Late Cretaceous ultramafic lamprophyres and carbonatites from the Delitzsch Complex, Germany

Chemical Geology 353 (2013) 140–150

Contents lists available at ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Late Cretaceous ultramafic lamprophyres and carbonatites from the Delitzsch Complex, Germany J.C. Krüger ⁎, R.L. Romer, H. Kämpf GFZ German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam, Germany

a r t i c l e

i n f o

Article history: Accepted 13 September 2012 Available online 23 September 2012 Keywords: Rb–Sr and U–Pb dating Phlogopite Xenocryst Short duration of emplacement Delitzsch Complex

a b s t r a c t The Delitzsch Complex consists of late Cretaceous ultramafic lamprophyres and carbonatitic rocks. They form dikes and diatremes, emplaced into Palaeozoic to lower Permian volcanic and sedimentary rocks, and are covered by up to 120 m thick sequences of Tertiary sedimentary rocks. The complex includes a diversity of magmatic and subvolcanic rocks. The lithologies range from monchiquites and alkali picrites to dolomite– and calcite– carbonatites (rauhaugites, beforsites, and alvikites), and ultramafic lamprophyres (alnöites, aillikites). Contact relationships and the distribution of xenolithic material indicate that phases of carbonatitic and ultramafic lamprophyre magmatism overlapped. New U–Pb ages (72± 1 Ma on baddeleyite) from a dolomite–carbonatite (beforsite), Rb–Sr ages (73± 2 Ma on phlogopite) from an ultramafic lamprophyre (alnöite) in combination with modeling of the effect of the initial 87Sr/86Sr of phlogopite on the isochron ages of dolomite– and calcite– carbonatites demonstrate: (1) a short duration of magmatic activity for the main phases of the subvolcanic emplacement of the Delitzsch Complex; (2) phlogopite crystals in carbonatites of the Delitzsch Complex are xenocrysts; (3) the calculated initial 87Sr/86Sr composition of xenocrystic phlogopite equals the initial Sr composition of phlogopite from the alnöite. © 2012 Published by Elsevier B.V.

1. Introduction Alkaline rocks and associated carbonatites are unique samples that provide direct insight into the geochemical and isotopic development of the Earth's mantle and constraints for the processes leading to its spatial heterogeneity. Alkaline complexes associated with carbonatites are known from all over the world and several studies using the Sr, Nd, and Pb isotopic as well as trace element compositions of these rocks, focused on the petrogenetic relationship among these rocks. The isotopic characteristics of carbonatites are well documented (i.e. Nelson et al., 1988; Bell and Blenkinsop, 1989) and have been interpreted to reflect mixing products of material derived from the hypothetical (HIMU and EM1) mantle end-members (Bell et al., 1998) with Sr isotopic compositions typically less than εSr = 10, although there are carbonatites documented that exceed this value (e.g. Harmer and Gittins, 1998). A recurrent issue in carbonatite genesis is (e.g. Gittins, 1989) whether (i) carbonatites are ‘secondary’ melts, which were generated by liquid immiscibility or (ii) carbonate melts were derived from a partially molten carbonated mantle peridotite. Experimental data are equivocal. Immiscibility seems to occur at shallower depths, whereas at greater mantle depths a primary carbonate melt is more likely to be generated (e.g. Kiseeva et al., 2012). If carbonatitic melts are produced by partial melting of a carbonated mantle peridotite, the spatial relationship

⁎ Corresponding author. E-mail address: [email protected] (J.C. Krüger). 0009-2541/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.chemgeo.2012.09.026

with alkaline rocks in terms of a cogenetic evolution should be reevaluated (Harmer, 1997). A second issue, long debated, is the temporal relation between ultramafic rocks and related carbonatites in alkaline complexes (e.g. Nelson et al., 1987). Geochronologic studies of alkaline complexes (e.g. Tappe et al., 2009, 2011) support a close genetic relationship between carbonatites and alkaline rocks. The temporal evolution of such complexes seems to be a centerpiece for understanding the mechanisms that are involved to form these relatively rare and small magmatic bodies as it puts important constrains to the still open debate of carbonatite genesis and its relationship to associated alkaline rocks. Importantly, results from experimental studies and conditions of carbonatite genesis should be verified on existing field relations between alkaline rocks and carbonatites to evaluate the reliability of these models. The Delitzsch ultramafic lamprophyre–carbonatite Complex provides new insights into the relationships between carbonate and silicate-dominated ultramafic rocks in alkaline massifs, in particular among carbonatites and various types of ultramafic lamprophyres. The complex is located in the eastern part of Germany and kept secret for a long time by the former GDR, because of its potential for REE, Nb–Ta, and U deposits. The Delitzsch Complex was discovered during uranium exploration and is probably the largest Mid-European carbonatite body. The carbonatitic and ultramafic rocks from dikes were first found by Meissner (1967) in drill cores. Over 500 drill holes were sunk in the 1970s and 1980s (Seifert et al., 2000) and the cores were meticulously sampled by the SDAG Wismut Company (Soviet/East-German Uranium mining Co.) to assess the economic

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

potential of the area. Although petrographic, geochemical, and isotopic investigations of the complex started immediately after it had been found, the results of these early investigations were not available except for internal reports — many of them in Russian. The exploration of the Storkwitz diatreme in combination with geologic, petrographic, geochemical, and isotopic investigations of the Delitzsch Complex started in the 1980s by a cooperation between geologists of the SDAG Wismut Company and the Geological Survey of East Germany (Zentrales Geologisches Institut, Berlin). These early investigations eventually were summarized by Röllig et al. (1989, 1990, 1995) and Wasternack (2008), who presented detailed geological, lithological, as well as the first geological multi-stage eruption sequence scheme of the Delitzsch Complex (Table 1). Wand et al. (1989) presented the first isotope data and demonstrated that δ18O, δ13C and initial 87Sr/ 86Sr of the rocks fall in the ‘carbonatite’ field. First modern results of mineral composition, petrography and geothermobarometry of the rock types were published by Seifert et al. (2000). Röllig et al. (1990) also refers to K/Ar phlogopite ages published in various Russian SDAG Wismut reports in the 1970s and 1980s. These K/Ar phlogopite ages of the alnöite and the dolomite–carbonatite breccia, which were interpreted to date the time of emplacement encompassing a range from 110 Ma to 70 Ma (Kozyrev, 1977; Fedoriziev et al., 1989). This broad age range was interpreted as evidence for a long-lived magma system beneath the complex, producing the intrusive sequence established on the basis of cross-cutting relationships (Röllig et al., 1995). Twenty years after the German reunification, the age of the Delitzsch Complex is still only poorly known and there are only a few, modern petrographic, geochemical and isotopical descriptions of the complex. Previous K–Ar ages are unreliable for three reasons: (i) large age ranges are recorded — even for samples from the same dike, (ii) the lack of documentation of the original data and the correction procedures in the unpublished reports of Kozyrev et al. (1977) and Fedoriziev et al. (1989), and (iii) the possibility of the presence of excess radiogenic argon. New U–Pb and Rb–Sr ages from different rocks from this area supersede earlier K–Ar ages and give insights into the temporal development of the Delitzsch Complex and the relations between carbonatites and ultramafic lamprophyres. 2. Geological setting The Delitzsch Complex is located at the southern border of the Mid-German Crystalline Zone, which represents the former suture between Laurussia and Gondwana, being active during the closure

141

of the Rheic Ocean during the Variscan Orogeny (Fig. 1a). The Delitzsch Complex is situated in the so-called Wrench and Thrust Zone (cf. Kroner et al., 2007; Kroner et al., 2010), which includes Palaeozoic sedimentary rocks that have been deposited on the Cadomian basement and were strongly reworked during the Variscan orogeny. Variscan structural elements were repeatedly reactivated during reorganization of the regional stress field, such as the development of the Oslo Rift, the opening of the Tethys and the Atlantic, and the Alpine Orogeny. These tectonic reactivations of older elements led to the formation of various horst and graben structures with up to 3 km of uplift (Wagner, et al., 1997). The Delitzsch Complex is situated on the intersection of an E–W trending structural low and a seismically active long-lived N–S trending zone (Bankwitz et al., 2003), which is characterized today by low magnitude swarm earthquakes. Carbonatites and ultramafic lamprophyres, developed as distinct geological bodies forming dikes, sills and pipe-shaped small intrusions (Röllig et al., 1990) have been found in an area of more than 450 km 2 (Wasternack, 2008). They are overlain by up to 120 m thick sequences of Tertiary sedimentary rocks (Standke, 1995). The Delitzsch Complex is emplaced into a heterogeneous series of Palaeozoic to lower Permian volcanic and sedimentary rocks. Dikes of alnöites, monchiquites, aillikites, dolomite– and calcite– carbonatite, as well as the small diatremes of Storkwitz and Serbitz near Delitzsch (Fig. 1c) give evidence for subvolcanic intrusions in the depth range up to 600 m (Wasternack, 2008). The diatremes of Storkwitz and Serbitz are characterized by intrusive dolomite– carbonatite breccias containing abundant angular xenoliths of metamorphic, igneous, and metasedimentary rocks, as well as rounded xenoliths of coarse-grained dolomite–carbonatite (rauhaugite), fenites, and glimmerites, suggesting that carbonatites and related lamprophyres occur at greater depths than sampled by the drilling campaign. According to Röllig et al. (1995), the Delitzsch Complex is characterized by a multi-stage intrusive sequence, summarized in Table 1. 3. Sample descriptions and petrography Sample selection is based on the eruption sequence according to Röllig et al. (1995) to include the entire compositional range and time sequence of subvolcanic magmatism (Table 1). Furthermore, sampling was restricted to rocks with the least indication for contamination. Nonetheless, most samples show abundant xenoliths or xenocrysts of crustal wall-rocks that had been acquired during the

