Chemical zoning of muscovite megacrystal from the Brazilian Pegmatite Province

EARTH SCIENCES

Chemical zoning of muscovite megacrystal from the Brazilian Pegmatite Province

Rúbia R. Viana^I; Hanna J. Evangelista^II; Willem B. Stern^III

^ICoordenação do Programa de Pós-Graduação em Geociências, Departamento de Recursos Minerais, ICET, Universidade Federal de Mato Grosso, Av. Fernando Correa da Costa, s/n, Coxipó, Campus Universitário, 78068-900 Cuiabá, MT, Brasil

^IIDepartamento de Geologia, Universidade Federal de Ouro Preto, Morro do Cruzeiro, 35400-000 Ouro Preto, Minas Gerais, Brasil

^IIIMineralogisch-Petrographisches Institut, University of Basel, Basel, Switzerland

^{Correspondence to}

ABSTRACT

Macroscopically homogenous muscovite plate from the Cruzeiro pegmatite, located in the Eastern Pegmatite Province in Minas Gerais, may show complex distribution patterns of some trace elements. In geochronological and petrological studies, as for example in the distinction of magmatic and post-magmatic mica, the cause of zoning could be taken into consideration. The complex chemical zoning in the studied mica plate can be best explained by growth in an evolving magma followed by alteration due to percolation of hydrothermal fluids. Enrichment of Rb towards the border is interpreted as resulting from the chemical evolution of the residual magma during crystal growth. The depletion in (^IVAl+^VIAl) as well as the increase in (Fe+Mg) and Si along a fracture could be due to the hydrothermal celadonitic substitution of muscovite. This alteration also caused depletion in the contents of Rb, Ga, Y, Nb, Sn, and Zn and residual concentration of Ti. Elements such as Ga, Y, Nb, Sn, and Zn, rarely considered in the discussion of differentiation or alteration processes in micas, have been shown to be as significant as the alkali-elements.

Key words: muscovite, chemical zoning, hydrothermal alteration, Cruzeiro pegmatite.

RESUMO

Um grande cristal de muscovita, macroscopicamente homogêneo, procedente do Pegmatito Cruzeiro, localizado na Província Pegmatítica Oriental, em Minas Gerais, exibe padrão de distribuição complexa para alguns elementos traços. Em estudos geocronológicos e petrológicos, como, por exemplo, na separação entre micas magmáticas e pós-magmáticas, a causa de zoneamento deve ser levada em consideração. O complexo zoneamento químico no cristal de mica estudado é melhor explicado pelo crescimento em um magma evoluído, seguido pela alteração, proveniente da percolação de fluidos hidrotermais. O enriquecimento de Rb nas bordas é interpretado como resultado da evolução química do magma residual durante o crescimento do cristal. A diminuição em (^IVAl+^VIAl), bem como o aumento de (Fe+Mg) e Si ao longo da fratura é explicado pela substituição hidrotermal celadonítica da muscovita. A alteração hidrotermal causou, também, a diminuição nos conteúdos de Rb, Ga, Y, Nb, Sn e Zn ao longo desta fratura, além da concentração residual de Ti. Elementos tais como, Ga, Y, Nb, Sn, e Zn, pouco considerados em discussão de diferenciação ou processos de alteração, mostraram significância tanto quanto os elementos alcalinos.

Palavras-chave: muscovita, zoneamento químico, alteração hidrotermal, pegmatito Cruzeiro.

INTRODUCTION

Two kinds of zoning were found in micas from pegmatites from the gem-producing Eastern Brazilian Pegmatite Province (EBPP): discontinuous zoning, due to the overgrowth of muscovite on biotite as described by Viana et al.(2003), and chemical zoning of crystals of asingle type of mica, investigated in the present paper.

The Cruzeiro pegmatite is located in Governador Valadares region, Minas Gerais State, Brazil (Fig. 1). This pegmatite is composed of three subvertical dikes reaching a thickness of 50 meters.The pegmatite ishosted by quartzite of the Serra da Safira Sequence (Federico et al.1998). It shows well defined internal zonation, being composed of quartz, feldspar, muscovite, gem-tourmaline and less commonly beryl (aquamarine variety), garnet, niobotantalate, spodumene, and rarely amblygonite (Bilal et al. 2000, Cassedane et al. 1980).

