Age of cleft monazites in the eastern Tauern Window: constraints on crystallization conditions of hydrothermal monazite

Abstract

Monazite-bearing Alpine clefts located in the Sonnblick region of the eastern Tauern Window, Austria, are oriented perpendicular to the foliation and lineation. Ion probe (SIMS) Th–Pb and U–Pb dating of four cleft monazites yields crystallization ages of different growth domains and aggregate regions ranging from 18.99 ± 0.51 to 15.00 ± 0.51 Ma. The crystallization ages obtained are overlapping or slightly younger than zircon fission track ages but older than zircon (U–Th)/He cooling ages from the same area. This constrains cleft monazite crystallization in this area to ~300–200 °C. LA-ICP-MS data of dated hydrothermal monazites indicate that in graphite-bearing, reduced host lithologies, cleft monazite is poor in As and has higher La/Yb values and U concentrations, whereas in oxidised host rocks opposite trends are observed. Monazites show negative Eu anomalies and variable La/Yb values ranging from 520 to 6050. The positive correlation between Ca and Sr concentration indicates dissolution of plagioclase or carbonates as the source of these elements. The data show that early exhumation and cleft formation in the Tauern is related to metamorphic dome formation caused by the collision of the Adriatic with the European plate and that monazite crystallization in the clefts occurred later. Our data also demonstrate that hydrothermal monazite ages offer great potential in helping to constrain the chronology of exhumation in collisional orogens.

1 Introduction

Monazite, (LREE,Th,U)PO₄ represents an excellent mineral for combined ²³²Th/²⁰⁸Pb, ²³⁸U/²⁰⁶Pb and ²³⁵U/²⁰⁷Pb isotopic dating of geologic processes (e.g., Parrish 1990). Monazite is very resistant to radiation damage (e.g., Meldrum et al. 1998, 1999, 2000; Nasdala et al. 1999), and U–Th-Pb resetting by solid diffusion is negligible (Cherniak et al. 2004). The numerous studies reporting monazite age “resetting” attribute it to dissolution-precipitation processes in the presence of fluids along with a modification of the chemical composition of the monazite (e.g., Seydoux-Guillaume et al. 2002; Hetherington et al. 2010; Harlov et al. 2011; Williams et al. 2011; Janots et al. 2012; Didier et al. 2013).

Monazite ages can be determined from Th-U–Pb concentrations (e.g., Suzuki and Adachi 1991; Suzuki et al. 1994; Montel et al. 1994, 1996; Scherrer et al. 2000). This assumes that the monazite has not incorporated common Pb and that the common Pb has a negligible influence on the age determination. This dating approach is especially useful in polymetamorphic regions, where rock-forming monazite generally shows complex, small-scale growth zoning (e.g., Berger et al. 2006; Krenn et al. 2011a; Schulz and von Raumer 2011). These can be dated only with the high spatial resolution provided by electron microprobe analysis (EMPA), unless the monazites are less than 100 Ma old such that EMPA is insufficient to precisely analyze the amount of Pb (e.g., Montel et al. 1996). Furthermore, initial/common Pb becomes important in young monazite, especially in hydrothermal environments (Krenn et al. 2011b; Seydoux-Guillaume et al. 2012).

Rare hydrothermal cleft monazite is commonly mm-sized in contrast to monazite from metamorphic rocks, which is normally <100 µm. Hydrothermal monazite growth patterns can display both dissolution and growth stages (Janots et al. 2012). Individual growth domains, identifiable by trace element variations (REE + Y, Th, U and Pb), are sufficiently large enough to be dated precisely utilizing in situ isotopic techniques (secondary ion mass spectrometry, SIMS, and inductively coupled plasma mass spectrometry, ICP–MS). In ideal cases it is even possible to resolve growth duration (Janots et al. 2012; Berger et al. 2013).

Alpine clefts formed in the presence of fluid at the ductile–brittle transition during or after the peak of metamorphism (e.g., Mullis et al. 1994; Mullis 1996). They represent open, originally fluid-filled voids in veins and fissures. They form during tectonic activity and are oriented roughly perpendicular to the lineation and foliation. Due to interaction of hot fluid (<500 °C) with the wall rock, dissolution of minerals occurs in the wall rock leading to the formation of porosity and an alteration halo around the cleft. New minerals precipitate in the open cleft space and in the wall rock pores. For this reason, many clefts display hydrothermal bleaching (dissolution of mafic minerals from the adjacent rock walls). The cleft mineralization does not represent a metamorphic paragenesis but a crystallization sequence that has grown during continuous fluid-rock interaction and dissolution and reprecipitation stages in response totectonic movements, exhumation, and cooling. This results in growth domains observed in the cleft monazite (e.g., Janots et al. 2012) and other cleft minerals.

