Structural and Chemical Characteristic of Tourmaline, and Mineralogy of Associated Micas from Tourmaline Bearing Quartzite of Kombé II (Bafia Group, Central Africa Fold Belt); Implication on the Metamorphic Conditions ()
1. Introduction
Tourmaline is an accessory mineral usually present in a large variety of rocks: sedimentary, metamorphic, plutonic rocks and their hydrothermal aureoles. The wide range of possibility of tourmaline appearance makes it a good candidate as a petrogenetic indicator ( [1] [2] [3]). Despite this ubiquity of tourmaline, tourmaline rich rocks or tourmalinites are very rare. It is even suggested that rocks with more than 15 vol% tourmaline are enigmatic ( [4]).
In the Bafia Group of the Cameroonian portion of the Central African Fold Belt (CAFB), tourmaline has been described as an accessory mineral in metamorphic rocks ( [5] [6] [7]). Tourmaline bearing quartzite has been mentioned for the first time in Kombé II by [5] Ganwa (1998). In this paper focusing on the tourmaline bearing quartzite, we refine the structure of the tourmaline variety dravite, report the chemical composition thereof and of associated micas (muscovite and biotite), in order to determine qualities of tourmaline crystal and its implications on metamorphic conditions.
2. Geological Setting
The Kombé II area belongs to the Bafia group which is considered as the southernmost termination of the Adamawa-Yadé Domain (ADY) of the Central Africa Fold Belt (CAFB) in Cameroon ( [8] [9]) and which has been recently considered as a piece of Archaean/Palaeoproterozoic crust detached from the Congo craton in the early Neoproterozoic times ( [10]) (Figure 1). It has been described as a Neoproterozoic immature sedimentary sequence, with detritus of various sources (Archean, Paleoproterozoic, Mesoproterozoic, [8] [9]) associated with orthogneisses of Paleoproterpzoic ages ( [11] [12]). The Bafia area is made up of metamorphic rocks, mainly gneisses, amphibolites, micaschists and quartzites ( [5] [6]). Gneisses have various compositions (amphibole-biotite gneiss, garnet-biotite gneiss, biotite gneiss) with intercalation of amphibolites, micaschists and quartzites. Formerly, the Bafia Group was known under the name of Bafia series and formed with the Yaoundé series the Groupe of Yaoundé ( [13]). Since the studies carried out by [5] [6] [8] [9] [14], Bafia series is denominated as a specific litho-structural group as a whole. The particularity of the Bafia group with respect to the other areas of the ADY in Cameroon is the NNE-SSW stretching structures, regionally highlight by hill chains with ridge lines underlain by quartzite, and the occurrence of Pan-African meta-plutonites ( [5] [7] [9]). Quartzites are either pure quartzites or minerals bearing quartzites such as two micas quartzite, muscovite quartzite, garnet quartzite and tourmaline quartzite ( [5]). Pure quartzite is located at the top of the hills (Figure 2) where it is interleaved with mica-schist. Due to differential weathering huge blocks of
Figure 1. (a): Position of the Bafia area in the Pan-African fold belt of Cameroon. 1: Pan-African domain; 2: Northern edge of Congo craton; CL: Cameroon line; SF: Sanaga Fault; CCSZ: “Central Cameroonian Shear Zone”. (b): Geological sketch map of the Bafia area (modified after Weecksteen (1957), Dumort (1968)) showing the KombéII area. 1: Tertiary volcanism; 2: Cretaceous sediments; 3: Granite; 4: Mica-schist and quartzite; 5: Undifferentiated gneisses; 6: Amphibolites, pyroxenites; 7: Granulites, 9: Mylonite; 10: Strike and dip; 11: Tectonic line, 12 Faults.
quartzite are not in place, some of which are fold hinges (Figure 3(a)); sometime, blocks of quartzite are supported by small portions of mica-schist, forming a type of “hoodoo” structure (chéminée de fée) (Figure 3(b)). Quartzite rich in tourmaline has been reported for the first time in the Bafia group to the East of the Kombé II village at Lilpagang by Ganwa [5] (Figure 2).