Table 1 The multi-stage eruption sequence (I=oldest, VI=youngest) for the Delitzsch UML-CR complex, Germany, according to Röllig et al. (1995). Stage

Event

Depth levela

Rock types

Sample (location)

I II

Intrusion of carbonatitic magma body Intrusion of ultramafic and alkali lamprophyres

Hypabyssal Subvolcanic

Dolomite–carbonatite Ultramafic lamprophyres (alnöite, aillikite, monchiquites)

III

Formation of diatremes (‘intrusive breccia’)

Subvolcanic

Dolomite–carbonatite (beforsite) with xenoliths (UML and dolomitecarbonatite)

IV V

Intrusion of lamprophyres within diatremes of stage III Formation of beforsite dikes

Subvolcanic Subvolcanic

Ultramafic and alkalilamprophyres Dolomite–carbonatite (beforsites) without xenoliths

VI

Formation of carbonate dikes

Subvolcanic

Calcite–carbonatite (alvikite), partly with xenoliths

– ALN 51°33′11″N 12°17′31″E W2 51°32′05″N 12°17′26″E – 5551-2, -7, -10 51°32′05″N 12°17′26″E SER 51°33′24″N 12°14′48″E KMD6, KMD8 51°32′05″N 12°17′26″E

a

Depth level according to Röllig et al. (1995), based on depth in drill core and textural relation.

142

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

a

Rock types in boreholes Ultramafic lamprophyres

c re s Sutu

Iapetu

iss Te

FMC

M

TBU MZ

e

BM

Ultramafic lamprophyres and carbonatites

in

Z SX

S

.L

Z

RH

AM

re

nq

RM

Front ine

p

Al

Gondwana Delitzsch Complex

ALN B

SER

AM Armorican Massif, FMC French Massif C., RM Rhenish Massif, BM Bohemian Massif, RHZ Rheno-Hercynian Zone, SXZ Saxo-Thuringian Zone, TBU Teplá-Barrandian Unit, MZ Moldanubian Zone, S Sudetes, M Moravo-Silesian Zone

b

Carbonatites

Bitterfeld

or

e

utu

-T re ey

ssiaRheic S

Lauru

Rheic Sutu r

c

i

t

t

e

r

f

e

l d

h

Bitterfeld

Brehna

o

r

s

t

KMD6, KMD8 W2, 5551Schenkenberg Delitzsch

K 5 km

y

Upper Permian to Lower Triassic Vendian Cambrian Molasse deposits Delitzsch pluton

h

n

a h

Serbitz diatreme

Delitzsch

o

r

s

Kyhna

t

Storkwitz diatreme 0

Permian Delitzsch pluton

1

2 km

Fig. 1. (a) Geological map of central Mid-Europe with major tectonic elements and Variscan basement massifs. The Mid-German Crystalline Zone is marked by the Rheic Suture as it is the former border between Gondwana and Laurussia. The complex of Delitzsch is located between the Rheic Suture and the Saxo-Thuringian Zone (SXZ). (b) Geological map of the Delitzsch area, presenting the stratigraphic units around the Permian pluton of Delitzsch. Lamprophyres and carbonatites were emplaced in Palaeozoic to lower Permian rocks, overlain by Tertiary sediments and molasse deposits. Some drill holes, intersecting lamprophyre and carbonatite dikes were collared in the Bitterfeld horst. (c) Drill collar positions of holes that encountered lamprophyres and/or carbonatites. The highest concentration of lamprophyres and carbonatites is in the immediate proximity of the Permian Delitzsch pluton, between two structural highs, the Bitterfeld Horst and the Kyhna Horst. The diatremes of Serbitz and Storkwitz are marked. Sample locations are indicated by arrows with sample signature. Two calcite–carbonatite samples (KMD6, KMD8) and the dolomite–carbonatite samples (W2, 5551-3, 5551-7, 5551-10) were obtained in the diatrems of Storkwitz. A calcite–carbonatite sample stem from the diatremes of Serbitz (SER) and the alnöite sample (ALN) was collared south of the Bitterfeld Horst. (a) Simplified after Franke (2000), (b) modified after Seifert et al. (2000), and (c) modified after Röllig et al. (1995).

ascent of the carbonatitic and ultramafic melts through the crust. Almost all carbonatite rocks from the subvolcanic units (Stages II–VI) are breccias, probably reflecting the rapid, partly explosive emplacement of the dikes. Sub-solidus reactions after the emplacement of the dikes (i.e. Nasraoui et al., 2000) are only of minor significance as Wand et al. (1989) demonstrated that δ 18O and δ 13C of various carbonatite samples from the Delitzsch Complex plot in the ‘primary igneous carbonatite’ box, as first defined by Taylor et al. (1967), which indicates a near-primary character. The original nomenclature of carbonatites and ultramafic lamprophyres, partly used by earlier authors (i.e. Rock, 1991) is superseded and replaced by the nomenclature of Tappe et al. (2005). For completeness, the original terminology (e.g., Röllig et al., 1990, 1995) is given in parentheses.

3.1. Alnöite (ultramafic lamprophyre) The alnöite dike was encountered from a depth of 180 m to 280 m in a drill hole, collared in the northern part of the complex (Fig. 1c). The alnöite, which forms a dike or a root zone of a diatreme that is locally strongly fractured, is a black to greenish-black porphyritic rock with abundant melilite, olivine, phlogopite, and compositionally diverse pyroxene (i.e., diopside, aegerine–augite). Accessory minerals include apatite, magnesian–chromite, ilmenite, and magnetite. Phlogopite, locally reaching up to 20 vol.%, occurs as dispersed phenocrysts with a kink band structure, which is common in phlogopite-bearing rocks like alnöites, minettes, and lamproites. This structure indicates pronounced deformation during the rapid and probably explosive

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

ascent of the magma. The crystal rims show no secondary reaction and no indications of disequilibrium textures (i.e., dissolution). Drill core logging and thin section investigations demonstrated that the dike is very heterogeneous with several zones of contrasting amounts of xenolithic and xenocrystic material, locally reaching up to 50 vol.%. Quartz and feldspar are the dominant xenocrysts, which are clearly derived from the disintegration and incorporation of crustal rocks. Alnöite with little or no xenolithic and xenocrystic material is found at depth levels below 240 m (sample ALN). 3.2. Dolomite–carbonatite (beforsite) The medium- to fine-grained dolomite–carbonatite samples (5551-2, 5551-7, and 5551-10) are from a dike near Storkwitz (Fig. 1c). Some of the beforsite dikes reach a thickness of several meters. The beforsite consists of up to 80 vol.% of dolomite. Phlogopite locally reaches up to 10 vol.% and occurs as fine dispersed crystals up to several millimeters in grain size. Apatite, pyrochlore, fluoro-carbonate, zircon, and baddeleyite are accessory minerals (Seifert et al., 2000). 3.3. Dolomite–carbonatite breccia (beforsite breccia) The breccia sample W2 contains 10–45 vol.% angular xenolithic material from the adjacent wall-rocks (shales and felsic magmatites), xenocrysts, which include richterite (e.g. Tappe et al., 2009), aegirine, and alkali-feldspar, as well as rounded xenoliths of carbonatite, glimmerite, fenite, and ultramafic lamprophyre that have been picked up at deeper levels and became fragmented and partially rounded during transport to the surface. The matrix of this rock type consists of ferroan dolomite with sporadic siderite (Seifert et al., 2000, Table 2). Phlogopite (~5 vol.%) occurs as dispersed crystals, locally reaching a grain size of up to 2–3 cm. Accessory minerals include apatite, magnetite, and pyrochlore. 3.4. Calcite–carbonatite (alvikite) The alvikite sample (SER) was obtained from a vein near Storkwitz (NW part of the complex; Fig. 1c) from a depth of about 375 m. Alvikite veins typically are only a few centimeters thick, fine-grained and composed of more than 50 vol.% calcite. The texture varies from granular to intersertal. Accessory minerals include ilmenorutile, apatite, pyrochlore, fluoro-carbonate, and zirkelite (Seifert et al., 2000). Phlogopite (~5 vol.%, max. 10–15 vol.%) is the most abundant non-carbonate mineral. The grain size of phlogopite crystals ranges from 200 μm to about 1 mm. Rare phlogopite crystals reach up to 3 mm in diameter. The crystals are well dispersed in the carbonate matrix and cumulate aggregates were not recognized. 3.5. Calcite–carbonatite breccia (alvikite breccia) The alvikite–breccias (KMD6, KMD8) were obtained from one drill core at 400 m and 450 m depth, respectively. Calcite is the main mineral in both samples and phlogopite crystals are the most abundant accessory minerals. Phlogopite reaches up to 5–8 vol.% with a grain size similar to phlogopite from the calcite–carbonatite (Section 3.4). Xenolithic material is abundant and seems to be predominantly derived from the crust, as indicated by a shift in the Sr isotopic whole-rock compositions toward crustal values. 4. Analytical methods and results 4.1. U–Pb dating of zircon and baddeleyite (dolomite–carbonatite) The dolomite–carbonatite samples 5551-2, 5551-7, and 5551-10 were dissolved in dilute HNO3. Apart from fragments of K-feldspar and quartz, the insoluble silicate residue contained two single crystals