Muscovite of variable size is common in all pegmatites from the EBPP, mostly as pseudo-hexagonal or fishtail shaped books. The larger crystals are found in the intermediate zone of Cruzeiro Pegmatite, in some cases exceeding 30 cm. A large muscovite plate from this zone was submitted to detailed chemical analysis in order to examine it compositional zoning and to understand the growth patterns of giant crystals. A thin fracture line crosscutting the plate from one border to the other was the only heterogeneity observed.

The chemical variation within the large muscovite plate was investigated by special analytical procedures, enabling the detection of trace elements with a higher degree of accuracy than by conventional electron microprobe analysis.

GEOLOGICAL SETTING

A large quantity and variety of gemstones, particularly aquamarine and tourmaline, is produced in the Eastern Brazilian Pegmatite Province (EBPP), which comprises an area about 800 km long and 150 km wide. The pegmatites are spread over eastern Minas Gerais, western Espírito Santo and southern Bahia States (Fig. 1). The EBPP is characterized by a particular geotectonic setting in a Neoproterozoic-Cambrian orogenic belt generated during the Brasiliano-Pan-African cycle, which consisted of a set of orogenies that lasted from about 850 to 550 Ma (e.g., Oliveira et al.1997, Pinto and Pedrosa-Soares 2001). The majority of the pegmatites of the EBPP are related to granite intrusions into the Brasiliano mobile belts generated during the consolidation of the Gondwana supercontinent. Biotite and its muscovite overgrowth from the Ipê pegmatite, located near Governador Valadares (Fig. 1), have been dated by the K/Ar method enabling to establish a crystallizationage of 575Ma as well as a cooling rate of 3.3°C for the pegmatite (Viana et al. 2003).

SAMPLING AND EXPERIMENTAL METHODS

An 80 ×40 cm mica plate from the intermediate zone ofthe Cruzeiro pegmatite was selected for analysis.Mica discs were cut out on 56 points of the muscovite plate. An even, clean crystal surface of 30 mm diameter was analyzed in the Geochemical Laboratory Institute of Mineralogy and Petrography of Basel University (Switzerland) by means of X-ray Fluorescence Analysis - XFA, (SRS-3400 spectrometer of Siemens-Bruker-AXS, Germany, Specplus software) without further preparation using the analytical routine discussed by Stern (2001). No grinding process took place, and hence no contamination (W,Co) induced by sample dressing.No elements were lost in this way due to volatilization (F, Cl) as is the case with certain conventional preparationtechniques such as vitrification (glass beads). The specimen thickness was measured by means of a micrometer and used for thickness correction, essential when high-energy spectral lines are used for analysis.

Since analytical reliability decreases with decreasing concentrations (Fig. 2), elements which are present mostly well above the detection limit were preferably selected for analysis in this study.

The results of the chemical analyses reported in Table I show that the main composition changes little, that is, the chemical variation of the major elements is within the analytical frame of detection. However, some trace elements vary with a factor 2 to 3, which reflects indeed the changing environment during crystal growth.

PETROLOGICAL CONSIDERATIONS TO ELEMENT DISTRIBUTION PATTERNS

The locations of the analyzed points and the distribution maps of the compositional contour lines for selected elements are presented in Figure 3. The distribution maps show very similar patterns, although the trends might be the opposite, as for example for Ti and Ga (Fig. 3). Thesimilar distribution of the contour lines is an indication that the chemical zoning is due to the operation of some kind of geological process, i.e., it is not the result of imprecise analytical data that would certainly result in chaotic distribution patterns.

In order to evaluate the element distribution maps within the muscovite plate, the following geological processes are potential causes for the chemical variation in pegmatite minerals:

1 – The chemical evolution of residual magma during the crystallization of the pegmatite magma is responsible for an increase of Rb and other alkali-elements and a decrease of the K/Rb-ratio during fractionation (e.g. Morteani et al. 1995). Consequently, there is an enrichment of Rb and depletion of K/Rb-ratio from center to border of pegmatite minerals such as mica during growth. In melts remaining during fractionated crystallization there is an increase in Rb and decrease in Sr and Ba (Neiva et al. 1987).