Although micas and feldspars are frequent in alpine fissures, attempts to use them for dating crystallization in Alpine clefts have failed, in many cases, due to excess Ar accumulating in these minerals (e.g., Purdy and Stalder 1973).

Alpine clefts in some metasediments and metagranitoids (Niggli et al. 1940; Mannucci et al. 1986) have long been known to occasionally contain well-developed monazite crystals but only recently have dating attempts proved to be successful (Gasquet et al. 2010; Janots et al. 2012; Berger et al. 2013). The first successful cleft monazite study in the Lauzière massif, Western Alps (Gasquet et al. 2010), yielded two groups of ²³²Th/²⁰⁸Pb ages at 5–7 and 10–11 Ma, obtained by laser ablation inductively coupled plasma mass spectrometry (LA–ICP–MS). Corresponding ²³⁵U/²⁰⁷Pb ages were considered imprecise (low ²⁰⁷Pb) and ²³⁸U/²⁰⁶Pb ages meaningless due to ²⁰⁶Pb _excess (Gasquet et al. 2010). By using a SIMS, Janots et al. (2012) showed that hydrothermal monazite generally provides the most reliable dates with ²³²Th/²⁰⁸Pb dating due to ²⁰⁶Pb_excess and the low radiogenic Pb component in the U–Pb system. Janots et al. (2012) also showed that ²⁰⁶Pb _excess (due to isotopic disequilibrium in the ²³⁸U-²⁰⁶Pb decay chain) can reach 70 % of the total measured ²⁰⁶Pb.

In combination with textural information, Janots et al. (2012) were able to show that two cleft monazites in the Gotthard and Aar Massifs, Switzerland, started to crystallize at 15.2 ± 0.3 and 14.1 ± 0.3 Ma, respectively. The youngest rims were dated at 13.5 ± 0.4 Ma. Monazites occurring in shear zones from the southern Aar Massif, yielded crystallization stages between 8.03 ± 0.22 Ma and 6.25 ± 0.60 Ma (Berger et al. 2013). The resolvable age difference shows that in both studies growth was episodic and lasted for 1–2 Ma. Moreover, in the Aar and Gotthard Massifs, monazite crystallization occurred a few million years after the cleft formation (Janots et al. 2012).

In this study we will present the crystallization ages of cleft monazites from the eastern Tauern window. We will show that the chemical and isotopic composition of monazite is influenced by the host rock composition that cleft formation is linked to Miocene metamorphic doming and that monazite crystallisation occurred later during exhumation.

2 Tectonic setting

2.1 Alpine context

In the Eastern Alps, Austroalpine nappes, derived from the Adriatic plate, predominate and overlie the Penninic and Subpenninic nappe stack (Fig. 1). This entire nappe stack has been affected by large-scale E-W oriented extension and doming in Early to Middle Miocene triggering unroofing of tectonically lower, Penninic and Sub-Penninic nappes (Fig. 1). These units are now exposed in the Tauern tectonic window (e.g., Schmid et al. 2004). In the Tauern Window (Fig. 1), Penninic crystalline basement nappes with Mesozoic sedimentary cover and structurally lower European plate units, including part of the distal deposits (Valaisan basin), occupy the core of the window. Both the Tauern and Rechnitz Penninic windows reached amphibolite facies metamorphism during the Alpine orogeny. The PT peak of this Barrovian metamorphism is dated at ~30 Ma in the Tauern Window (e.g., von Blanckenburg et al. 1989; Christensen et al. 1994; Inger and Cliff 1994). The metamorphic ages are thus younger than metamorphism in all other parts of the Eastern Alps (e.g., Hoinkes et al. 1999), but comparable to ages (e.g., Hunziker et al. 1992) obtained in the Lepontine metamorphic dome (Todd and Engi 1997).