143

of zircon and several tiny crystals of baddeleyite. The zircon crystal forms are a simple tetragonal prism and simple tetragonal pyramids, typical for magmatic zircon. Edges and corners of the zircon crystals are distinct and not corroded. The zircon crystals are perfectly clear, colorless, and have no inclusions. The platy baddeleyite crystals are clear brown with perfect striations. Most crystals represent fragments of larger crystals that have been irregularly broken perpendicular to the crystallographic C-axis. Both zircon crystals were analyzed for U–Pb as single crystals, whereas several baddeleyite crystals were combined to produce adequately large samples. Zircon and baddeleyite samples were weighed and spiked with a mixed 205Pb–235U tracer. The samples were dissolved in concentrated HF using Parr autoclaves and kept at 220 °C for four days. After evaporating HF, the samples were redissolved in 6N HCl in the autoclaves at 220 °C overnight (Romer et al., 1996). U and Pb were separated using the chemical procedure described by Krogh (1973). U and Pb were loaded together with silica-gel on a single Re-filament and were analyzed on a Finnigan MAT262 multi-collector mass-spectrometer operated in static multicollection mode using Faraday collectors and a secondary electron multiplier. For measurement conditions and analytical results see Table 2. The two zircon single crystals from the dolomite–carbonatite (sample 5551-2) yield concordant to slightly disconcordant 206Pb/238U ages at 454±6 Ma and 431±4 Ma (2σ). These ages are too old to agree with the field geological constraints for the emplacement age of the Delitzsch Complex. Instead, the ages fall in the known age ranges of Variscan basement rocks that includes age groups of 340–330 Ma and 325–280 Ma for metamorphic and magmatic rocks as well as 480–430 Ma old magmatic rocks related to the Mid-German Crystalline Zone (Anthes and Reischmann, 2001). The zircon crystals are interpreted to be inherited from the basement and to have been incorporated into the carbonatite during the explosive emplacement of the dikes. Baddeleyite from the Delitzsch Complex yielded concordant data with a 206Pb/ 238U age at 72 ± 1 Ma (2σ, MSWD = 0.016) (Fig. 2). Baddeleyite is not present in felsic rocks like those hosting the dikes of the Delitzsch Complex or cut by the magma conduits. Instead baddeleyite is more common in silica-undersaturated rocks such as carbonatites and lamprophyres (e.g. Amelin and Zaitsev, 2002). Consequently, we suggest that the baddeleyite age represents the time of emplacement of the dolomite–carbonatite (Stage V) from the Delitzsch Complex. 4.2. Rb–Sr dating of phlogopite (alnöite) Phlogopite crystals without macroscopically visible intergrowths of other minerals were selected under the binocular microscope. The phlogopite typically is dark-brownish, single thin transparent and clear plates with a pseudo-hexagonal shape. For each sample, about 10 mg of phlogopite with grain sizes ranging between 180 μm (alnöite) and 3 mm (calcite–carbonatite) was separated. Phlogopite samples were dissolved in concentrated HF and each sample was split into two fractions for isotopic composition (IC) — and concentration determination of Rb and Sr by isotopic dilution (ID) measurements. For the ID measurements a mixed 84Sr– 87Rb tracer was used. For preparation of the whole-rock samples see Table 3. Sr was separated from all samples (ID, IC, whole-rock) using standard cation exchange techniques described in Romer et al. (2001, 2005). For measurement conditions and analytical results see Table 3. A five point isochron for the alnöite phlogopite (Stage II) samples yields in an Rb–Sr age of 73 ±2 Ma (2σ, MSWD = 3.3) (Fig. 3a). This age falls near the lower end of the range defined by unpublished K–Ar age determinations and overlaps with the 206Pb/238U age of the baddeleyite crystals from the dolomite–carbonatite. Phlogopite of all other samples does not define similarly robust isochrons as the alnöite sample. Instead, the phlogopite samples plot at moderate Rb/Sr and the associated whole-rock samples plot at very

low Rb/Sr, defining a pattern not too different from the one of a ‘two-point’ isochron. In such a situation, the slope of the regression line and, thus, the apparent age strongly depends on the initial (i.e., the whole-rock composition), as the phlogopite samples alone do not allow to closely constraining the slope of the regression line. As some of these ‘two-point’ isochrons yield apparent ages that appear anomalously young in the context of the geological setting of the complex, it is quite possible that the initial isotopic composition of carbonatite and phlogopite were not the same or that the Sr isotopic composition of the carbonatite had been modified after phlogopite crystallization by assimilation of wall-rocks.

4.3. Phlogopite compositions in alnöite and related carbonatites Electron microprobe analysis of phlogopite was carried out using a JEOL JXA-8230 (GFZ German Research Centre for Geosciences, Potsdam), operating at 20 kV voltage, 20 nA beam current and 20–30 s counting time. The major element composition of phlogopite (222 analyses, see Appendix) was determined on polished thin sections by wavelengthdispersive spectrometry (WDS). Some samples were affected by extensive alteration, therefore only phlogopite crystals without indications of serpentinization or chloritization were analyzed. The microprobe data show that phlogopite from the alnöite and the carbonatites has distinct compositional characteristics, which were used to group the data (Table 4). For representative analyses of each group see Table 5. Phlogopite from the alnöite can be subdivided into two groups with different compositions. The first group (L1) is characterized by high Mg# from 0.89 to 0.92 with TiO2 ranging from 3.8 wt.% close to 5 wt.% (0.43– 0.88 apfu) and Al2O3 between 15 wt.% and 17.7 wt.% (2.59–2.92 apfu). The second group (L2) of phlogopite from the alnöite is marked by moderate Mg# from 0.82 to 0.84 with TiO2 ranging from 4.2 wt.% to 5.3 wt.% (0.46–0.58 apfu) and Al2O3 between 13.1 wt.% and 16.5 wt.% (2.21– 2.84 apfu). Phlogopite data from the various carbonatite samples can be combined and subdivided into three groups. In general, the composition is more variable than in phlogopite from the alnöite. The majority of phlogopite data from the carbonatite samples belongs to one, compositionally relatively variable group (C1), which is characterized by Mg# from 0.79 to 0.85 with TiO2 ranging from 3.2 wt.% to 6.4 wt.% (0.37– 0.72 apfu) and Al2O3 between 14.9 wt.% and 16.5 wt.% (2.65– 2.82 apfu). The other two phlogopite groups from carbonatites define relative narrow compositional fields. One group (C2) is marked by Mg# from 0.77 to 0.80 with TiO2 ranging from 2.0 wt.% to 2.5 wt.%

76

Dolomite-carbonatite 0.0116

206Pb/238U

Samples were analyzed as Pb+ at 1220 to 1260 °C and as UO2+ at 1320 to 1360 °C, respectively, on a Finnigan MAT 262 multi-collector mass-spectrometer at Deutsches GeoForschungsZentrum using Faraday collectors and ion-counting. Lead isotope ratios corrected for fractionation with 0.1%/a.m.u. c Corrected for fractionation, blank, trace lead, and initial lead with the composition 206Pb/204Pb = 18.7 ± 0.1, 207Pb/204Pb = 15.50 ± 0.03, and 208Pb/204Pb = 38.4 ± 0.2. During the measurement period total blanks were less than 15 pg for lead and less than 1 pg for uranium. Concentrations were determined using a 205Pb–235U mixed tracer. d Apparent ages were calculated using the constants recommended by IUGS (λ235U = 9.8485E-10 y−1, and λ238U = 1.55125E-10 y−1). b

a

73 ± 80 60 ± 100 72.4 ± 7.5 71.8 ± 15.8 72.3 ± .7 72.0 ± .9 4.49 4.44 5.03 8.51 147 109 0.059 0.044 Baddeleyite 5 (5551–7) 6 (5551–10)

203 187 1990 1925 Zircon 1 (5551–2) 2 (5551–2)

0.009 0.011

Pbtot U

46.664 28.855

94.45 93.77

1.06 1.79

0.01128 ± 0.00011 0.01123 ± 0.00014

0.0739 ± 0.0079 0.0733 ± 0.017

0.0475 ± 0.0048 0.0472 ± 0.010

431 ± 9 445 ± 14 431 ± 4 454 ± 6 0.0554 ± 0.0012 0.0546 ± 0.0019 0.5290 ± 0.013 0.5490 ± 0.022 0.0699 ± 0.0007 0.0729 ± 0.0011 9.25 10.08 4.77 4.66 149.8 208.1

85.98 85.26

207

Pb/235U 207

Pb/238U 206 206

Pb/204Pb

207

Pb/204Pb

208

Pb/204Pb

206

Pb/238U

207

Pb/235U

207

Pb/206Pb

Apparent ages (Ma)d Atomic ratiosc Radiogenic Pb (at.%)c Pb/204Pb measured ratiosb 206

Concentrations (ppm) Weight (mg) Samplea

Table 2 U–Pb analytical data of zircon and baddeleyite from the Delitzsch Complex, Germany.

430 ± 48 397 ± 80

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150 Pb/206Pb

144

74

72

0.0112 70

0.0108 68 Concordia Age = 72±1 Ma (2σ,decay-const. errs included)

0.0104

66

MSWD (of concordance) = 0.016 Probability (of concordance) = 0.90

data-point error ellipses are 2σ

0.0100 0.05

0.06

0.07

0.08

0.09

207Pb/235U

Fig. 2. Concordia diagram showing analyses of baddeleyite from the dolomite– carbonatite (Stage V) from the Delitzsch Complex. Baddeleyite (samples 5551-7, 5551-10) crystal fractions yield concordant data with a 206Pb/238U age at 72 ± 1 Ma (2σ, MSWD = 0.016, probability of concordance = 0.90). Data from Table 2.