2 – Zoning due to hydrothermal alteration and/orgrowth.In this case it is supposed that the produce enrichment in the celadonitic component that leads to increase Si and (Fe+Mg) and diminish (^IVAl+^VIAl) when compared to magmatic muscovite (Gomes and Neiva 2000, Demster et al. 1994). Hydrothermal muscovite is also supposed to be poorer in Nb and Ta (Neiva 1987).

3 – Coalescence of growing crystals by synneusis (Roycroft 1991), which should generate independentzoning in each of the attached crystals surrounded by late stage zones joining the various parts (Shelley 1993). Corrosion followed by later growth of normal euhedral faces is also possible (Roycroft 1989, 1991).

RESULTS AND DISCUSSION

The following discussion about the chemical variation within the studied mica plate takes into account the above mentioned possible causes of zoning in pegmatite minerals.

Analyzing the selected bivariant diagrams (Fig. 4) it is possible to observe that there are a clear negative correlation between Ti, Mg and Fe and positive correlation between Nb, Rb, Sn, Zn, Ga, Y and Mn, both versus the Al content. In Figure 4, is also shown a positive correlation of Ti content in function of the Mg content. On the other hand, the Fe content versus both Mg and Si present two different trends, negative for lower and positive for higher Mg and Si contents. This tendency can be related to presence of Fe²⁺ and Fe³⁺ in the samples. No correlation was observed with K content, meaning that K/Rb ratio is controlled only by Rb.

Distribution patterns of indicator elements such as Rb and K/Rb in the studied megacrystal is unrelated to crystal center-border geometry (Fig. 3).Rb shows agradual increase towards the upper border and a left to right-oriented low crosscutting the lower half of thecrystal, which is more or less coincident with the fracture line. Elements showing similar trends to Rb are Ga, Y, Nb, Sn, Zn, Na, and (^IVAl+^VIAl). K/Rb ratio, on the other hand, shows an opposite trend, which is also found for Ti and (Fe+Mg) and less pronounced also for Si. Other elements, such as Ba, K, Pb, F, not shown in Figure 3, have erratic distribution patterns, while Ta and Sr, which are elements considered of petrological importance, present contents lower than the limit of detection.

Considering the mentioned variation patterns,growth during evolution of residual magma could not be the only cause of zoning, because chemical variation of diagnostic elements such as Rb, which should graduallyincrease from center to border, shows distribution patterns unrelated to crystal geometric contour. However, the much higher Rb-contents found along the upper border (Fig. 3) suggest that the influence of growth during magmatic differentiation cannot be excluded.

Coalescence of several crystals by synneusis is improbable because the various areas corresponding to the old crystals that should present similar concentric distribution patterns cannot be recognized. The observed lower contents in Rb, Ga, Y, Nb, Ta, Sn, Zn, Na, and (^IVAl+^VIAl) as well as the higher values of K/Rb, Ti, Si, and (Fe+Mg) along a more or less E-W oriented stripe in the southern half of the plate seem to be related to the fracture line (see Fig. 3).

The impoverishment in (^IVAl+^VIAl) and the in-crease in (Fe+Mg) and Si near to the fracture region could be dueto the celadonitic substitution of muscovite as discussed by Demster (1992) and Gomes and Neiva (2000):

This substitution is considered to be characteristic of hydrothermal alteration/growth (Demster et al. 1994, Gomes and Neiva 2000). In the studied case alteration could have been caused by hydrothermal fluids circulating along the fracture, implying in a deformational event of late Brasiliano age, following the crystallization of the pegmatite. Elements that were also depleted during the alteration include Rb, Ga, Y, Nb, Sn, Zn, Na. Ti-contents, on the other hand, may have been increased due to (i) residual concentration because of the immobile character of Ti or (ii) reaction/alteration of other phases, e.g. rutile needles included in muscovite.

The similarity of the trends of Ga, Y, Nb, Sn, and Znto the trend of Rb, which is considered as being animportant petrological indicator, demonstrates that incase of the studied mica these elements show the same geochemical behavior as Rb. Therefore elements such asGa, that shows significant variation, can also be valuable indicators for discrimination of different mica generations in rocks.