3. Analytical Methods
The tourmaline bearing quartzites were investigated by optical polarizing microscopy and back-scattered electron (BSE) imaging using a Fei INSPECT S50 (Department of Lithospheric Research, University of Vienna). The mineral chemistry was established using a CAMECA SX-100 electron microprobe (Department of Lithospheric Research, University of Vienna). The conditions of operation were: 15 kV accelerating voltage, 20 nA beam current, 20 s counting time on peak position, and a PAP correction procedure for data reduction. Analyses were carried out with a defocused 5 μm electron beam, minimizing the loss of Na and K. Calibration was based on the following standards: quartz (Si), corundum (Al) albite (Na), olivine (Mg), almandine (Fe), wollastonite (Ca), rutile (Ti), spessartine (Mn), orthoclase (K), Mg-chromite (Cr) and Ni-oxide (Ni).
For the crystallographic characterization tourmaline was mounted on a Bruker Apex CCD diffractometer equipped with graphite-monochromated MoKa radiation at the “Institut für Mineralogie und Kristallographie, Geozentrum,
Figure 2. Geological map of the Kombé II area ( [5]), showing the location of the tourmaline bearing quartzite (Black star) 1: Quartzite; 2: Garnet biotite gneiss with intercalation of amphibolite and garnet amphibole gneiss; 3: Biotite muscovite gneiss; 4: Trajectory of S1 foliation; 5: Strike and dip of S1; 6: L2 lineation; 7: Axis of F2 folds; 8: Trajectory of thrusting; 9: Fractures; 10: Rivers.
Universität Wien. Redundant data were collected for an approximate sphere of reciprocal space, and were integrated and corrected for Lorentz and polarization factors using the Bruker program SaintPlus (Bruker AXS Inc. 2001). The structure was refined using tourmaline starting models and the Bruker SHELXTL v 6.1 program package, with neutral-atom scanning factors and terms for anomalous dispersion. The structure refinement was performed with anisotropic thermal parameters for all non-hydrogen atoms.
4. Petrography of the Quartzite
At Lilpagang, quartzite outcrops are found to the South East of Kombé II village in the form of a cliff (Figure 4). This cliff is made up of decimetric to metric
Figure 3. (a) Bloc of quartzite not in place, showing a fold hinge; (b) “hoodoo” structure forming by differential weathering between quartzite and micaschist.
Figure 4. Cliff at Lilpagang forest, made up of migmatitic biotite gneiss with intercalation of tourmaline bearing quartzite layers.
layers of tourmaline bearing quartzite, interleaved with migmatitic biotite gneiss. Quartzite and gneiss underwent ductile deformation with an S1 foliation; in the quartzite, the S1 foliation is marked by milimetric thick muscovite rich layers, which alternate with 2 to 5 cm thick quartz rich layers. Tourmaline crystals are disseminated between quartz in the quartz rich layers (Figure 5(a)) and also associated with muscovite in the thin mica rich layers (Figure 5(b)).
The whole rock geochemistry of two samples (Table 1) shows that tourmaline bearing quartzite is characterized by 72.04 - 73.99 wt% of SiO2, 11.29 - 11.86 wt% Al2O3, 3.53 - 6.08 wt% Fe2O3, 3.49 - 5.78 wt% K2O, 2.12 - 2.14 wt% Na2O; MgO and CaO are <2 wt%, while TiO2, MnO and P2O5 are <1 wt%. Trace elements Ba, Cr, Nb, Ni, Rb, Sr, V, Y, and Zr are present. Under the microscope, the quartzite layers are granoblastic and heterogranular in texture, made up of quartz (ca. 81
Figure 5. Tourmaline crystals disseminated in the quartz rich layer (a), and S1 Foliation surface of the tourmaline bearing quartzite showing association of tourmaline and muscovite (b).
Table 1. Whole rock chemical composition of tourmaline bearing quartzite.
vol%), muscovite (ca. 13 vol%), tourmaline (ca. 4 vol%), biotite (ca. 1.5 vol%) and chlorite (ca. 0.5 vol%). Quartz forms xenomorphic and elongated crystals of 1.9 mm length in average. Quartz often portrays undulatory extinction. Muscovite forms small anhedral crystals dispersed between quartz grains in the quartzite layers or large flake associated to biotite and tourmaline in the interlayers. Biotite flakes are smaller in size than muscovite. Tourmaline is subhedral with variable grain sizes up to 1 mm long. Small crystals are dispersed between quartz in the quartzite layers whilst large crystals are associated with muscovite and biotite in the thin micaceous layers. One should note that under microscope, tourmaline, biotite, muscovite and quartz show sharp contact between minerals; they form an assemblage of the same generation with perfect equilibrated texture (Figure 6 and Figure 7).