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150 Table 3 Rb–Sr analytical data of phlogopite minerals and their host rocks from the Delitzsch Complex, Germany. Concentrations (ppm) Rb Ultramafic lamprophyre (alnöite) ALN1 1.392 309 ALN2 2.154 207 ALN3 2.374 211 ALN4 0.292 255 ALN WRb 73 Calcite–carbonatite (alvikite) SER1 0.283 SER2 0.211 SER3 0.225 SER4 0.358 SER WRb SER CFd

87 86

Rbc/ Sr

87

Src/86Sr

0.708

Sr 158 114 119 150 545

5.651 5.264 5.132 4.910 0.387

Sr/ 86Sr

Weight (mg)

Alnöite

0.710

0.709405 ± 9 0.708824 ± 11 0.708676 ± 6 0.708604 ± 8 0.703874 ± 7

0.706

87

Samplea

a

145

0.704 315 341 338 334 – 16

255 281 583 438 – 3997

3.574 3.514 1.675 2.203 – 0.011

0.707945 ± 8 0.707937 ± 26 0.705902 ± 14 0.706258 ± 5 0.703623 ± 8 0.703598 ± 7

Calcite–carbonatite–breccia (alvikite–breccia) KMD61 1.032 384 301 KMD62 1.373 361 267 KMD63 1.376 383 310 KMD64 1.817 312 238 KMD6 WRb – – d KMD6 CF 35 769 KMD81 1.055 262 183 KMD82 1.056 287 193 KMD83 0.833 317 186 KMD84 0.977 317 184 KMD8 WRb – – KMD8 CFd 29 1981

3.684 3.919 3.581 3.789 – 0.131 4.153 4.290 4.938 4.995 – 0.042

0.707736 ± 17 0.707725 ± 5 0.707501 ± 2 0.707759 ± 4 0.705626 ± 3 0.705000 ± 19 0.708106 ± 7 0.707958 ± 11 0.708628 ± 7 0.708705 ± 8 0.705193 ± 15 0.704373 ± 5

Dolomite–carbonatite–breccia (beforsite–breccia) W21 1.210 337 143 W22 0.992 356 153 W23 1.191 346 144 W2 WRb – – W2 CFd 26 1174

6.854 6.740 6.938 – 0.063

0.710620 ± 4 0.710120 ± 4 0.710603 ± 6 0.704585 ± 10 0.704380 ± 5

a The phlogopite samples represent inclusion-free, clear, dark brownish, thin platy crystals ranging in grain size from 50 μm (ultramafic lamprophyre) to 3 mm (calcite– carbonatite). The samples are multi-grain, when applicable of contrasting grain size and include only inclusion-free, clear crystals. Phlogopite was purified by carefully breaking up the flakes using a mortar, rinsing in ethanol, and removing released impurities. The final phlogopite samples were rinsed shortly with 10% acetic acid. Phlogopite samples were dissolved in concentrated HF and each sample was split in two fractions for isotopic composition (IC) — and isotopic dilution (ID) measurements. For the ID measurements a mixed 84Sr–87Rb tracer was used. Ion exchange chromatographic separation of Rb and Sr was done as described by Romer et al. (2001, 2005). b The whole rock samples (WR) were mechanically crushed until a grain size of 60 μm was achieved. The WR powders were dissolved in concentrated 4–5 ml HF in Teflon vials at 150–160 °C for four days. The samples were redissolved in 2N HNO3 to break fluorides, dried, and transferred into chloride-form using 6N HCl. Ion exchange chromatographic separation of Sr was done as described by Romer et al. (2001, 2005). c Strontium and Rb were loaded on single Ta-filaments. The isotopic composition of Sr (IC and ID) was determined on a Thermo Scientific Triton multi-collector mass spectrometer using dynamic multi-collection. Rb was analyzed on a Thermo Scientific Triton multicollector mass spectrometer using static multi-collection. Repeated measurement of Sr standard NBS 987 during the measurement period gave 0.710249±0.000004 (2σ reproducibility for n=12 independent analyses). Analytical uncertainties of the individual measurements are reported as 2σm. Total procedural blanks are less than 50 pg Sr.87Sr/86Sr was normalized to 86Sr/88Sr=0.1194. The 2σ uncertainty of 87Rb/86Sr is better than 2%. The 2σm uncertainty of 87Sr/86Sr refers to the last digit(s). d The whole rock samples were mechanically crushed until a grain size und 60 μm was achieved. The rock powders were dissolved in 2N HCl for 30 min to dissolve the carbonate amounts selectively. The remaining silicate amounts were removed. The carbonate fraction (CF) samples were dried, and transferred into chloride-form using 6N HCl. Ion exchange chromatographic separation of Sr was done as described by Romer et al. (2001, 2005).

(0.22–0.28 apfu) and Al2O3 between 13.7 wt.% and 14.8 wt.% (2.38– 2.51 apfu). The other group (C3) of phlogopite from the carbonatite samples plots at low Mg# from 0.65 to 0.66 with TiO2 ranging from 4.0 wt.% to 4.5 wt.% (0.46–0.51 apfu) and Al2O3 at 15.5 wt.% (2.77 apfu).

Rb-Sr age= 73±2 Ma Initial87Sr/86Sr=0.70347 ± 0.00004 0.702 0

2

4

6

87

Rb/ 86Sr

b

Spl Phl

Fig. 3. (a) Rb–Sr isochron for phlogopite from the alnöite (Stage II) from the Delitzsch Complex. The five-point isochron gives an age of 73±2 Ma with an initial 87Sr/86Sr of 0.70347± 0.00004 (2σ, MSWD=3.3). Data from Table 3. The filled symbols are phlogopite multi-grain samples, the open symbol represents the whole-rock sample. (b) BSE (back-scattered electron) image of a phenocrystic phlogopite in the ultramafic lamprophyre (alnöite) with a porphyritic structure and a fine-grained matrix. The bright mineral, partly included in phlogopite, is a magnesian chromite crystal. The rapid ascent and the related deformation are responsible for the local deformation of phlogopite crystals. There is only one generation of phlogopite. Phl = phlogopite, Spl = magnesian–chromite.

The Al2O3 content in all samples defines a relatively narrow field and is broadly similar, whereas the TiO2 content is more variable among the defined groups. Furthermore, phlogopite from the alnöite has a Cr content near the detection limit of the microprobe, as Cr is mostly bound to magnesian–chromite (up to 26 wt.% Cr) that coexists with the phlogopite. Thus, the low Cr content in these phlogopite Table 4 Grouping of phlogopite data from the Delitzsch Complex based on compositional variations. Group Samples included

Rock type

Compositional characteristics (relative abundances)

L1

ALN

Alnöite

L2

ALN

Alnöite

C1 C2

SER, W2, KMD6 KMD6, W2

C3

KMD6

Dolomite–, calcite– carbonatite Dolomite–, calcite– carbonatite Calcite–carbonatite

High Mg#, moderate TiO2, Cr2O3 b.d.a Moderate Mg#, moderate TiO2, Cr2O3 b.d.a Moderate Mg#, moderate–high TiO2, low–high Cr2O3 Moderate Mg#, low TiO2, Cr2O3 b.d.

a

Low Mg#, moderate TiO2, Cr2O3 b.d.

The absence of Cr in phlogopite samples is due to the presence of earlier or simultaneous crystallization of magnesian–chromite, recognized in the samples.

146

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

Table 5 Representative major element composition of phlogopite from the alnöite and the carbonatites of the Delitzsch Complex. Oxides, wt.% groupa

Alnöite (ALN)

Calcite–carbonatite (SER)

Calcite–carbonatite breccia (KMD6)

Dolomite– carbonatite breccia (W2)

L1

L2

C1 (i.c)b

C1 (o.c.)b

C1 (mantle)

C1

C2

C3

C1

C2

Na2O K2O CaO BaO MgO MnO FeO Al2O3 TiO2 Cr2O3 SiO2 Cl Mg# Total

0.34 9.16 0.01 0.50 24.24 0.05 4.63 16.60 4.44 b.d 35.23 0.05 0.90 95.39

0.52 9.09 0.02 1.81 20.17 0.08 7.41 16.09 4.96 b.d. 35.69 0.01 0.83 95.84

0.44 9.53 0.01 0.48 18.93 0.01 6.45 16.27 5.04 1.34 37.65 0.01 0.84 96.16

0.46 9.33 0.03 0.50 18.42 0.08 6.85 16.12 4.69 0.94 38.07 0.02 0.83 95.51

0.55 9.56 0.01 0.59 18.99 0.04 6.76 16.08 4.57 0.80 37.57 0.01 0.83 95.57

0.36 9.71 0.01 0.44 18.21 0.04 6.47 15.19 6.28 1.48 36.17 0.01 0.83 94.57

0.55 9.53 0.01 0.26 19.91 0.18 9.57 13.70 2.36 b.d. 38.45 0.01 0.78 94.69

0.51 9.26 0.02 0.42 15.09 0.19 14.27 15.52 4.43 0.02 35.08 b.d. 0.65 94.84

0.89 9.31 b.d. 0.54 19.55 0.07 8.13 15.66 3.82 0.46 37.01 0.01 0.81 95.56

0.92 8.92 0.02 0.32 20.75 0.09 9.47 14.50 2.00 b.d. 38.88 0.01 0.79 96.00

Elements, atoms per formula Na K Ca Ba Mg Mn Fe Al Ti Cr Si F Cl

unit (calculated to 11 atoms of oxygen) 0.01 0.15 0.12 1.68 1.69 1.74 0.00 0.00 0.00 0.03 0.54 0.03 5.19 4.38 4.04 0.01 0.01 0.00 0.56 0.89 0.77 2.81 2.76 2.75 0.48 0.54 0.54 0.00 0.00 0.15 5.06 5.20 5.39 0.00 0.00 0.00 0.46 0.09 0.00

0.13 1.72 0.00 0.03 3.96 0.01 0.83 2.74 0.51 0.11 5.49 0.00 0.00

0.15 1.76 0.00 0.03 4.09 0.00 0.82 2.74 0.50 0.09 5.43 0.02 0.00

0.10 1.82 0.00 0.03 3.99 0.01 0.80 2.63 0.70 0.17 5.32 0.09 0.00

0.16 1.79 0.00 0.02 4.38 0.02 1.18 2.38 0.26 0.00 5.67 0.08 0.00

0.15 1.79 0.00 0.03 3.40 0.02 1.81 2.77 0.50 0.00 5.31 0.01 0.00

0.25 1.73 0.00 0.03 4.25 0.01 0.99 2.69 0.42 0.05 5.39 0.05 0.00

0.26 1.65 0.00 0.02 4.48 0.01 1.15 2.47 0.22 0.00 5.63 0.05 0.00

b.d.

