantification of these two mineral
phases. From these mappings, muscovite is easily identi-
fiable, with aluminum exhibiting high intensity in mus-
covite crystals and lower silicon intensity (approximately
70% and 20%, respectively) (Figure 2b, c). The use of the
Si/Al ratio helps differentiate muscovite and lepidolite.
However, it is impossible to differentiate between lepidolite
and potassium feldspar, which have similar behaviours in
µXRF mapping for these two elements (Figure 2b, c). By
using these two elemental mappings, the lepidolite content
would be incorrect, as lepidolite and potassium feldspar
are not differentiated using this approach. Additionally,
another limitation of this approach is the lightness of the
elements used for mineralogical classification. It has been
shown repeatedly that the detection and quantification of
aluminum by X-ray fluorescence methods can be problem-
atic (La Tour, 1989 Young et al., 2016). However, in the
context of an automated mineralogy method, the goal is to
develop a reliable method that allows for rapid analysis of
a large number of samples without significant modification
of data processing.
The last approach relies on the use of elemental map-
pings of interfoliar cations. The strong substitution of
potassium for other cations (Rb, Cs, etc.) in the interfoliar
site of these micas can also be a way to differentiate mus-
covite and lepidolite (Korbel et al., 2023). The mapping of
potassium and rubidium is presented in Figure 3.
These mappings demonstrate the possible identification
of three mineral phases: muscovite, where only potassium is
detected lepidolite, which shows high intensity for rubid-
ium and lower intensity for potassium (intensity of approx-
imately 60% and 80%, respectively) and finally, potassium
feldspar, which shows the opposite trend to lepidolite
Table 1. Crystallography of micas found in the Beauvoir granite
Mineral Lepidolite Muscovite
Formula K(Li
2 Al
1 )Si
4 O
10 (OH,F)
2 KAl
2 (Si
3 Al
1 )O
10 (OH)
2
Tetrahedron Si, Al Si, Al
Octahedron Si, Al, Li Si, Al
Interfoliar cation K, Rb, Cs, etc. K, Rb, Cs, etc.
Figure 2. (a) Thin section observed in cross polarized light along its (b) Si map and (c) Al map. Qz
=quartz, Ab =albite, Kfd =K-feldpar, Lpd =lepidolite et Msc =muscovite
silicon and aluminum. One of the main methods to differ-
entiate these two minerals is to calculate the Si/Al ratio of
the two minerals, with lepidolite having a higher Si/Al ratio
than muscovite (Guggenheim, 1981 Kosakevitch, 1976).
However, XRF mappings of silicon and aluminum
have limitations for the quantification of these two mineral
phases. From these mappings, muscovite is easily identi-
fiable, with aluminum exhibiting high intensity in mus-
covite crystals and lower silicon intensity (approximately
70% and 20%, respectively) (Figure 2b, c). The use of the
Si/Al ratio helps differentiate muscovite and lepidolite.
However, it is impossible to differentiate between lepidolite
and potassium feldspar, which have similar behaviours in
µXRF mapping for these two elements (Figure 2b, c). By
using these two elemental mappings, the lepidolite content
would be incorrect, as lepidolite and potassium feldspar
are not differentiated using this approach. Additionally,
another limitation of this approach is the lightness of the
elements used for mineralogical classification. It has been
shown repeatedly that the detection and quantification of
aluminum by X-ray fluorescence methods can be problem-
atic (La Tour, 1989 Young et al., 2016). However, in the
context of an automated mineralogy method, the goal is to
develop a reliable method that allows for rapid analysis of
a large number of samples without significant modification
of data processing.
The last approach relies on the use of elemental map-
pings of interfoliar cations. The strong substitution of
potassium for other cations (Rb, Cs, etc.) in the interfoliar
site of these micas can also be a way to differentiate mus-
covite and lepidolite (Korbel et al., 2023). The mapping of
potassium and rubidium is presented in Figure 3.
These mappings demonstrate the possible identification
of three mineral phases: muscovite, where only potassium is
detected lepidolite, which shows high intensity for rubid-
ium and lower intensity for potassium (intensity of approx-
imately 60% and 80%, respectively) and finally, potassium
feldspar, which shows the opposite trend to lepidolite
Table 1. Crystallography of micas found in the Beauvoir granite
Mineral Lepidolite Muscovite
Formula K(Li
2 Al
1 )Si
4 O
10 (OH,F)
2 KAl
2 (Si
3 Al
1 )O
10 (OH)
2
Tetrahedron Si, Al Si, Al
Octahedron Si, Al, Li Si, Al
Interfoliar cation K, Rb, Cs, etc. K, Rb, Cs, etc.
Figure 2. (a) Thin section observed in cross polarized light along its (b) Si map and (c) Al map. Qz
=quartz, Ab =albite, Kfd =K-feldpar, Lpd =lepidolite et Msc =muscovite