Figure 6. BSE image of the association tourmaline, muscovite, biotite, quartz, showing the data points (green dots) of the EMPA analyses. Mus: muscovite, Bt: biotite, Tur: tourmaline, Qtz; quartz. Note the crystal chemical zonation in tourmaline.
Figure 7. BSE image of the association tourmaline, muscovite, quartz, showing the data points (green dots) of the EMPA analyses. Mus: muscovite, Tur: tourmaline, Qtz; quartz.
5. Structure Refinement of Tourmaline
Structure refinements of tourmaline crystals give following values: X(Na0.62Ca0.16K0.010.21) Y(Mg1.20Al1.20Fe2+0.54Ti4+0.06) Z(Al5.10Mg0.90) (BO3)3 T[Si5.98Al0.02O18] (OH)3[(OH)0.54O0.34F0.12]; this structure is characterized by the following lattice parameters: a = 15.946(1) Å, c = 7.197(1) Å , R1 = 1.24%. wR2 = 3.53%, <X-O> = 2.681 Å, <Y-O> = 2.025 Å, <Z-O> = 1.927 Å, <T-O> = 1.621 Å,
= 1.374 Å.
The studied tourmaline belongs to Fe-bearing dravite. The approximately formula is X(Na0.95[]0.05)Y(Mg2.39Fe0.61)Z(Al5.10Mg0.90)(BO3)3T[Si6O18](OH)3[(O,OH)0.88F0.12]. There is only a small F content and a minor vacancy (0.5 apfu) at the X site. The F content as calculated from the structure refinement is around 0.23 wt%. This is in agreement with the chemical composition of the mineral (cf. Table 2) which shows nil values for the F and a value of X site vacancy less than 0.3. The Mg at Z site includes also Al-The Fe is mainly Fe2+ and some Fe3+. There is a significant disorder of Al-Mg between the Y and the Z site. There might be a tiny amount of Al at the T site. This is corroborated by the chemical analyses (cf. table) all comprises between 0.007 and 0.138 in the T site.
6. Mineralogy
6.1. Tourmaline
In BSE images tourmalines show zonation patterns that are not well organized (cf. Figure 6). This optical zonation is also reflected in the chemical composition of the crystal. Considering the crystal Tur17 of the Figure 6, one notes an increase of the Mg/(Fe + Mg) ratio from the core (data set point 45, Table 2) with a value of 0.794 to the rim where one has values of 0.805 (data set point 48, rim in contact with quartz) and 0.806 (data set point 50, rim in contact with biotite). The chemical composition at the rim of tourmaline crystal seems not to have been influenced by neighboring minerals as can be seen in the Figure 7. In fact, the dataset point 4 in contact with muscovite has a Mg/(Fe + Mg) ratio of 0.800 (Table 2) close similar to the ratio of data point set 3 (0.798) in contact with quartz (Table 2). Nevertheless, the Fe ratios vary in a very narrow interval regardless the position of the data set point in a crystal.
Principal constituents at the X-site show that the tourmalines crystals belong to the alkali subgroup (Figure 8(a)) of [16] Hawthorne and Henry (1999). In the corresponding diagram, two clusters of data set can be observed with XCa less (bounded by grey rectangle) or more than 0.124. No discrimination can be observed with respect to the position of data points in the mineral grains. Inside the grey rectangle for instance, are data points from the core (7, 11, 18, 38, 45; table), data points from the rim (20; Table 2) and data points from intermediate domain (1, 32, 39, 42, 46, 50; table). The same observation can be made in the rest of the data. It appears that no simple zonation can be seen in respect to the position of the data point in the crystal.
Table 2. Chemical composition of tourmaline.
The studied tourmaline crystals classify as dravite according to the classification of [15] Henry et al. 2002 (Figure 8(b)). In the classification diagram, representative points are aligned parallel to the Y axis. This arrangement reflects the variation of the vacancy + Na content (cf Table 2) and a constant Mg/(Mg + Fe) ratio of about 0.8. In the Al-Fe(tot)-Mg ternary diagram ( [2]), the tourmalines plot in both fields of metapelites and metapsamite coexisting with Al-saturating phase, and metapelites without an Al-saturating phase (Figure 9(a)). The Al-rich phase in the hosting tourmaline bearing quartzite may be the muscovite the chemistry of which shows more than 32 wt% of Al2O5. In the Ca-Fe(tot)-Mg diagram the representing points are in the field of Ca-poor metapelites, metapsammites and quartz-tourmaline rocks (Figure 9(b)). The Ca-poor character of the host rock can be seen through its mineralogical composition made up of quartz, biotite (CaO ≤ 0.008 wt%), muscovite (CaO ≤ 0.015 wt%) tourmaline (CaO ≤ 1.35 wt%).