Below detection limit of the microprobe. Grouping as described in Section 4.3 (also see Table 4). b i.c.: inner core; o.c.: outer core. a

samples indicate either simultaneous crystallization of phlogopite and magnesian–chromite or early crystallization of magnesian–chromite, followed by the crystallization of phlogopite in a chromium depleted environment. Phlogopite from the carbonatite samples shows a bimodal compositional pattern. Whereas two groups (C2 and C3) are marked by low chromium contents (below detection limit), the other group (C1) shows a highly variable Cr2O3 content ranging from 0.1 wt.% to 1.75 wt.%. There is a compositional zoning within some phlogopite crystals, especially present in sample SER, but there are also grains with a relative homogeneous chemical composition (i.e. samples KMD6 and ALN). In general, the decrease of Mg# correlates with a drop in the Cr2O3 and TiO2 content with an increase of BaO from core to rim. 5. Discussion/age modeling The Rb–Sr age of phlogopite of the alnöite, which represents Stage II, and the U–Pb age of baddeleyite from a dolomite–carbonatite, representing Stage V, bracket the emplacement age of each phase of magmatic activity from the Delitzsch Complex. Consequently, the two ages (73 ±2 Ma and 72±1 Ma) of the alnöite and the dolomite– carbonatite, respectively are identical within their error limits and constrains the main phase of subvolcanic emplacement is reliably to fall between 75 Ma and 71 Ma. Based on the relative eruption sequence of Röllig et al. (1995), these ages also constrain for the emplacement age of the carbonatites of Stage VI (samples SER, KMD6, KMD8) and the dolomite– carbonatite–breccia (sample W2) of Stage III (cf. Table 1). The age of the dolomite–carbonatite–breccia of Stage III should at least overlap within its uncertainties with the baddeleyite age of the dolomite– carbonatite and the phlogopite age of the alnöite and, thus, should fall

in a time range between 75 Ma and 71 Ma. Samples related to Stage VI of the eruption sequence (SER, KMD6, KMD8) should yield isochron ages, that are younger (within respective error limits) than the baddeleyite age of 72±1 Ma for the dolomite–carbonatite of Stage V. Using the Sr isotopic whole-rock and phlogopite compositions of the various carbonatite samples (Table 3) to calculate the emplacement age yields Rb–Sr isochron ages with (i) significant data scatter, (ii) a broad range of apparent ages, some of which are anomalously young, and (iii) a general tendency toward younger ages, than obtained for the alnöite and the dolomite–carbonatite. These features would be obtained if the initial Sr isotopic composition of phlogopite and host-rock were not identical. An isotopic contrast in initial 87Sr/86Sr of phlogopite and corresponding host-rock may have several different reasons, including (i) the contamination of the host-melt by wall-rock assimilation after the crystallization of phlogopite and (ii) phlogopite and host-magma are not cogenetic, i.e., phlogopite in the carbonatite represents xenocrystic material from the mantle source and may have been picked up by the carbonatite melts from the sublithospheric mantle. To clarify the influence of these two end member scenarios for a non-ideal system, we use (i) selectively dissolved whole-rock samples to minimize the effect of silicate wall-rock material and (ii) the major element composition of phlogopite from the alnöite and the carbonatites to identify their origin. 5.1. Effect of wall-rock contamination The rapid and possibly explosive ascent of carbonatitic and lamprophyric melts through the crust resulted in the fragmentation of wall-rocks and the entraining of different amounts of xenoliths and xenocrysts. The short duration between emplacement and crystallization of the magmas may account for limited material exchange

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

between xenoliths and melts, in many cases resulting only in thin reaction rims around the xenoliths, and preserving even xenocrysts in extreme chemical disequilibria with the melts, as for instance K-feldspar and quartz in carbonatite. Furthermore, the well preserved edge-shaped forms of crystals and broken crystal fragments of K-feldspar and quartz in some segments of the alnöite and in the carbonatites do not support major chemical interaction between disaggregated material and host rock. The problem of xenocryst assimilation is most clearly demonstrated by the Paleozoic age of zircon crystals in the Delitzsch dolomite–carbonatite from Stage V (cf. Table 1). Similarly, Claesson et al. (2000) documented that the age spectrum of zircon crystals in a carbonatite sample from the Kola Peninsula shows maxima that correspond to various orogenic and crust-forming events known for that region. Nonetheless, even if the time between assimilation of wall-rocks and final cooling of the ascending magma was short, it is unlikely that no chemical interactions occurred at all during this time. Xenoliths and xenocrysts shift the isotopic whole-rock composition of the samples toward the isotopic composition of the wall-rocks and may, for instance, increase the 87Sr/86Sr ratios of the rock. In particular, if xenolithic and xenocrystic material is not assimilated to a larger extent by the carbonate melts, selective dissolution of carbonates may give 87Sr/86Sr values that are close to the true initial Sr isotopic composition of the carbonatitic melt. Selective dissolution of carbonatite samples was made using 2N HCl for 30 min to separate silicate (xenolithic) material from carbonate material. The isotopic 87Sr/86Sr ratios of the selectively dissolved carbonate samples are shown in Table 3. This method does not eliminate all assimilation effects that modified the magma during ascent, but it helps to remove all silicate-bounded contributions. Modeled isochrons using the isotopic 87Sr/86Sr composition of selective dissolved bulk-rock samples give (i) a narrower range of apparent ages for the carbonatite samples, than untreated whole-rock samples (ii) a general decrease in MSWD values, which means a better fit of the regression line constructed by phlogopite and host rock, and (iii) a narrower range for the initial isotopic 87Sr/86Sr ratios of the carbonatites from about 0.7036 to 0.7049. Although there is a general improvement of the correlation between the Sr isotopic composition of the selectively dissolved carbonatites and the phlogopite samples, the results are still may be influenced by contamination effects that are not fully removable. Furthermore, the Rb–Sr age of the alnöite and the U–Pb age of the dolomite–carbonatite do not support the calculated isochron ages in respect to the relative eruption sequence (Table 1). 5.2. Chemical composition of phlogopite minerals from the Delitzsch alnöite and related carbonatites as source indicator The chemical composition of phlogopite is an indicator for the conditions, which predominated at the time of crystallization. Furthermore, it allows a direct comparison with other phlogopite data from various mantle-derived rocks. We used phlogopite data from carbonatites (Reguir et al., 2009), lamprophyres (Platt and Mitchell, 1982; Bachinski and Simpson, 1984; Morogan and Wolley, 1988), MARID‐like rocks (Sweeny et al., 1993; Grégoire and Bell, 2002) and from high-pressure mantle xenoliths (Delaney and Smith, 1980). The phlogopite data from the alnöite fall in two groups (L1 and L2), which form relative restricted clusters on major element plots (Fig. 4a, b, c). The Mg# of group L1 is similar to the data from mantle xenolith mica and MARID‐like rocks, but the TiO2 and Al2O3 contents are far high and the Cr2O3 is low to match with these data. The second group of alnöite phlogopite (L2) shows a similar behavior, except that it plots at lower Mg# with slightly higher TiO2 contents. The low Cr2O3 content is due to the presence of magnesian–chromite, as mentioned above, but the unusually high TiO2 content precludes a derivation from a MARID‐type source. Phlogopite data from other alnöites and also from minette‐like rocks sensu lato give a better fit in Al2O3 and