ACKNOWLEDGMENTS

This work was partially supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de Amparo a Pesquisas do Estado de Minas Gerais (FAPEMIG) (Brazil) and by the Intra-muros funding by GeochemicalLaboratories/Basel University.

Manuscript received on May 16, 2005; accepted for publication on August 23, 2006; presented by ALCIDES N. SIAL

• BILAL E, CORREIA-NEVES JM, FUZIKAWA K, HORN AH, MARCIANO VRPRO, FERNANDES MLS, MOUTTE J, MELLO FM AND NASRAOUI M. 2000. Pegmatites in southeastern Brazil. RBG 30: 234237
• CASSEDANE JP, CASSEDANE JO AND SAUER DA. 1980. The Cruzeiro mine past and present. Mineral Rec 11: 363370.
• DEMSTER TJ. 1992. Zoning and recrystallization of pengiticmicas: implications for metamorphic equilibration. Contr Min Petr 109: 526537.
• DEMSTER TJ, TANNER PWG AND AINSWORTH P. 1994. Chemical zoning of white micas. Am Mineral 79: 536544.
• FEDERICO M, ANDREOZZI GB, LUCHESI S, GRAZIANI G AND CÉSAR-MENDES J. 1998. Compositional variation of tourmaline in the granitic pegmatite dykes of the Cruzeiro mine, Minas Gerais, Brazil. Can Mineral 36: 415431.
• GOMES MEP AND NEIVA AMR. 2000. Chemical zoning of muscovite from the Ervedosa granite, northern Portugal. Mineral Mag 64: 347358.
• MORTEANI G, PREINFALK C, SPIEGER W AND BONALUMI A. 1995. The Achala granitic complex and the pegmatites of the Sierras Pampeanas (northwest Argentina): A study of differentiation. Econ Geol 90: 636647.
• NEIVA AMR. 1987. Geochemistry of white micas from Portuguese tin and tungsten deposits. Chem Geol 63: 299317.
• NEIVA AMR, NEIVA JMC AND PARRY SJ. 1987. Geochemistry of the granitic rocks and their minerals from Serra da Estrela, Central Portugal. Geo Cosmo Acta 51: 439454.
• OLIVEIRA MJR, FÉBOLI WL AND PINTO CP. 1997. Geologia estrutural e tectônica. In: Projeto Leste. Província Pegmatítica Oriental. Programa de Levantamento Geológico Básico do Brasil, CPRM. Belo Horizonte, MG, Brasil, p. 98119.
• PEDROSA - SOARES AC AND WIEDEMAN - LEONARDOS CM. 2000. Evolution of the Araçuaí Belt and its connection to the Ribeira Belt, Eastern Brazil. In: TECTONIC EVOLUTION OF SOUTH AMERICA, 31st International Geologic Congress, Rio de Janeiro, RJ, Brazil, p. 265285.
• PINTO CP AND PEDROSA-SOARES AC. 2001. Brazilian gem provinces. Aus Gem 21: 216.
• ROYCROFT P. 1989. Zoned muscovite from the Leinster Granite, SE Ireland. Mineral Mag 53: 663665.
• ROYCROFT P. 1991. Magmatically zoned muscovite from the peraluminous two-mica granites of the Leinster batholith, southeast Ireland. Geology 19: 437440.
• SHELLEY D. 1993. Igneous and Metamorphic Rocks under the Microscope. Chapman & Hall, London, 445 p.
• STERN WB. 2001. XRF-Analysis of Geological Standard Materials: Performance of "Standardless" Measuring Routines for Major and Minor Components. ICP Inf Newsl 27: 196200.
• VIANA RR, MANTÄRI I, HENJES-KUNST F AND JORDTEVANGELISTA H. 2003. Age of pegmatites from eastern Brazil and implications of mica intergrowths on cooling rates and age calculations. J South Am Ear Sci 16: 493501.

Publication Dates

• Publication in this collection
  27 Aug 2007
• Date of issue
  Sept 2007

History

• Received
  16 May 2005
• Accepted
  23 Aug 2006