6.2. Biotite
Chemical composition of the analyzed biotite flakes are shown in Table 3; biotite
Figure 8. Classification of studied tourmaline. (a) Principal tourmaline subgroups diagram of Hawthorne and Henry ( [16]): major compositional groups are defined by the principal constituents at X-site. (b) Diagram Mg/(Mg + Fe) vs /(X-vacancy/(X-vacancy + Na), according to the classification of [15].
is very poor in calcium (virtually Ca-free or equal to 0.001 apfu); There are no opaque minerals in the biotite cleavages, showing that the analyzed minerals are not affected by chloritization. The Fe/(Fe + Mg) is around 0.3 at the transition between phlogopite and biotite of [18] (Figure 10). In the ternary compositional
Figure 9. Tourmaline environmental diagrams of Henry and Guidotti ( [2]); (a) Al-Fe(tot)-Mg diagram with numbered fields corresponding to the following rock types: (1) Li-rich granitoidpegmatites and aplites, (2) Li-poor granitoids and associated pegmatites and aplites, (3) Fe3+-rich quartz-tourmaline rocks (hydrothermally altered granites), (4) Metapelites and metapsammite coexisting with an Al-saturating phase, (5) Metapelites without an Al-saturating phase, (6) Fe3+-rich quartz-tourmaline rocks, calc-silicate rocks, and metapelites, (7) Low-Ca meta-ultramafics and Cr−, V-rich metasediments, and (8) Meta-carbonates and meta-pyroxenites, (b) Ca-Fe(tot)-Mg diagram with the numbered fields corresponding to the following rock types: (1) Li-rich granitoidpegmatites and aplites, (2) Li-poor granitoids and associated pegmatites and aplites, (3) Ca-rich metapelites and calc-silicate rocks, (4) Ca-poor metapelites, metapsammites and quartz-tourmaline rocks, (5) Meta-carbonates, and (6) Meta-ultramafics.
Table 3. Chemical composition of biotite.
Figure 10. Classification of biotite based on [18].
system of (Alvi + Fe3+)–Mg–(Fe2+ + Mn) after [19], biotite classifies as phlogopite (Figure 11). In the FeOtMgOAl2O3 ternary diagram (Figure 12), representative points plot in the field of biotite unaccompanied by other mafic minerals [17].
6.3. Muscovite
The muscovite flakes (Table 4) of the tourmaline bearing quartzite exhibit the following chemical characteristic: total Al: 5.019 - 5.413 apfu, Ti: 0.072 - 0.104 apfu, Na: 0.124 - 0.175 apfu, Si: 6.136 - 6.366 apfu, Mg: 0.301 - 0.476 apfu, Mn and Cr contents are negligible. In the Mg-Ti-Na ternary diagram ( [22]) the muscovites plot mainly in the field of primary muscovite. Only one data point is in the field of secondary muscovite, while few points are on the limit between the two fields (Figure 13). In the (Fe2+ + Mg) versus (Si – 6) apfu diagram (Figure 14) muscovites are above the muscovite-phengite solid solution. Such muscovite is classified as slightly phengitic celadonitic muscovite.
7. Discussion
The Bafia group to which belongs the Kombé II area, is considered as the southern part of the Adamawa-Yadé domain of the Pan-African Central Africa Fold Belt in Cameroon ( [6] [23]). It is a meta-volcano sedimentary sequence, made up of gneisses, amphibolites and quartzites ( [8] [10]). In the study area, dravite is present notably only in the quartzite layer intercalated with gneisses at the bottom of a hill at Lilpagang, while quartzites at the top of the hills are pure or muscovite-rich quartzites. The occurrence of dravite in that layer means that required conditions are met to allow the crystallization of the mineral. It means that the sedimentary protholite was sufficiently rich in alumina to allow crystallization of muscovite, and rich in boron to favour the growth of dravite. Differences of crystals size of tourmaline in the quartzite plate and in the thin muscovite rich layers can be explained by the fact that the platy micas can accommodate larger crystals than the more granular quartz. Even though the origin of the
Figure 11. Plot of biotite on Foster’s classification diagram [19].