147

TiO2, although there is still a slight deviation, especially in group L1, which does not match with any of the literature data. The phlogopite data from the Delitzsch carbonatites define three groups. The major group C1 and group C2 seem to display a complete fractionation series (Fig. 4a, b) from 1.8 wt.% to 0.0 wt.% Cr2O3 and from 6.5 wt.% to 2 wt.% TiO2 with Mg# between 0.85 and 0.77. Furthermore, the Cr2O3 and TiO2 contents are untypical for phlogopite from carbonatites. A typical feature of carbonatite phlogopite is the elevated MnO content (Reguir et al., 2009), which could not be observed in group C1 (see Appendix). Group C3 has very low Mg# and correlates with data from other carbonatites (Fig. 4a), but the likewise high TiO2 content of this group also seems to preclude a genetic relation. The best fit of data is with phlogopite from the Delitzsch alnöite (Fig. 4b, c, d) and with other alnöites and minette‐like rocks. Phlogopite data from the Delitzsch carbonatites and minette-like rocks show a similar Al2O3 and TiO2 contents, even for phlogopite with very high TiO2. Phlogopite from the Delitzsch carbonatites and the alnöite is similar and overlaps compositionally with phlogopite data from minette-like rocks. This would imply that phlogopite in the Delitzsch carbonatites is dominantly xenocrystic and is derived from the same, lamprophyre-like source. 5.3. Modeling the initial 87Sr/ 86Sr compositions of xenocrystic phlogopite from the Delitzsch carbonatites Phlogopite from the Delitzsch carbonatites is characterized by a high TiO2 content, partly enriched in Cr2O3 and a low to moderate MnO content and, therefore closely resembles phlogopite that crystallized from a lamprophyric source rather than phlogopite that crystallized from a ‘carbonatite’ like environment. Thus phlogopite in the Delitzsch carbonatites may represent xenocrysts. As a consequence thereof, model age calculations using the isotopic composition of the carbonate host rocks may lack geochronologic significance. The lack of knowledge of the initial 87Sr/ 86Sr composition of the source of xenocrystic phlogopite prevents the estimation of robust Rb–Sr isochron ages for the Delitzsch carbonatites. Rather than using the Rb–Sr data to constrain isochrons, the Rb–Sr data can be used to estimate the initial 87Sr/ 86Sr of the xenocrystic phlogopite. The new U–Pb and Rb–Sr ages of the dolomite–carbonatite and the alnöite, respectively in combination with the emplacement sequence determined by Röllig et al. (1995) indicate an emplacement of the Delitzsch Complex from 75 Ma to 71 Ma. These ages can be used to estimate the initial 87Sr/ 86Sr composition of xenocrystic phlogopite from the Delitzsch carbonatites. Modeled initial 87Sr/ 86Sr compositions for phlogopite from the carbonatites are similar to the initial 87Sr/86Sr composition of the alnöite. The range of the modeled initial 87Sr/86Sr compositions of the carbonatite samples is narrow, in particular for sample W2 (0.70316–0.70355), sample KMD6 (0.70370–0.70391), and for sample KMD8 (0.70346–0.70373). The range of the modeled initial 87Sr/ 86Sr composition for sample SER deviates slightly and is more radiogenic (0.70408–0.70422). This rough estimate of the initial Sr composition of the xenocrystic phlogopite suggests that phlogopite from the carbonatites and from the alnöite may be derived from an isotopically similar, lamprophyric source. Assuming phlogopite in the Delitzsch carbonatites is derived from the same source as phlogopite from the alnöite, the range of modeled initial Sr composition of the xenocrystic phlogopite constricts severely (Table 6), in particular for sample W2 (0.70348), KMD6 (0.70343) and KMD8 (0.70346). These results are identical within the error of 0.70347 ± 4, the initial 87Sr/ 86Sr composition of phlogopite from the alnöite. Only the sample SER is slightly outside this group, with a modeled initial 87Sr/ 86Sr composition of 0.70336. As sample SER shows least indications of wall-rock contamination (Table 3), it seems that the initial 87Sr/ 86Sr composition of 0.70336 (modeled using the isotopic whole-rock composition of the alnöite) resembles the ‘original’ initial isotopic Sr composition of phlogopite more

148

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

2.0

1.6

W2 (carbonatite) KMD6 (carbonatite) SER (carbonatite) ALN (alnöite)

1.4

a

alnöites minettes mantle xenoliths carbonatites MARID rocks

b 6 5

TiO2 wt %

Cr2O3 wt %

7

Mica (phlogopite) data from

1.8

1.2 1.0 0.8 0.6

4 3 2

0.4 1

0.2 0 0.6

0.7

0.8

0.9

0 0.6

1

0.7

0.8

Mg#

0.9

1

Mg# 22

25

c

d

Al2O3 wt %

Al2O3 wt %

20

15

10

17

12 5

0 0.4

0.5

0.6

0.7

0.8

0.9

1

7

0

1

2

3

4

5

6

7

TiO2 wt %

Mg#

Fig. 4. Chemical composition of phlogopite samples from the Delitzsch alnöite (blue squares) and from the Delitzsch carbonatites (red, green, brown circles, see legend in subpanel a). Each circle represents a single analysis on the mantle or the core of a phlogopite crystal. The smaller symbols represent phlogopite compositional data from various authors (see Appendix) and are from lamprophyres, in particular from alnöites (yellow squares) and from minettes (orange squares). Other phlogopite compositional data are from mica from mantle xenoliths (light gray triangles), carbonatites (dark gray triangles), and MARID-like rocks (black crosses). (a) Cr2O3 content (wt.%) vs. Mg# plot. Phlogopite data from the Delitzsch alnöite plot in two groups (Table 4) with a Cr2O3 content below the detection limit. The low Cr content is probably a result of earlier or simultaneous crystallization of magnesian–chromite (Fig. 3b) with a Cr content of up to 26 wt.%. Phlogopite data from the Delitzsch carbonatites fall in three groups (Table 4). Group C1 is characterized by variable Cr2O3 contents, which indicate a fractionation series. Sample SER shows a pronounced zoning with high Cr2O3 contents from the core to low Cr2O3 contents to the rim. Phlogopite of group C3 plots at low Mg#, indicating a late stage fractionation. (b) TiO2 content (wt.%) vs. Mg# plot. Phlogopite from the Delitzsch alnöite and carbonatites is marked by a high TiO2 content (2 wt.% to 6.4 wt.%), which is untypical for phlogopite from mantle xenoliths or carbonatites. Group L2 of phlogopite from the alnöite broadly overlaps with phlogopite from sample SER (group C1). (c) Al2O3 content (wt.%) vs. Mg# plot. Phlogopite data from the Delitzsch carbonatites form restricted clusters at Al2O3 contents between 14.5 wt.% and 15.5 wt.%, which overlap with group L2 of phlogopite from the Delitzsch alnöite. Phlogopite data from minette-like rocks show a similar Al2O3 content as phlogopite from the Delitzsch Complex, but plot over more variable Mg#. (d) Al2O3 content (wt.%) vs. TiO2 content (wt.%) plot. Phlogopite from the Delitzsch alnöite overlaps with group C1 of phlogopite from the Delitzsch carbonatites and with phlogopite data from minette-like rocks and other alnöites, even for samples with very high TiO2 contents.

Table 6 Apparent isochron ages and modeled initial isotopic 87Sr/86Sr compositions of phlogopite samples from carbonatites from the Delitzsch Complex, Germany. Sample

Apparent agea

Modeled initial isotopic

87

Sr/86Sr composition

MSWD

Modeled initial isotopic 87Sr/86Sr compositions based on alnöite whole-rock signaturea W2 71.7 ± 5.5 Ma 0.703479 ± 75 8.3 SER 95.0 ± 15 Ma 0.703360 ± 20 61 KMD6 79.9 ± 3.9 Ma 0.703434 ± 56 6.9 KMD8 75.1 ± 4.4 Ma 0.703461 ± 66 10.8 Note: calculated apparent ages are only model ages as the modeling is focused on the initial isotopic 87Sr/86Sr composition of xenocrystic phlogopite from the Delitzsch carbonatites. a Isochrons were calculated using Isoplot (Ludwig, 2009), with the 87Rb decay constant recommended by Minster et al. (1982; 1.402·10−11 a−1). This decay constant is considered to be more accurate than the widely used constant of 1.42·10−11 a−1 (Steiger and Jäger, 1977) and, thus, enables a better comparison between Rb–Sr and U–Pb age determinations (for discussion see Amelin and Zaitsev, 2002). Standard 2-sigma uncertainties of ±2.0% for Rb/Sr ratios were applied in the isochron age calculations. The 2σm uncertainty of 87Sr/86Sr refers to the last digit(s).

accurately than the calculated initial isotopic Sr composition of phlogopite based on an age duration between 75 Ma and 71 Ma. Calculated isochrons of xenocrystic phlogopite based on an initial 87Sr/ 86Sr composition of 0.70336, however, yield model ages that are older than the U–Pb age of the dolomite–carbonatite and the Rb–Sr age of the alnöite (Table 6) and are in conflict with the determined emplacement sequence after Röllig et al. (1995). A possible explanation for this discrepancy is that this phlogopite (i) crystallized from a melt with slightly more radiogenic 87Sr/ 86Sr, which would result in a higher initial 87Sr/ 86Sr value or (ii) differs from phlogopite from the other carbonatite samples in that the melt, from which it crystallized, evolved significantly during phlogopite crystallization. Phlogopite from sample SER differs from phlogopite from the other carbonatite samples by a pronounced compositional zoning (Table 5) with a complete fractionation series from high Mg# and high Cr2O3 contents from the core to low Mg# and low Cr2O3 to the rim. All other phlogopite samples show a more homogeneous composition, which

