The timing of metamorphism in the Odenwald–Spessart basement, Mid-German Crystalline Zone

Abstract

New in situ electron microprobe monazite and white mica ⁴⁰Ar/³⁹Ar step heating ages support the proposition that the Odenwald–Spessart basement, Mid-German Crystalline Zone, consists of at least two distinct crustal terranes that experienced different geological histories prior to their juxtaposition. The monazite ages constrain tectonothermal events at 430 ± 43 Ma, 349 ± 14 Ma, 331 ± 16 Ma and 317 ± 12 Ma/316 ± 4 Ma, and the ⁴⁰Ar/³⁹Ar analyses provide white mica ages of 322 ± 3 Ma and 324 ± 3 Ma. Granulite-facies metamorphism occurred in the western Odenwald at c. 430 and 349 Ma, and amphibolite-facies metamorphism affected the eastern Odenwald and the central Spessart basements between c. 324 and 316 Ma. We interpret these data to indicate that the Otzberg–Michelbach Fault Zone, which separates the eastern Odenwald–Spessart basement from the Western Odenwald basement, is part of the Rheic Suture, which marks the position of a major Variscan plate boundary separating Gondwana- and Avalonia-derived crustal terranes. The age of the Carboniferous granulite-facies event in the western Odenwald overlaps with the minimum age of eclogite-facies metamorphism in the adjacent eastern Odenwald. The granulite- and eclogite-facies rocks experienced contrasting pressure–temperature paths but occur in close spatial proximity, being separated by the Rheic Suture. As high-pressure and high-temperature metamorphisms are of similar age, we interpret the Odenwald–Spessart basement as a paired metamorphic belt and propose that the adjacent high-pressure and high-temperature rocks were metamorphosed in the same subduction zone system. Juxtaposition of these rocks occurred during the final stages of the Variscan orogeny along the Rheic Suture.

Appendices

Appendix 1: ⁴⁰Ar/³⁹Ar dating

We performed ⁴⁰Ar/³⁹Ar dating at the Argonlab of the TU Bergakademie Freiberg, Germany (Pfänder et al. 2014). The micas were handpicked and ultrasonically cleaned in acetone and deionised water. After drying, they were wrapped into Al foil and loaded in 5 × 5 mm wells on 33-mm Al discs stacked together for irradiation. Cadmium-shielded neutron irradiation of samples and fluence monitors was done for 7 h in the RODEO facility of the HFR research reactor in Petten, The Netherlands. The irradiated micas were unwrapped, and 2.5–3.0 mg aliquots were loaded in 3 × 1 mm (diameter × depth) wells on an oxygen-free copper disc and transferred to the sample chamber. We performed step heating using a floating 10.6 µm CO₂ laser system with a defocused beam at 3 mm diameter, followed by gas purification applying two AP10 N getter pumps, one at room temperature and one at 400 °C. Heating time was 5 min; cleaning time was 10 min per step. Ar isotope compositions of individual temperature steps were measured in static mode using an ARGUS noble gas mass spectrometer equipped with five faraday cups and 10¹² Ohm resistors on mass positions 36–39 and a 10¹¹ Ohm resistor on mass position 40. Typical blank levels were 2.5 × 10⁻¹⁶ mol⁴⁰Ar and 8.1 × 10⁻¹⁸ mol³⁶Ar. Measurement time was 7.5 min per step acquiring 45 scans at 10-s integration time each. Mass bias was corrected assuming linear mass-dependent fractionation and using an atmospheric ⁴⁰Ar/³⁶Ar ratio of 295.5. Raw data reduction employed an in-house developed Matlab^® toolbox; inverse isochron and plateau ages were calculated using Isoplot 3.7 (Ludwig 2008). Ages were calculated against Fish Canyon Tuff sanidine (FCT) as flux monitor (28.305 ± 0.036 Ma; Renne et al. 2010). Corrections for interfering Ar isotopes were done using the ratios as given in Electronic Supplement Table S1 and applying 5 % uncertainty. Repeatedly measured HDB1 biotite, irradiated in several batches along with FCT and unknowns, yielded an average age of 24.68 Ma at an external reproducibility better 0.9 % (1σ, n = 5). Complete step heating data and intensity intercept values are presented in the Electronic Supplement Table S1.