Fig. 1

Tectonic map of the Tauern Window according to Schmid et al. (2013). See inset for location. The tectonic window consists of a metamorphic dome forming the core of large anticlines produced during collision of the Europen and Adriatic plates. The formation is coeval with orogen-parallel extension (e.g. Ratschbacher et al. 1991; Frisch et al. 2000). Several Tertiary plutons occur in the region of the right-lateral Periadriatic line. BSZS Brenner shear zone system, KSZS Katschberg shear zone system, MF Mölltal Fault, NF Niedere Tauern Fault, PF Pustertal Fault, SEMP Salzach-Ennstal-Mariazell-Puchberg Fault, ZWD Zwischenbergen-Wöllatratten-Drau Fault

2.2 Exhumation and cooling of the Tauern Window

Peak metamorphic conditions of up to 650 °C were reached in the structurally deepest units of the Tauern metamorphic dome by the end of the early Oligocene (Thöni 1999). This was in response to the Late Eocene to Early Oligocene collision of the European with the Adriatic continental plate that causes nappe stacking (Schmid et al. 2013 and references therein). Peak metamorphic temperatures reached ~500 °C in the Sonnblick area (Figs. 1, 2). Post-nappe folding (Hoinkes et al. 1999; Scharf et al. 2013; Figs. 1, 2) is related to N–S shortening and associated E-W oriented extension dated at 17.5–16.5 Ma (Steininger et al. 1996). These ages correspond to mica and zircon fission track data obtained in the eastern part of the Tauern Window (Dunkl et al. 2003; Scharf et al. 2013). Around 13 Ma this extensional phase and the rapid exhumation of Tauern Window rocks was terminated (Steininger et al. 1996).

Fig. 2

Tectonic map of the eastern Tauern Window based on Scharf et al. (2013) and Schmid et al. (2013). Structural data are from Exner (1962) and gold vein information from Feitzinger and Paar (1991). Cleft and gold vein orientations mark trends. The length of the signs has no meaning. Numbered stars mark localities of dated cleft monazites (see Table 1): 1 Erfurter Steig; 2 Griesswies; 3 Lohninger Quarry; 4 Gjaidtroghöhe. Cross section, adapted from Schmid et al. (2013), shows sample localities in their tectonic position. The nappe parts shown above the line of the present topography indicate the amount of material eroded. TAUERN3 is placed at the locality from where the landslide material was derived. For lithology see Fig. 1

Luth and Willingshofer (2008) and Rosenberg and Berger (2009) compiled age data providing information about post-collisional cooling of the Eastern Alps and analysed the cooling pattern and the cooling rates of the Tauern Window. Their compilation showed that the eastern Tauern Window had cooled below the zircon fission track closure temperature of ca. 280–240 °C (Yamada et al. 1995; Bernet and Garver 2005; Reiners 2005; Bernet 2009) between 19 and 16 Ma (Dunkl et al. 2003; Foeken et al. 2007; Wölfler et al. 2012). Apatite fission track data show that the bulk of the Tauern Window had not cooled below ~110 °C before 15 Ma (Staufenberg 1987; Foeken et al. 2007; Wölfler 2008), in contrast to the older cooling ages in Austroalpine units surrounding the study area (e.g., Hejl 1998; Dunkl et al. 2003; Wölfler et al. 2012).

3 Materials and methods

3.1 Sample locations and description

The monazite grains in this study were sampled by crystal searchers from steeply dipping, roughly NE-SW striking clefts located in the Sonnblick region of the eastern Tauern Window, Austria (Figs. 2, 3a, b). Only in the Erfurter Steig and Griesswies cases (Table 1) it was possible to visit the exploited mineral cleft. The other two clefts were quarried away or not accessible. Cleft monazite is rarely recognised in the field, because samples are generally covered by dirt. The monazites are from clefts hosted in metasedimentary rocks of the Modereck and Glockner nappe systems (Fig. 1; Schmid et al. 2013) that also occupy a synclinal position between the Sonnblick and Hochalm sub-domes (Fig. 2). The clefts (Fig. 3a, b) characteristically occur at the lower end of subvertical quartz veins located in lower amphibolite facies metamorphic metasedimentary gneisses and schists. The clefts are subvertical and oriented perpendicular to the main foliation and lineation (Fig. 2). Their orientation is at a small angle to the Au-bearing veins in the area (Feitzinger and Paar 1991; Fig. 2).

Fig. 3

Mineral clefts and monazites. a Series of parallel clefts in a 10 m sized outcrop of two-mica schist from the Griesswies, Rauris Valley. Their orientation is perpendicular to the foliation and lineation. b Quartz vein at Griesswies, Rauris Valley, showing the beginning of the rotation of the quartz vein into the foliation resulting in an oblique orientation relative to the foliation and lineation. Younger set of fractures is oriented perpendicular to the foliation and the lineation, which limits the exploited cleft room located at the lower end of the quartz vein. Image width is 2 m. c Secondary electron image showing TAUERN2 monazite crystals from Griesswies, Rauris Valley, growing on euclase. d 3 mm sized and twinned cleft monazite (TAUERN1) on a quartz crystal. Erfurter Steig, Rauris Valley