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

implies that they were mostly equilibrated with their source. It is quite possible that the zoning of sample SER indicates that phlogopite was not in full equilibrium with the source from which phlogopite from the alnöite and the other carbonatites crystallized. Nonetheless, the results of modeling of the initial Sr isotopic composition of xenocrystic phlogopite from the Delitzsch carbonatites give a close range of initial isotopic Sr compositions, which overlaps with the initial isotopic Sr composition of the alnöite. As the chemical composition of xenocrystic phlogopite from the carbonatites implies that phlogopite in the Delitzsch carbonatites are mostly related to lamprophyric phlogopite, it is likely that the xenocrystic phlogopite in the Delitzsch carbonatites is derived from an ‘alnöitic’ source. 5.4. Short duration emplacement of the Delitzsch Complex The results of baddeleyite dating of Stage V dolomite–carbonatite and phlogopite dating of Stage II alnöite indicate a short duration for the main phase of carbonatite and lamprophyre magmatism, which agrees the modeling of the initial Sr isotopic compositions of Stage III and VI carbonatites. Early K–Ar ages (Kozyrev, 1977; Fedoriziev et al., 1989) from the Delitzsch Complex indicated a broad range of emplacement ages for carbonatites and ultramafic lamprophyres. This discrepancy most likely is caused by the correction of initial 40Ar. The correction of non-radiogenic 40Ar by assuming that it is dominated by atmospheric argon with 40Ar/36Ar = 295.5 (Steiger and Jäger, 1977) systematically underestimates initial 40Ar, if it is derived from a source with high 40Ar/ 36Ar, such as the mantle (40Ar/36Ar >7000, e.g. Farley and Craig, 1994). Underestimation of initial 40Ar results in too old apparent K–Ar ages (examples for metamorphic rocks are given by Lee et al., 1990 and Boundy et al., 1997). Different contributions of excess Ar may also account for contrasting apparent K–Ar ages within the same dike (e.g., 73 Ma and 100 Ma, cf. Kozyrev, 1977; Fedoriziev et al., 1989). The results of our Rb–Sr and U–Pb age overlap at the lower limit of earlier K–Ar results, which is in line with the interpretation of excess mantle argon in Delitzsch phlogopite. Thus, earlier determined K–Ar ages define a too broad age range for the emplacement of the Delitzsch Complex and only the youngest K–Ar ages may be close to the emplacement age. Calculated ages for the carbonatite samples based on the isotopic whole-rock composition of the alnöite, yield model ages that define a relative narrow emplacement range, except for the sample SER (Table 6). The two solid ages of the alnöite (Section 4.2) and the dolomite–carbonatite (Section 4.1), respectively lie inside the modeled emplacement range. These results support the earlier observations from Röllig et al. (1995), who assumed a close genetic and temporal relation between carbonatites and ultramafic rocks on the basis of crosscut relationships, but do not correlate with the determined relative eruption sequence in a complete manner. It is possible that the emplacement of the Delitzsch Complex occurred as multiple pulses with more than one eruption related to a special type of rock in a geologic short time. It should be noted that the above discussion ignores the possibility of phlogopite and baddeleyite to have formed before eruption and being not reset completely. In this case the youngest ages would provide the best age estimation for the emplacement of the carbonatites. If baddeleyite and phlogopite had been picked up from metasomatized mantle, these minerals may have formed already during the metasomatism of this mantle, which occurred before the carbonatite and ultramafic lamprophyre magmatism. Examples of minerals that did not reset their geochronologic system even at very high temperatures have been demonstrated for the Rb–Sr system in granulites (e.g. Kühn et al., 2000) and for the Ar–Ar in mantle phlogopite (Hopp et al., 2008). These minerals did not reset their isotopic clocks at temperatures far above the widely used temperature estimates for their isotopic closure as daughter isotopes are lost from the crystal by diffusion and are not been taken up by crystals in their surroundings or are not transported away and therefore stay in its host-mineral. Such closed

149

system behavior occurs in dry systems or rocks with no minerals that can accommodate Sr in their structure (e.g. Kühn et al., 2000; Romer and Rötzler, 2011). Given the high compatibility of Sr in carbonatite melts, phlogopite in such a melt at mantle depth and temperature generally should reset the Rb–Sr system. 6. Conclusions New U–Pb ages on baddeleyite from a dolomite–carbonatite (beforsite) and Rb–Sr ages on phlogopite from an alnöite (ultramafic lamprophyre) combined with modeling of initial 87Sr/ 86Sr compositions on carbonatites supersede earlier K–Ar ages that indicated an broad emplacement of the Delitzsch Complex between 110 Ma and 70 Ma. Our new age data demonstrate that (i) the emplacement of the main phase of the subvolcanic stockwork (Stages II–V) of the Delitzsch Complex occurred in a short time (75–71 Ma) and that (ii) carbonatite and lamprophyre magmatism seem to be coeval within the calculated uncertainties, which indicates a close genetic relation between lamprophyres and carbonatites of the Delitzsch Complex. The chemical composition of phlogopite from the carbonatites is similar to the one of phlogopite that crystallized in a lamprophyric environment and does not show the chemical composition typical for phlogopite crystals or cumulates crystallized from a carbonatitic melt. High TiO2 contents in phlogopite and an enrichment in Cr2O3 rule out an origin from a carbonatite source. The chemical composition of such phlogopite is more typical for minette‐like mica and many of the ‘secondary metasomatic’ mica of kimberlites and mantle xenoliths, whereas phlogopite of ‘primary metasomatic’ origin (i.e. phlogopite bearing mantle peridotite) or ‘MARID’ like micas are characterized by higher Mg#, lower TiO2 and Al2O3. Thus, phlogopite in the Delitzsch carbonatite samples is xenocrystic. The xenocrystic nature of phlogopite also is in line with the modeling of the initial 87Sr/ 86Sr isotopic composition of phlogopite. The determined isotopic Sr initials of phlogopite from the carbonatites resemble the isotopic Sr initial of phlogopite from the alnöite, which favors a cogenetic evolution of carbonate and lamprophyric melts, in particular phlogopite from lamprophyres and carbonatites are derived from one parental melt. Acknowledgments We thank Oona Appelt and Dieter Rhede (both GFZ German Research Centre for Geosciences, Potsdam) for their help with the microprobe analysis and Bodo Ehling and Carl-Heinz Friedel (both Landesamt für Geologie und Bergwesen, Sachsen Anhalt, Halle) and Jürgen Wasternack (Biesenthal) for their help with sampling of the drill cores and geological discussions. Further we thank A. Hiller, geological archive of the Wismut GmbH Chemnitz for literature research. We are grateful to Johannes Glodny (GFZ German Research Centre for Geosciences, Potsdam) for his help and advice with the preparation of the phlogopite samples. We thank the two reviewers Hugh O'Brien and Tom Anderson as well as Guest Editor Sebastian Tappe for constructive and detailed reviews. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.chemgeo.2012.09.026. References Amelin, Y., Zaitsev, A.N., 2002. Precise geochronology of phosorites and carbonatites: the critical role of U-series disequilibrium in age interpretations. Geochimica et Cosmochimica Acta 66, 2399–2419. Anthes, G., Reischmann, T., 2001. Timing of granitoid magmatism in the eastern MidGerman crystalline rise. Journal of Geodynamics 31, 119–143. Bachinski, S.W., Simpson, E.L., 1984. Ti-phlogopites of the Shaw's Cove minette: a comparison of micas with other lamprophyres, potassic rocks, kimberlites, and mantle xenoliths. American Mineralogist 69, 41–56.