Figure 12. FeOtMgOAl2O3 ternary plot to discriminate biotite association with mineral phases ( [20]). Metamorphic and igneous fields are from [21]. I is the field of biotite associate with muscovite or topaz; II is biotite unaccompanied by other mafic minerals; and III biotite associated to amphibole, pyroxene or olivine.
Figure 13. Ternary Mg–Ti–Na diagram for muscovite with primary and secondary fields from [22].
Figure 14. (Fe2+ + Mg) versus Si – 6 (apfu) diagram for the muscovite of the tourmaline bearing quartzite. The black circle is the phengite position whereas the ideal muscovite is positioned at the origin of the diagram: Fe2+: total Fe (Fe2+ + Fe3+).
boron is not investigated, it can be assumed that the element is transported by a fluid. This fluid can be a product of devolatisation during metamorphic process, fluid hosted in that level of the sedimentary sequence, or fluid of external origin. Furthermore, Mg/(Mg + Fe) ratios in the studied dravite rage from 0.784 to
Table 4. Chemical composition of muscovite.
0.805 (cf Table 2), which is in the range of the value (0.4 - 1.0) determine by [24] for metamorphic tourmaline. The metamorphic origin of tourmaline is confirmed by its presence in the metamorphic mineral assemblage biotite-muscovite-tourmaline-quartz) showing a perfect equilibrated metamorphic texture (cf Figure 6 and Figure 7). References [25] and [26] used FeO/(FeO + MgO) ration in tourmaline as indicator of source environments. As demonstrated by many authors working on tourmaline ( [27] [28]) the composition of the studied dravite seems to be influenced by the composition of the host rock. This can be deduced from the mineralogy of the tourmaline bearing quartzite made up for example of Ca poor minerals (biotite, muscovite, quartz), hosting dravite with Ca content equal or less than 1.35 weigh percent. The values for vacancy + Na vary from 0.900 to 0.771. Na varies from 0.699 apfu to 0.518 apfu with the corresponding vacancy of 0.202 apfu and 0.371 apfu. Even though 0.202 is not the lowest value of the vacancy, globally, the Na apfu value decreases with the increase of the vacancy. The variation of the vacancy +Na could be link with the optical zonation of the tourmaline crystals; as one can see in Figure 6, data points 45 to 50 are localized in different domains of the tourmaline crystal and they show different values of Na apfu and vacancy (cf Table 2).
The structure of the dravites shows a low vacancy at the X site, which militates for a crystallization of the tourmaline at a high temperature of more than 750˚C. One should note that the layer of tourmaline bearing quartzite is intercalated with the biotite gneisses which show partial melting phenomena ( [5]). This is in good agreement with the above temperature. Ganwa ( [5]) shows that metamorphic path in the area reaches a possible temperature peak at 825˚C. Nevertheless, even though the presence of Al in the T site and the low F content are not favorable for the high temperature of crystallization of the tourmaline, it is likely that the studied tourmaline crystallized at about 750˚C. It is likely that tourmaline start to crystallize at the end of the granulite facies overprinting, especially during the anatexis in the adjoining gneisses. This temperature should be favorable for the production and expulsion of boron rich metamorphic fluid in the quartzite layer, or the circulation of and external boron rich fluid in the quartzite layer. The present study should lead to metallogenic research in the study area. The R1 value of 1.24% means that the crystal structure of the tourmalines is of high quality. It has been shown that gold mineralization and tourmaline are genetically linked ( [29] and [30] suggest that recognition of tourmaline “should stimulate exploration interest in any area in which they occur”. It has been demonstrated in the Tsa Da Glisza prospect in Yukong territory (Canada) that tourmaline has potential as an indicator mineral for emerald mineralization ( [27]).
8. Conclusion
In addition to the quartzite underlining the hills summit in the Bafia Group, one has tourmaline quartzite as centimetric to decametric layers interleave with gneiss to the East of Kombé II. Studied tourmaline is belonging to the alkali subgroup, especially Fe-bearing dravite, and is associated to phlogopite and phengitic celadonitic muscovite. These crystals are of high quality, forming at high temperature during the migmatisation of the gneiss.
Acknowledgements
The first author is great full to the association of youth of Kombé II, and especially to Bassinha Bonaventure, a hunter who guided the first author to outcrops in the forest. Laboratory analyses related to the present work have been carried out during the post-doctoral research of the first author at the University of Vienna, funded by the Austrian Science Fund FWF, project M 1371-N19. We thank Claudia, Franz Biedermann of the University of Vienna for technical assistance.