Table 1 Locations of the four analyzed Tauern cleft monazites

Monazites from the clefts (Fig. 3c, d) are yellow, brownish-yellow to rose in colour. They occur either as a crust consisting of 50–100 µm-sized crystals (Fig. 3c) or well-developed mm-sized individuals (Fig. 3d). Monazite is associated with albite (pericline), quartz, anatase/rutile, adularia, muscovite-phengite, carbonate, sulphides, chlorite, and rarely with hematite, tourmaline and the Be minerals euclase, BeAlSiO₄(OH), and phenakite, Be₂SiO₄ (Table 1). Orthoclase (adularia), quartz, and albite-oligoclase (“pericline”) were the first minerals to grow in the cleft. Monazite grew late on quartz, feldspar, euclase (Fig. 3c) and tourmaline (Fig. 3d), and is commonly coeval with rutile or anatase. In general, only clinochlore crystallised later than monazite (e.g., Niedermayr 1980).

Monazite TAUERN1 (star 1 in Fig. 2) is a crystal sitting on cleft quartz from the Erfurter Steig, Rauris, Salzburg (Fig. 3d). Monazite TAUERN2 (Fig. 2) forms a rose-coloured crust consisting of 50-100 µm-sized crystals in a cleft of the Griesswies, Rauris, Salzburg, where it occurs in association with albite and euclase (Table 1; Fig. 3c). Monazite TAUERN3 (Fig. 2) is a sample from the Lohninger quarry, Rauris, Salzburg, where gneiss blocks from the Permo-Triassic Wustkogel Formation (Frasl 1958) are quarried from pre-historic landslide material that originated from the western slope of the valley (Pestal et al. 2009). Monazite TAUERN4 (Fig. 2) is from the Gjaidtroghöhe, Grosses Fleisstal (Carintia). Monazite crusts and well-developed samples have been selected in order to test if their crystallization age is different.

3.2 Analytical techniques

Backscatter electron (BSE) images were obtained on a JEOL JXA8200 electron microprobe (EMPA) at the University of Copenhagen.

Th-U–Pb isotope analyses of monazite were obtained on a Cameca IMS1280 SIMS instrument at the Nordsims facility (Swedish Museum of Natural History). Analytical methods closely follow those described by Harrison et al. (1995) and Kirkland et al. (2009), using a −13 kV O₂ ⁻ primary beam of ~6 nA and nominal 15 µm diameter. The mass spectrometer was operated at +10 kV and a mass resolution of ca. 4300 (M/ΔM, at 10 % peak height) with data collected in peak hopping mode using an ion-counting electron multiplier. U–Pb and Th–Pb data were calibrated against an in-house reference monazite, C83-32 (Corfu 1988). Analytical details and correction procedures closely follow those described in Kirkland et al. (2009) and Janots et al. (2012). Pb isotope signals were corrected for common Pb contribution using measured ²⁰⁴Pb and a present day Pb isotope composition predicted by the model of Stacey and Kramers (1975). In monazite, ²⁰⁴Pb is affected by an unresolvable molecular interference from doubly charged ²³²Th¹⁴⁴Nd¹⁶O₂ ⁺⁺ (²⁰⁶Pb and ²⁰⁷Pb are also affected to a smaller degree by ThNdO₂ ⁺⁺), which can result in an over estimate of the amount of common Pb. The extent of this interference was monitored using ²³²Th¹⁴³Nd¹⁶O₂ ⁺⁺ at mass 203.5 and a correction applied whenever the count rate exceeded the average background count on the ion-counting detector by three times its standard deviation. Age calculations were done using the decay constant recommendations of Steiger et al. (1977) and plots use the routines of Isoplot (Ludwig 2001) with uncertainties presented at the 2σ level. Th–Pb ages presented in the results section were calculated using common Pb and polyatomic Nd-overlap corrections.

Major and trace element concentrations of monazites were obtained by LA-ICP-MS system at the University of Bern using a GeoLas Pro 193 nm ArF Excimer laser system (Lambda Physik, Germany) coupled with an ELAN-DRCe quadrupole ICP-MS (Perkin Elmer, USA). Measurements closely followed procedures detailed in Pettke et al. (2012), employing an energy density of 4–5 Jcm⁻² on the sample surface at 10 Hz laser repetition rate. Spot diameters varied between 24 and 60 µm. Analyses were made next to ion probe age dating spots. The ICP-MS settings were optimised to maximum signal to background intensity ratios with (²³²Th¹⁶O)⁺ production rates below 0.5 % and robust plasma conditions as monitored by equal sensitivities of U and Th. Data reduction was done using SILLS (Guillong et al. 2008), with improved calculation of the limit of detection (Pettke et al. 2012), using the method without internal standard (Liu et al. 2008) and a sum of analysed oxides of 100 %. The synthetic SRM 610 glass from the National Institute of Standards and Technology (NIST) was used as the external standard.

