Geochronological Data of Dykes and Sills Intruding Proterozoic Sequences of the Tombador Formation, Bahia-Brazil

 

 

Battilani, G.B. ¹; Vasconcelos, P.M.²; Gomes, N.S.³; Guerra, W.J. ²

 

1. Fundação Gorceix – Nupetro, Ouro Preto–MG, Brasil. gislaine@nupetro.com.br, wjguerra@nupetro.com.br

2. Dept. of Earth Sciences,University of QueenslandBrisbane, Qld 4072 ,Australia. paulo@earth.uq.edu.au

3. Dept. de Geologia, Universidade Federal de Ouro Preto - DEGEO/UFOP, Brasil. newton@degeo.ufop.br

 

Abstract
 

Laser incremental heating ⁴⁰Ar/³⁹Ar analysis of muscovite crystals from muscovite-martite dykes and sills spatially associated with diamonds deposits in Chapada Diamantina, Bahia, Brazil, yield well defined and internally consistent plateau ages of 1.514 + 5 and 1512 + 6 Ma for these units. ⁴⁰Ar/³⁹Ar analysis of muscovite crystals from muscovite-martite dykes and sills crosscut by quartz and tourmaline veins yield reproducible plateau ages of 512 + 0.9, 516.5 + 0.8, and 516.1 + 1.6 Ma. We interpret the ca. 1515 Ma plateau ages as the age of intrusion of the magmas into wet sediments of the Tombador Formation. We interpret the ca. 480 Ma plateau ages for the Bahia samples as ages of peak metamorphic/hydrothermal alteration associated with the Brasiliano thermo-tectonic event in each region. Our results impose a minimum deposition age constraint of ca.1515 Ma for the Tombador Formation sediments. This is also the minimum age of introduction of detrital diamonds into the middle Proterozoic paleoplacer deposits.

 

Keywords: ⁴⁰Ar/³⁹Ar geochronology, Serra doEspinhaço, Tombador Formation, intrusive rocks, diamonds

 

 

Introduction

In the Chapada Diamantina area, Bahia, the muscovite-martite dykes and sills were first described by Herrgesell (1984), who compared these units with the hematitic phylitte from Minas Gerais and concluded that they have an igneous origin. The main distinction between the occurrence of these rocks in Minas Gerais and Bahia is their metamorphic grade. The Minas Gerais occurrences are metamorphosed to greenschist facies, while the Bahia samples are hosted by essentially unmetamorphosed sedimentary rocks (Battilani et al., 2000).

We will present below ⁴⁰Ar/³⁹Ar geochronology data of dykes and sills from diamond producing areas in Chapada Diamantina, Bahia. Recente research work carried out by Battilani et al. (in preparation) reports the occurrence of microdiamonds in the intrusive unit. Because the dykes and sills investigated in this study clearly crosscut the Tombador Formation, the lowermost member of the Chapada Diamantina Group, their age would provide a minimum age for this group, which is unconstrained. This is also, probable, the minimum age of introduction of detrital diamonds into the middle Proterozoic paleoplacer deposit that are founded at the Chapada Diamantina area.

All analytical procedures adoted in this work can be consulted in Battilani et al. (in preparation). We selected three samples from the Chapada Diamantina area, Bahia, for ⁴⁰Ar/³⁹Ar geochronology. One sample represents the most pristine igneous texture observed, one sample displays evidence of incipient tourmalinization and is crosscut by quartz veins and one sample displays advanced tourmaline substituion and pseudomorphose of K-feldspar.

The samples were handpicked and placed into aluminium disks together with Fish Canyon sanidine standards using the geometry illustrated in Vasconcelos et al. (2002). The disk were covered with a lid and wrapped in aluminium foil, vacuum encapsulated in a silica glass tube, and irradiated for 14 hours at the Oregon State University Triga Reactor (CLICIT Facility). After a three-month cooling period, the samples were analysed by the laser incremental heating ⁴⁰Ar/³⁹Ar method at UQ-AGES (University of Queensland Argon Geochronology in Earth Sciences) laboratory using procedures described by Vasconcelos (1999a, b) and Vasconcelos et al. (2002).

To determine the age of Tombador and the alteration of muscovite-martite dykes and sills we dated samples of martite-muscovite dykes from

Chapada Diamantina, Bahia. The only K-bearing phase in the samples analysed is muscovite; therefore, the geochronology results either reveal the age of precipitation of muscovite in these samples or the timing of re-heating of the samples beyond 350-400 °C. From the Chapada Diamantina area, Bahia, we dated samples displaying the most pristine igneous textures (GB156) and samples displaying initial stages of quartz veining and tourmaline alteration (GB86 and GB58).