150

J.C. Krüger et al. / Chemical Geology 353 (2013) 140–150

Bankwitz, P., Schneider, G., Kämpf, H., Bankwitz, E., 2003. Structural characteristics of epicentral areas in central Europe: study case Cheb Basin (Czech Republic). Journal of Geodynamics 35, 5–32. Bell, K., Blenkinsop, J., 1989. Neodymium and strontium isotope geochemistry of carbonatites. In: Bell, K. (Ed.), Carbonatites—Genesis and Evolution. Unwin Hyman, London, pp. 278–300. Bell, K., Kjarsgaard, B.A., Simonetti, A., 1998. Carbonatites — into the twenty-first century. Journal of Petrology 39, 1839–1845. Boundy, T.M., Hall, C.M., Li, G., Essene, E.J., Halliday, A.N., 1997. Fine-scale isotopic heterogeneities and fluids in the deep crust: a 40Ar/39Ar laser ablation and TEM study of muscovites from a granulite–eclogite transition zone. Earth and Planetary Science Letters 148, 223–242. Claesson, S., Vetrin, V., Bayanova, T., Downes, H., 2000. U–Pb zircon ages from a Deconian cabonatite dyke, Kola Peninsula, Russia; a record of geological evolution from the Archaean to the Paleozoic. Lithos 51, 95–1008. Delaney, J.S., Smith, J.V., 1980. Chemistry of micas from kimberlites and xenoliths‐II. Primary- and secondary-textured micas from peridotites xenoliths. Geochimica et Cosmochimica Acta 44, 857–872. Farley, K.A., Craig, H., 1994. Atmospheric argon contamination of ocean island basalt olivine phenocrysts. Geochimica et Cosmochimica Acta 58, 2509–2517. Fedoriziev A.M., Vieweg, M., Bräuer, H., Herrmann, G., Selzov, B.M., Arakeljanitz, M.M., Anderson, E.B., Antipov, a.A., Sorokina, G.G., 1989. Results of isotope — chronological research of rocks and ores from southern part of the GDR in between 1983–1989. unpublished report (in Russ.), Geological Archiv, Wismut GmbH, Archiv-Nr. 54846, 241 p. Franke, W., 2000. The mid-European segment of the Variscides: tectonostratigraphic units, terrane boundaries and plate tectonic evolution. In: Franke, W., Haak, V., Oncken, O., Tanner, D. (Eds.), Orogenic Processes: Quantification and Modelling in the Variscan Belt: Geol. Soc., London, Spec. Publ., 179, pp. 35–61. Gittins, J., 1989. The origin and evolution of carbonatite magmas. In: Bell, K. (Ed.), Carbonatites—Genesis and Evolution. Unwin Hyman, London, pp. 405–427. Grégoire, M., Bell, D.R., 2002. Trace element geochemistry of phlogopite-rich mafic xenoliths: their classification and their relationship to phlogopite-bearing peridotites and kimberlites revisited. Contributions to Mineralogy and Petrology 142, 603–625. Harmer, R.E., 1997. Carbonatites from source to surface. Journal of African Earth Sciences 25, 1–3. Harmer, R.E., Gittins, J., 1998. The Case for Primary, Mantle-derived Carbonatite Magma. Journal of Petrology 39, 1895–1903. Hopp, J., Trieloff, M., Brey, G.P., Woodland, A.B., Simon, N.S.C., Wijbrans, J.R., Siebel, W., Reitter, E., 2008. 40Ar/39Ar-ages of phlogopite in mantle xenoliths from South African kimberlites: evidence for metasomatic mantle impregnation during the Kibaran orogenic cycle. Lithos 106, 351–364. Kiseeva, E.S., Yaxley, G.M., Herman, J., Litasov, K.S., Rosenthal, A., Kamenetsky, V.S., 2012. An experimental study of carbonated eclogite at 3.5–5.5 GPa—implications for silicate and carbonate metasomatism in cratonic mantle. Journal of Petrology 53, 727–759. Kozyrev I.I., Turkin, l.A., Moltschnov, W.I., Selzov, B.M., Buruikin, a.A., Grigoriev, W.W., Krupin, G.N., Tretjakov, B.G., Scherbakov, A.S., Seiler, H., Müller, B., Fasman, U, Schenke, G., 1977. Results of geological prospection in the area of Halle-Delitzsch in between 1971–1976. unpublished report (in Russ.), Geological Archiv, Wismut GmbH, Archiv-Nr. U4080, 129 p. Krogh, T.E., 1973. A low-contamination method for hydrothermal decomposition of zircon and extraction of U and Pb for isotopic age determinations. Geochimica et Cosmochimica Acta 37, 485–494. Kroner, U., Hahn, T., Romer, R.L., Linnemann, U., 2007. The Variscan orogeny in the Saxo-Thuringian zone — heterogeneous overprint of Cadomian/Paleozoic PeriGondwana crust. Geological Society of America Special Papers 423, 153–172. Kroner, U., Linnemann, U., Romer, R.L., 2010. The Saxo-Thuringian Zone of the Variscan Orogen as part of Pangea. In: Linnemann, U., Romer, R.L. (Eds.), Pre-Mesozoic Geology of Saxo-Thuringia — From the Cadomian Active Margin to the Variscan Orogen. Schweizerbart, Stuttgart. Kühn, A., Glodny, J., Iden, K., Austrheim, H., 2000. Retention of Precambrian Rb/Sr phlogopite ages through Caledonian eclogite facies metamorphism, Bergen Arc Complex, W-Norway. Lithos 51, 305–330. Lee, J.K.W., Onstott, T.C., Hanes, J., 1990. An 40Ar/39Ar investigation of the contact effects of a dyke intrusion, Kapuskasing Structural Zone, Ontario. Contributions to Mineralogy and Petrology 105, 87–105. Ludwig, K., 2009. Isoplot v. 3.71 — A geochronological toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publication 4, 70 p. Meissner, B., 1967. Magnetic anomaly of Delitzsch. unpublished report, VEB GFE, Freiberg. Minster, J.F., Birck, J.L., Allègre, C.J., 1982. Absolute age of formation of chondrites studied by the 87Rb-87Sr method. Nature 300, 414–419. Morogan, V., Wolley, A.R., 1988. Fenitization at the Alnö carbonatite complex, Sweden; distribution, mineralogy and genesis. Contributions to Mineralogy and Petrology 100, 169–182.

Nasraoui, M., Toulkeridis, T., Clauer, N., Bilal, E., 2000. Differentiated hydrothermal and meteoric alterations of the Lueshe carbonatite complex (Democratic Republic of Congo) identified by REE study combined with a sequential acid-leaching experiment. Chemical Geology 165, 109–132. Nelson, D.R., Chivas, A.R., Chapell, B.W., McCulloch, M.T., 1987. Geochemical and isotopic systematics in carbonatites and implications for the evolution of ocean–island sources. Geochimica et Cosmochimica Acta 52, 1–7. Nelson, D.R., Chivas, A.R., Chappell, B.W., McCulloch, M.T., 1988. Geochemical and isotope systematics in carbonatites and implications for the evolution of ocean–island sources. Geochimica et Cosmochimica Acta 52, 1–17. Platt, R.G., Mitchell, R.H., 1982. The Marathon Dikes: ultrabasic lamprophyres from the vicinity of McKellar Harbour, N.W. Ontario. American Mineralogist 67, 907–916. Reguir, E.P., Chakhmouradian, A.R., Halden, N.M., Malkovets, V.G., Yang, P., 2009. Major- and trace element compositional variation of phlogopite from kimberlites and carbonatites as a petrogenetic indicator. Lithos 112S, 372–384. Rock, N.M.S., 1991. Lamprophyres. Blackie, Glasgow. Röllig, G., Kampe, A., Viehweg, M., 1989. Zur Wirksamkeit von äquatorialen und meridionalen Strukturen im Gebiet der Mitteldeutschen Schwelle zwischen Halle und Schönwalde. Zeitschrift für Angewandte Geologie 35, 297–300. Röllig, G., Viehweg, M., Reuter, N., 1990. The ultramafic lamprophyres and carbonatites of Delitzsch/GDR. Zeitschrift für Angewandte Geologie 36, 49–54. Röllig, G., Kampe, A., Steinbach, V., Ehling, B.-C., Wasternack, J., 1995. Der Untergrund des deutschen Braunkohlereviers. Zeitschrift für Geologische Wissenschaften 23, 3–26. Romer, R.L., Schärer, U., Steck, A., 1996. Alpine and pre-Alpine magmatism in the rootzone of the western Central Alps. Contributions to Mineralogy and Petrology 123, 138–158. Romer, R.L., Förster, H.-J., Breitkreuz, C., 2001. Intracontinental extensional magmatism with a subduction fingerprint: the late Carboniferous Halle Volcanic Complex (Germany). Contributions to Mineralogy and Petrology 141, 201–221. Romer, R.L., Heinrich, W., Schröder-Smeibidl, B., Meixner, A., Fischer, C.-O., Schulz, C., 2005. Elemental dispersion and stable isotope fractionation during reactive fluidflow and fluid immiscibility in the Bufa del Diente aureole, NE-Mexico: evidence from radiographies and Li, B, Sr, Nd, and Pb isotope systematics. Contributions to Mineralogy and Petrology 149, 400–429. Romer, R.L., Rötzler, J., 2011. The role of element distribution for the isotopic dating of metamorphic minerals. European Journal of Mineralogy 23, 17–33. Seifert, W., Kämpf, H., Wasternack, J., 2000. Compositional variation in apatite, phlogopite, and other accessory minerals of the ultramafic Delitzsch complex, Germany: implications for cooling history of carbonatites. Lithos 53, 81–100. Standke, G., 1995. Horizontkarten des Tertiärs in Sachsen. Zeitschrift für Geologische Wissenschaften 23, 103–117. Steiger, R.H., Jäger, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters 36, 359–362. Sweeny, R.J., Thompson, A.B., Ulmer, P., 1993. Phase relations of a natural MARID composition and implications for MARID genesis, lithospheric melting and mantle metasomatism. Contributions to Mineralogy and Petrology 115, 225–241. Tappe, S., Foley, S.F., Jenner, G.A., Kjarsgaard, B.A., 2005. Integrating ultramafic lamprophyres into the IUGS classification of igneous rocks: rationale and implications. Journal of Petrology 46, 1893–1900. Tappe, S., Steenfelt, A., Heaman, L.M., Simonetti, A., 2009. The newly discovered Jurassic Tikiiusaaq carbonatite–aillikite occurrence, West Greenland, and some remarks on carbonatite–kimberlite relationships. Lithos http://dx.doi.org/10.1016/j.lithos. 2009.03.002. Tappe, S., Pearson, D.G., Nowell, G., Nielsen, T., Milstead, P., Muehlenbachs, K., 2011. A fresh isotopic look at Greenland kimberlites: cratonic mantle lithosphere imprint on deep source signal. Earth and Planetary Science Letters 305, 235–248. Taylor Jr., H.P., Frechen, J., Degens, E.T., 1967. Oxygen and carbon isotope studies of carbonatites from the Laacher See District, West Germany and the Alnö District, Sweden. Geochimica et Cosmochimica Acta 31, 407–480. Wagner, G.A., Coyle, D.A., Duyster, J., Henjes-Kunst, F., Peterek, A., Schröder, B., Stöckhert, B., Wemmer, K., Zulauf, G., Ahrendt, H., Biscoff, R., Hejl, E., Jacobs, J., Menzel, D., Lal, N., Van den haute, P., Vercoutere, C., Welzel, B., 1997. Post-Variscan thermal and tectonic evolution of the KTB site and its surroundings. Journal of Geophysical Research 102, 221–232. Wand, U., Stiehl, G., Röllig, G., Mühle, K., Gerstenberger, H., Haase, G., 1989. O, N, C and Sr isotopes characteristics of carbonatites of the GDR. 5th. Working Meeting Isotopes in Nature. Leipzig, Proceedings, pp. 421–436. Wasternack, J., 2008. Ultramafit-Karbonatit-Komplex von Delitzsch. In: Pälchen, W., Walter, H. (Eds.), Geologie von Sachsen: Geologischer Bau und Entwicklungsgeschichte. Schweizerbart, Stuttgart, pp. 478–482.