4 Results

4.1 Age dating

All four dated cleft monazites are from the tectonically complex eastern region of the Tauern Window (Figs. 1, 2). Numbers in Fig. 4 refer to analysis spots.

Fig. 4

BSE images of monazites showing numbered ion probe spot analyses. a TAUERN1; b TAUERN2; c TAUERN3 and d TAUERN4. The indicated spots are larger than the ion probe spot size. Numbers refer to Tables 3, 4, 5, 6. The yellow spots indicate LA-ICP-MS analysis spots. D and R refer to domains and regions addressed in the text

Isotopic results for TAUERN1 show some compositional variations in the BSE image (Fig. 4a) that correlate with Th-U compositions (Tables 2, 3). A dark core (low in Th) is surrounded by brighter rim domains. Analysis point 21 is discarded due to its position on a crack expressed by an increased Th/U value of 27 (Table 3). Analyses in the darker core domain D1 in Fig. 4a gives a ²³²Th-²⁰⁸Pb age of 16.25 ± 0.55 Ma (MSWD = 3.0; spots 9–21; Fig. 5a). The three surrounding domains (D2–D4; Fig. 5a) give different ²³²Th-²⁰⁸Pb ages at 17.56 ± 0.55 Ma (MSWD = 1.6; spots 1–8), 15.00 ± 0.51 Ma (MSWD = 1.7; spots 22–29), and 18.99 ± 0.51 Ma (MSWD = 0.51; spots 30–34).

Table 2 Monazite LA–ICP–MS element concentrations (µg/g)

Table 3 SIMS Th-U–Pb isotopic ratios, element concentrations and ²³²Th-²⁰⁸Pb ages of monazite TAUERN 1

Fig. 5

²³²Th-²⁰⁸Pb and Tera-Wasserburg ages of cleft monazites from samples TAUERN1–4

U–Pb ages are more complex. The ²⁰⁷Pb/²⁰⁶Pb intercept of domain1 is not compatible with a common Pb composition, thus the Tera-Wasserburg diagram is inappropriate. In fact, only domain 4 shows negligible common Pb and the lower intercept of the Tera-Wasserburg diagram yields age of 19.10 ± 0.57 Ma, in agreement with the ²³²Th-²⁰⁸Pb age.

Isotopic results for TAUERN2 are compiled in Table 4. Individual analysis spots (Fig. 4b) represent ages of one or mixed analyses of two of these grains. Monazites are characterised by low amounts of Th (9–8740 µg/g; Table 4) with Th/U values between 0.02 and 11. Analysis spot 04 lies on a fracture showing increased common Pb and has been excluded. Analysis spots 2, 5, 6, 8 (Fig. 5b) from region 1 have Th >1000 µg/g and give a²³²Th-²⁰⁸Pb age of 15.12 ± 0.48 Ma (MSWD = 0.26). The other spots are too low in Th and have high initial Pb, thus robust ²³²Th-²⁰⁸Pb ages cannot be determined. However, in a Tera-Wasserburg diagram the U–Pb dataset yields a lower intercept age of 17.08 ± 0.27 Ma (MSWD = 1.19; Fig. 5b). Isotope analysis results are compiled in Table 4.

Table 4 SIMS Th-U–Pb isotopic ratios, element concentrations and ²³²Th-²⁰⁸Pb ages of monazite TAUERN 2

Isotopic results for TAUERN3 (Fig. 5c) show extreme Th/U ratios of up to ~1400. Four domains are distinguished texturally and by age determinations. The core of the crystal shows a complex growth zonation.

Core domain 1, characterised by similar U/Th values of the analysed spots, is the largest (Fig. 4c) and has the highest Th concentrations (Th > 40,000 µg/g). It has a ²³²Th/²⁰⁸Pb age of 15.49 ± 0.15 Ma (MSWD = 0.74; spots 23–48; Fig. 5c). Analysis points 23 and 32 have higher Th values and a higher apparent age. Domains 2 and 3 are darker in Fig. 4c but cannot be distinguished on the image. They have the lowest Th concentrations (Th < 20,000), but the highest Th/U ratios (>1090). Domain 2 gives a ²³²Th/²⁰⁸Pb age of 16.03 ± 0.27 Ma (MSWD = 0.51; spots 15–22; Fig. 5c). Domain 3 gives an age of 18.06 ± 0.42 Ma (MSWD = 0.51; spots 11–14; Fig. 5c), and domain 4 shows more scatter and yields an age of 17.18 ± 0.49 Ma (MSWD = 3.4; spots 1–10; Fig. 5c). Taking all 48 Th–Pb analyses together yields a geologically meaningless weighted mean age with a too high MSWD of 5.4. Due to the low U concentrations, U–Pb ages are imprecise, and the Tera-Wasserburg diagram gives an intercept age with a large error of 20.0 ± 5.2 Ma (MSWD = 0.93). Isotope analysis results are compiled in Table 5.