Two grains from the most pristine dyke sample from Chapada Diamantina (Bahia) yield concordant plateau ages of 1512 ± 6 and 1514 ± 5 Ma (Fig. 1). The first plateau is relatively well-defined, contains more then 70% of the ³⁹Ar released, and does not display any evidence of radiogenic ⁴⁰Ar^* loss. The slightly complicated pattern at the high temperature end of the spectrum is an artefact of a poorly programmed laser-heating schedule and does not represent any meaningful geological feature. The second plateau is very well-defined, it contains more than 95 % of the ³⁹Ar released, and it yields a plateau age concordant with the plateau age for the first grain.

The 1512 ± 6 and 1514 ± 5 Ma plateau ages obtained for sample GB156, the most pristine muscovite-martite dyke crosscutting the Tombador Formation, lie between the 1748 ± 1 Ma U-Pb age (Babinsky et al. 1999) for the underlying volcanic units of the Rio dos Remédios Group and the 1140 ± 140 Ma Pb-Pb (Babinksy et al. 1993) and the 1290 ± 52 Ma Rb/Sr results obtained for the Caboclo sediments by Brito Neves et al. (1980).

Interpreting our ⁴⁰Ar/³⁹Ar results, however, requires understanding the genesis of the mafic dykes and the muscovite-martite assemblage developed in these units. Petrographic investigation clearly indicates that the muscovite-martite assemblage does not represent the original mineralogy in the samples. The magnetite crystals, possibly primary igneous minerals, are highly fractured, corroded, and partially replaced by rutile. The occurrence of rutile microcrystals surrounding the martite crystals suggests that the original oxide in these samples was Ti-rich (ilmenite or Ti-magnetite), and that it lost Ti during oxidation to hematite. Muscovite textures, where crystals vary in grain size and display an interlocking texture, also suggest a hydrothermal origin. If the muscovite crystals in these samples are not primary, when did they form? A possible answer could be that the muscovite crystals result from K-metasomatism of K-feldspar crystals in an igneous intrusion during regional metamorphism of the sedimentary sequence. This hypothesis, however, is not

 

 

 

Figure 1: ⁴⁰Ar/³⁹Ar laser incremental heating single crystals of muscovite from one sample of muscovite-martite dyke from Chapada Diamantina, Bahia. This sample preserves the original igneous textures, and yields concordant ca 1515 Ma age for two distinct muscovite crystals.

 

consistent with the observation that the enclosing arenites of the Chapada Diamantina Formation are not metamorphosed. A second explanation is that the muscovite crystals result from alteration of the intruding units during or immediately after intrusion, when localized heat flow promoted the migration of K-rich solutions derived from the wet sedimentary pile into the hot magmas penetrating along fractures and weakness planes. If this interpretation is correct, the muscovite crystallization ages should lie very close to the emplacement of the dykes, as these dykes would have intruded the sediments very early in their depositional history. If this hypothesis is correct, the ⁴⁰Ar/³⁹Ar results impose a minimum age of 1512 ± 6 /1514 ± 5 Ma for the sediments of the Tombador Formation. Alteration of the magmas during or immediately after intrusion would account for the unusual muscovite-martite assemblage and the ca. 1515 Ma age for these rocks.

If the ⁴⁰Ar/³⁹Ar results presented here truly represent the age of dyke emplacement, they provide an important constraint on the introduction of diamonds in the Proterozoic sedimentary sequence. Since the dykes only crosscut barren units, and given that only units overlying the dyke-cut barren units are diamondiferous, it is possible to conclude that diamonds were introduced into the basin at ca. 1515 Ma.

The second sample of muscovite-martite dyke analysed from Bahia (GB58) displays similar igneous textures to sample GB156 and petrographic evidence of hydrothermal alteration, where quartz and tourmaline veinlets cross-cut an otherwise pristine igneous texture (Fig. 2). These evidences indicate that the muscovite-martite dykes were subject to a second hydrothermal event, post emplacement, that significantly altered the isotopic signatures recorded in the muscovite crystals. The two grains analysed for this sample yield complex

 

 

Figure 2: ⁴⁰Ar/³⁹Ar laser incremental heating of single crystals of muscovite from the sample GB58. The

disturbed spectra for two distinct grains of sample

2237 suggest that the ⁴⁰Ar/³⁹Ar system from this

rock was perturbed but not totally reset.

 

 

age spectra, displaying climbing age patterns, interpreted as evidence of partial Ar loss or the presence of mixed generations of muscovite in the sample. Neither spectra provide interpretable geochronological information.