Appendix 2: Th–U–Pb electron microprobe (EMP) monazite dating

The electron microprobe (EMP) Th–U–Pb dating is based on the observation that common Pb concentration in monazite (LREE, Th)PO₄ is negligible with respect to radiogenic Pb contents that result from the decay of Th and U (e.g. Suzuki et al. 1994; Montel et al. 1996). As Th concentrations in magmatic and metamorphic monazite are generally high, a sufficient amount of radiogenic Pb that is measureable by the EMP analysis accumulates in monazite within >100 myr (Schulz and Schüssler 2013). Thus, EMP analyses of bulk Th, U and Pb concentrations in monazite can be used for the calculation of a chemical model age and the associated error (e.g. Montel et al. 1996; Cocherie and Albarède 2001; Suzuki and Kato 2008; Spear et al. 2009).

Analyses of ThO₂, UO₂, PbO, light rare earth elements (LREE), Y₂O₅, CaO, SiO₂ and P₂O₅ concentrations in monazite grains (Electronic Supplement Table S2) were used to calculate monazite chemical ages following two different approaches. First, following the approach outlined by Montel et al. (1996), an age was calculated for each individual analysis, with the 1σ standard deviation resulting from counting statistics being in the order of c. 20–40 myr for the Palaeozoic samples (Electronic Supplement Table S2). Using these apparent age data, weighted average ages and related errors for monazite populations in the samples were then calculated using Isoplot 3.7 (Ludwig 2008) and are interpreted as the age of monazite growth or recrystallisation during metamorphism (Table 2 and Electronic Supplement Table S2). Second, ages were also determined using the ${\text{ThO}}_{2}^{*}$-PbO isochron method (the chemical isochron method ‘CHIME’ of Suzuki et al. 1994), where ${\text{ThO}}_{2}^{*}$ is the sum of the measured ThO₂ plus ThO₂ equivalent to the measured UO₂. The age is proportional to the slope of a regression line that is forced through zero in ${\text{ThO}}_{2}^{*}$ versus PbO space. The model ages obtained by the two different methods coincide exceptionally well in all samples analysed (Electronic Supplement Table S2, Fig. 4).

Th, U and Pb concentrations in monazite grains were determined for the calculation of monazite model ages, and Ca, Si, P, LREE and Y contents were measured for corrections and determination of the monazite mineral chemistry. All in situ thin section analyses were carried out with a JEOL JXA 8200 at the GeoZentrum Nordbayern, University of Erlangen-Nürnberg, Germany. The Mα1 lines of Th and Pb and the Mβ1 lines of U on a PETH crystal were selected for monazite analysis. Analytical errors of 2σ at 20 kV acceleration voltage, 100-nA beam current, 5-µm-beam diameter and counting times of 320 s (Pb Mα1), 80 s (U Mβ1) and 40 s (Th Mα1) on peak have been used for age calculations. The error on Pb concentrations typically ranges from 0.016 to 0.024 wt%. Synthetic orthophosphates from the Smithsonian Institution were used as standards for the REE analysis (Jarosewich and Boatner 1991). The Lα1 lines were chosen for the analysis of La, Y and Ce and the Lβ1 lines for Pr, Sm, Nd and Gd. The Si, P and Ca concentrations were analysed on the Kα1 lines. Calibration of PbO was carried out on a vanadinite standard. The U was calibrated on a glass standard with 5 wt% UO₂. Following the procedure outlined by Schulz and Schüssler (2013), monazite from a pegmatite in Madagascar (Madmon) was used for the calibration of ThO₂ and as a reference for the EMP analyses. The analytical results of the standard measurements are given in Electronic Supplement Table S3. As summarised by Schulz and Schüssler (2013), the age of the Madmon reference standard is well known and was determined by different techniques providing ages of 496 ± 9 Ma (concordant SHRIMP U–Pb age), 497 ± 2 Ma (TIMS Pb–Pb evaporation age) and 502 ± 6 Ma and 503 ± 6 Ma, respectively (EMP chemical model ages). The minor Y interference on the Pb Mα1 line was corrected by linear extrapolation after measuring several Pb-free yttrium glass standards with 5 and 12 wt% Y₂O₃ (Montel et al. 1996). The interference of Th Mγ on U Mβ was also empirically corrected, and a Gd interference on U Mβ requires correction if Gd₂O₃ in monazite is larger than 5 wt%. These parameters and measurement conditions were chosen to counter the analytical problems and limits of the method as discussed by various authors (e.g. Williams et al. 2006; Jercinovic et al. 2008; Suzuki and Kato 2008; Spear et al. 2009).