Table 5 SIMS Th-U–Pb isotopic ratios, element concentrations and ²³²Th-²⁰⁸Pb ages of monazite TAUERN 3

Isotopic results for TAUERN4 are compiled in Table 6. TAUERN4 (Fig. 4d) is an aggregate consisting of <10 µm-sized, zoned crystals with very variable Th concentrations ranging from 5 to 16,383 µg/g. Analysis spots with Th <200 µg/g gave unreliable ages. No correlation was detected between age and Th-U composition. For that reason all points are treated together resulting in an imprecise ²³²Th/²⁰⁸Pb age of 15.56 ± 0.70 Ma (MSWD = 9.1; Fig. 5d). The high MSWD reflects the large scatter in the data. U values are <300 µg/g. The Tera-Wasserburg diagram shows variable amounts of common Pb with possible excess ²⁰⁶Pb.

Table 6 SIMS Th-U–Pb isotopic ratios, element concentrations and ²³²Th-²⁰⁸Pb ages of monazite TAUERN 4

4.2 Monazite composition

LA-ICP-MS analyses are listed in Table 2 and analysis spots marked in yellow in Fig. 4. Analysis steps that showed Na spikes were interpreted as a mix with a fluid inclusion and discarded. All elements listed in Table 2 have been measured in comparable concentrations at three spots in the same growth zone. CI-chondrite-normalised (McDonough and Sun 1995) patterns of different domains in the four dated monazites show negative Eu anomalies (Fig. 6) and quite variable steepness, with La/Yb values ranging from 5600–6050 (TAUERN1) to 4430–4500 (TAUERN2), 521–930 (TAUERN3) and 1900–2000 (TAUERN4). Total REE concentrations show little variation (Table 2). By plotting La/Yb versus U, the four cleft monazites plot as individual clusters (Fig. 7a). The increased HREE concentrations in domains 1 and 2 (Table 2) of TAUERN3 correlate with an increased Y and Si content (thorite-huttonite, ThSiO₄, component). Average Th/U values vary from 4.4 to 7.2 (TAUERN1), 0.03–200 (TAUERN2), 740–1020 (TAUERN3) and 0.08–47 (TAUERN4). This compositional variation is displayed in Fig. 7b. A similar trend is observed when plotting As concentration (76–3233 µg/g; gasparite, REEAsO₄, component; Table 2) versus U (Fig. 7c). The monazites show 5400–7900 µg/g Y (xenotime, YPO₄, component) and ~1000–9200 µg/g Ca (brabantite, CaTh(PO₄)₂, component). Calcium correlates positively with 117–840 µg/g Sr (Table 2; Fig. 7d). Whereas As and Y are regularly distributed in the monazite, Ca and Sr show higher concentrations in the core regions of TAUERN 1 and 3 monazite. The analysed trace elements Be, Mg, Fe, and Te are below the detection limit. Sodium, Ti, and Bi are generally at or below the detection limit. Silicon and Bi are above the detection limit only in TAUERN3 (associated with tourmaline) and TAUERN4. Antimony is above the detection limit only in TAUERN4. Lithium is above the detection limit in some domains and regions of TAUERN1, 2 and 4 (Table 2). The fourstudied monazites also systematically show 0.42–1.01 µg/g Rb, 0.5 µg/g Mo and ~0.1-2.5 µg/g W.

Fig. 6

Chondrite-normalised LA–ICP–MS REE patterns of the four monazites from the Tauern Window. Variation in slope of the monazite REE patterns

Fig. 7

Compositional variation in TAUERN1–4. The arrows in plots a–c indicate trends related to more oxidizing/reducing conditions

5 Discussion

5.1 Age dating

Whereas the core and rims of TAUERN1 give well-resolved ²³²Th-²⁰⁸Pb ages of 15.00 ± 0.51 to 18.99 ± 0.51 Ma, the age distribution within the crystal is puzzling because there is no obvious relationship between the geometry of the growth zones and the age as observed in other cases (Janots et al. 2012; Berger et al. 2013). This suggests that growth was either more complicated than suggested by BSE imaging (Fig. 4a), or more likely, not all growth zones are cut by the section analyzed. We can also not exclude that the patchy core region of TAUERN3 represent a zone of monazite replacement. However, this needs to be confirmed by other monazite examples.