Three grains of muscovite clusters analysed for an intensely altered muscovite-martite dyke (sample GB86) yield relatively concordant plateau ages of 512.0 ± 0.9, 516.5 ± 0.8, and 516.1 ± 1.6 Ma (Fig. 3). The first grain, yielding the youngest plateau
 

 

 

Figure 3: ⁴⁰Ar/³⁹Ar laser incremental heating of single crystals of muscovite from sample GB86. The concordant plateau ages for three distinct grains of the sample 2240 suggest that this rock was reheated to >350-400 ^°C at ca. 512-516 Ma.

age, which is slightly discordant from the ages yielded by the other two grains, also yields the most
disturbed spectrum, with a plateau age defined by only approximately 60% of the ³⁹Ar released. The descending spectra may be interpreted as evidence of some recoil ³⁹Ar loss in the low temperature steps. The spectra and plateau ages obtained for the other two grains are very well defined and do not show any evidence of recoil ³⁹Ar loss.

This interpretation is consistent with the fact that samples displaying advanced stages of tourmalinization (sample GB86) yield relatively concordant plateau ages of 512.0 ± 0.9, 516.5 ± 0.8, and 516.1 ± 1.6 Ma, indicating almost complete resetting of the system.

Field, petrographic, geochemical, and geochronological relationships indicate that muscovite-martite dykes and sills intruding clastic sediments in the Chapada Diamantina Group (Tombador formation) formed early in the history of development of this rift basin. The intrusion of dykes and sills into wet sediments promoted the alteration of the original igneous assemblage. ⁴⁰Ar/³⁹Ar geochronology of muscovite crystals from samples preserving original igneous textures, but not mineralogy, indicate a probable intrusion age of ca. 1515 Ma. These results impose a minimum age constraint for the Tombador Formation. These results also identify the minimum age for the intruduction diamonds into the sedimentary basin.

⁴⁰Ar/³⁹Ar geochronology results for quartz-tourmaline altered muscovite-martite dykes and sills indicate partial resetting of the ⁴⁰Ar/³⁹Ar isotopic system in samples from Bahia.

 

Acknowledgements

The authors would like to acknowledge the financial support from the Brazilian Research Council (CNPq 200636/01-4) for a PhD scholarship to Gislaine Amorés Battilani; the Fundação Gorceix, for field support; and UQ-AGES for ⁴⁰Ar/³⁹Ar geochronology.

 

Referências

Babinsky, M.; Van Schmus, W.R.; Chemale Jr., F.; Brito Neves, B.B.; Rocha, A.J.D., 1993. Idade isocrônica Pb/Pb em rochas carbonáticas da Formação Caboclo em Morro do Chapéu, BA. In: SBG (Editor), Simpósio sobre o Cráton do São Francisco, Salvador, pp. 160-163.

Babinsky, M.; Pedreira, A.J.; Brito Neves, B.B.; Van Schmus, W.R. 1999. Contribuição à geocronologia da Chapada Diamantina. In: CPRM, SNET, VII, Lençóis. p. 118-120.

Battilani, G.A.; Vasconcelos, P.M.: Gomes, N.S.; Guerra, W.J. (no prelo). ⁴⁰Ar/³⁹Ar Geochronological Constraints on Dykes and Sills Intruding Diamondiferous Proterozoic Sequences, Eastern Brazil. (submitted to Precambrian Research).

Battilani, G.A; Varajão, A.F.D; Gomes, N.S. 2000. Metamorphic degree variations in Proterozoic sandstones of the Tombador Formation, Bahia State, Brazil. Zbl Geol Paläont 1(7-8): 917-926.

Brito Neves, B.B.; Cordani, U.G.; Torquato, J.R.F. 1980. Evolução geocronológica do Pré-Cambriano do Estado da Bahia. In: H.A.V. INDA and F.B. DUARTE (Editors), Geologia e Recursos Minerais do Estado da Bahia; Textos Básicos. SME/CPM, Salvador, pp. 1-101.

Herrgesell, G. 1984. Das Prakambrium am Ostrand der Nordlichen Serra do Cipó (Serra do Espinhaço, Minas Gerais, Brasilien), Albert-Ludwigs, Freiburg, Ph-D Thesis. 247 pp.

Vasconcelos, P.M. 1999a. 40Ar/39Ar Geochronology of Supergene processes in Ore Deposits. In: D.D. Lambert and J. Ruiz (Editors), Application of Radiogenic isotopes to ore deposit research and exploration. Reviews in Economic Geology. Society of Economic Geologists, INC, pp. 73-113.

Vasconcelos, P.M. 1999b. K-Ar and Ar-40/Ar-39 geochronology of weathering processes. Annual Review of Earth And Planetary Sciences, 27: 183-229.

Vasconcelos, P.M.; Onoe, A.T.; Kawashita, K.; Soares, A.J.; Teixeira, W. 2002. Ar-40/-Ar-39 geochronology at the Instituto de Geociencias, USP: instrumentation,  analytical procedures, and calibration. Anais da Academia Brasileira de Ciências, 74(2): 297-342.