TAUERN2 yielded a ²³²Th-²⁰⁸Pb mean crystallization age of 15.12 ± 0.48 Ma, based only on four spots containing sufficient Th. The well-defined U–Pb age of 17.08 ± 0.27 Ma, based on all the analysis spots from TAUERN2, is comparable to the monazites, despite occurring as a crust consisting of tiny monazites (Fig. 4b). This suggests that Be-mineral bearing clefts do not represent a separate, younger generation of clefts. Rather the occurrence of Be minerals seems to be controlled by the host rock containing Be-bearing biotite (e.g., Weiss et al. 2005).

TAUERN3 is comparable to TAUERN1 and yields ²³²Th-²⁰⁸Pb domain ages between 15.49 ± 0.15 and 18.06 ± 0.42 Ma. TAUERN4, which also consists of an aggregate of small monazite grains, gives an imprecise average ²³²Th-²⁰⁸Pb age of 15.56 ± 0.70 Ma. This age also falls in the age range obtained with the other samples.

The obtained age results indicate that all four cleft monazites crystallised between ~19 and 15 Ma and that there is no resolvable age difference between large monazites and monazite crusts. The data also confirm that hydrothermal monazite can only be dated reliably with the ²³²Th-²⁰⁸Pb system (see discussion in Janots et al. 2012), but that in rare cases, here strongly reduced host rocks, the U–Pb dating of low Th/U hydrothermal monazites may also yield reliable ages.

5.2 Mineral composition

Chemical and isotopic results indicate that each monazite occurrence has distinct U, Th, La/Yb, As and Ca (Sr) concentrations. Uranium and La/Yb are the most discriminative (Figs. 6, 7a). As all four clefts occur in rocks with identical peak metamorphic conditions (Scharf et al. 2013), the observed compositional variations between the cleft monazites seems to be controlled by the oxidation state of the fluid, which presumably is regulated by the oxidation state of the host rock. Whereas monazites TAUERN1 and 2 are from clefts in graphite-bearing schists (“Schwarzschiefer”), TAUERN3 is hosted in a strongly oxidised Permo-Triassic meta-arenite where monazite occurs in association with hematite. The gneiss hosting TAUERN4 has an intermediate oxidation state. The relationship between the cleft monazite composition and the oxidizing/reducing host rock lithology is underlined by the arrows in Fig. 7a–c. Under oxidizing conditions As is pentavalent and enters the monazite structure, replacing P (e.g., Janots et al. 2006, 2011; Ondrejka et al. 2007; Fig. 7c). Under more reducing conditions As is probably trivalent or monovalent and goes in cleft sulphides, mainly pyrite (Liang et al. 2013). The negative Eu anomaly in the chondrite-normalised monazite REE patterns (Fig. 6) is interpreted as being inherited from the sedimentary protolith or metamorphic paragenesis. Metamorphic minerals like garnet (HREE host) or hornblende (MREE host), that could influence the REE patterns of the cleft monazite, do not occur in the host rocks. For this reason dissolution of accessory minerals, e.g., monazite, xenotime, fluorapatite, allanite, feldspar and carbonates, likely provided the REEs for the cleft monazite. The fact that some clefts in the Lohninger quarry contain tens of grams of loose monazite (Josef Rathgeb pers. com., 2013) suggests that strings of REE-mineral enriched placers occur in the meta-arenite exploited in the quarry. The correlation of Ca and Sr (Fig. 7d) indicates dissolution of host-rock plagioclase or carbonates as a source. Increased Sr values in TAUERN2 indicate an additional source. Evidence of fluorapatite dissolution (high in LREE or MREE) is only recognizable in TAUERN4.

Monazite TAUERN3, which crystallised with tourmaline, shows very low concentrations of B that correlate with Li concentrations in four of the five analysed domains (Table 2). Monazite TAUERN2 that crystallised with the Be-mineral euclase in Be-bearing biotite schists (Weiss et al. 2005) hosts less than 0.4 µg/g Be (Table 2). The analysed Na in some monazite domains probably reflects the presence of tiny NaCl-bearing fluid inclusions.

The larger cleft monazites have chemically variable domains expressed by variable Th/U values (Fig. 4; Tables 2, 3, 4, 5, 6) in agreement with the data of Janots et al. (2012) and Berger et al. (2013). Monazite TAUERN3 shows extreme Th/U ratios of up to ~1400. The results show that TAUERN3 and TAUERN4 Th/U values plot in the field of hydrothermal monazite proposed by Janots et al. (2012). In contrast, TAUERN1 and 2 from the “Schwarzschiefer” (graphite-bearing host rock) show Th/U values characteristic of monazite in medium-grade metamorphic rocks. This suggests that more oxidizing conditions (TAUERN3 and 4) are required so that hexavelent U is not incorporated in monazite but remains in the fluid (Janots et al. 2012). Indeed, the U⁶⁺-bearing mineral cleusonite, (Pb,Sr)(U⁴⁺,U⁶⁺)(Fe²⁺,Zn)₂(Ti,Fe²⁺,Fe³⁺)₁₈(O,OH)₃₈, is reported in hematite-monazite-bearing clefts of the Lohninger Quarry.

Xenotime has been reported in cleft TAUERN2 (Niedermayr et al. 2010). If xenotime had crystallized coevally with monazite, it would be possible to use the monazite-xenotime thermometer (e.g., Heinrich et al. 1997). However, the mol. fraction of HREE_Gd-Ybin TAUERN2 is 0.06–0.07, indicating unrealistic temperatures of ~500–550 °C. It hence seems that xenotime crystallization was not in equilibrium with monazite.

The four studied cleft monazites thus document very nicely that cleft monazite composition is controlled by host-rock composition, which controls the composition of the mineralising fluid. The prevailing oxygen fugacity seems to have a major influence on monazite composition too.

5.3 Exhumation

The obtained ~19–15 Ma cleft monazite crystallization ages in the eastern part of the Tauern Window can be compared with published zircon fission track and zircon (U–Th)/He cooling data from the same area (Dunkl et al. 2003; Wölfler 2008). Our results show that cleft monazite crystallization ages are coeval to slightly younger than zircon fission track ages (Fig. 8) with an estimated closure temperature between 240 and 280 °C (Yamada et al. 1995; Bernet and Garver 2005; Reiners 2005; Bernet 2009). However, the monazite crystallization ages are older than the (U–Th)/He zircon ages (Wölfler et al. 2012) with a closure temperature in the range of 150–220 °C (Bernet and Garver 2005; Reiners 2005). Comparable observations have been made in the southern Aar Massif (Berger et al. 2013). This suggests that, in contrast to monazite in rocks, stepwise monazite crystallization in clefts may occur at or below the zircon fission track closure temperature of 280–240 °C during exhumation, and in combination with complex fluid/rock interaction in variably oxidised host rocks.

Fig. 8

Compilation of thermochronological data for the Sonnblick area given in Wölfler et al. (2012). The zircon fission track and (U–Th)/He data constrain the crystallization temperature of the cleft monazite ages from the Sonnblick region to ~300–200 °C. Mu muscovite, Bt biotite, Mnz monazite crystallization, ZFT zircon fission track, ZHe zircon (U–Th)/He, AFT apatite fission track, AHe apatite (U–Th/He). 1 Reddy et al. (1993); 2 Lambert (1970); 3 Waters (1976); 4 Dunkl et al. (2003); 5 Bertrand (2013); 6 this study; 7 Wölfler et al. (2012); 8 Wölfler (2008); 9 Staufenberg (1987); 10 Foeken et al. (2007). The grey box indicates temperature estimates for cleft formationin metamorphic domes, based on fluid inclusion studies (e.g., Luckscheiter and Morteani 1980; Mullis et al. 1994)

6 Conclusions

The cleft formation in the Sonnblick region of the Tauern Window occurred during metamorphic doming in association with E-W extension, as indicated by the cleft orientation perpendicular to foliation and stretching lineation (Fig. 2). This occurred at the stress field of the main ductile deformation. Cleft formation is most likely controlled by local competence contrast and/or fluid enhanced embrittlement during overall ductile deformation. The time and PT conditions of the cleft formation are not well known. However, in comparison with other Alpine metamorphic dome regions, formation occurred probably at 450 ± 50 °C (Luckscheiter and Morteani 1980; Mullis et al. 1994; Fig. 8). Monazite crystallization in the clefts occurred a few million years later, ~19–15 Ma ago, at the end of crystallization of quartz and adularia in the clefts, and at temperatures of ~300–200 °C. The Tauern Au veins are only at small angle to the alpine cleft orientation (Fig. 2). This suggests that they probably formed in the same stress field slightly before or after the cleft formation.