230

http://journals.tubitak.gov.tr/earth/

Turkish Journal of Earth Sciences Turkish J Earth Sci(2015) 24: 230-248© TÜBİTAKdoi:10.3906/yer-1407-21

New LA-ICP-MS U-Pb zircon dating for Strandja granitoids (SE Bulgaria): evidence for two-stage late Variscan magmatism in the internal Balkanides

Philip MACHEV1,*, Valentin GANEV2, Laslo KLAIN1

1Faculty of Geology and Geography, St. Kliment Ohridski University, Sofia, Bulgaria2Institute of Mineralogy and Crystallography, Bulgarian Academy of Sciences, Sofia, Bulgaria

* Correspondence: machev@gea.uni-sofia.bg

1. IntroductionThe questions about the nature and origin of the Variscan (Carboniferous-Permian) granitoid magmatism in Bulgarian territory have long been debated and here we present three groups of granitoids belonging to this event. The first one represents granitoid plutons in the external Balkanides (Stara Planina Ca-alkaline Formation; Dimitrov, 1946), which are best exposed in the Western Stara Planina. They were intruded in nonmetamorphosed sedimentary country rocks. Their ages are in the range between 311.9 ± 4.1 (St. Nikola pluton) and 304.6 ± 4.0 Ma (Petrohan pluton) (Carrigan et al., 2005), or 314 ± 4.8 Ma for Vezen pluton (Kamenov et al., 2002). These granitoids are correlated with the Variscan granitoids from West Europe (Dimitrov, 1946; Jaranoff, 1969).

The second group includes the granitoids from the intermediate Balkanides (Sredna Gora Zone). They are subdivided into three suites of magma emplacement (Zagorchev et al., 1973). The first and presumably oldest intrusive suite consists primarily of biotite-bearing granites and hornblende-bearing granodiorites. Granitoids of this suite are strikingly similar to the granitoids exposed in the Western Stara Planina Mountains. The second and third intrusive suites consist of strongly peraluminous, two-

mica leucocratic granites. The Sredna Gora granites were intruded in high-grade, partly migmatized metamorphic rocks and their age ranges between 312.0 ± 5.4 Ma (Koprivshtitsa pluton) and 289.5 ± 7.8 (Strelcha pluton). The protolith age of the metamorphites is 616–595 Ма and the age of metamorphism is 336 Ма (Сarrigan et al., 2006). Velichkova et al. (2001) reported data for a retrograde metamorphic event along shear zones in the range of 120–78 Ма.

The third group is represented by the granites in the Rhodope massif. They crop out in the cores of metamorphic domes (Arda, Byala Reka, and Kesebir) and differ in age: 300 ± 11 Ma in Arda (Peytcheva et al., 2004); 256 ± 27, 255 ± 2.6, 331 ± 8, and 296 ± 6.6 Ma in Byala Reka (Peytcheva et al., 1992, 1995), and 328 Ma in Kesebir (Peytcheva et al., 1998). The main feature of granites from the Rhodope massif is that they were not uniformly metamorphosed (up to migmatization in the Arda dome) in Late Alpine time: 36.5 ± 0.5 Ma in Arda (Peytcheva et al., 2004), 34.8 ± 0.8 and 35 ± 2 Ma in Byala Reka (Peytcheva, 1997), and 35–36 Ma in Kesebir (Peytcheva et al., 1998).

In this sense, the granitoids from Strandja can be attributed to the first group of granitoids.

The pre-Cretaceous metamorphic basement in SE

Abstract: The Strandja Massif (Sakar-Strandja Zone) forms an important link between the Balkan Zone (external Balkanides) of Bulgaria, which is commonly correlated with the Variscan orogen in Central Europe, and the Western Pontides of Turkey. The Bulgarian part of the massif is composed of a metamorphic basement (various granite gneisses, paragneisses, and schists) traditionally interpreted as having Precambrian age, Triassic-Jurassic metasedimentary cover, and Upper Cretaceous volcanosedimentary sequences. The basement is intruded by large granitic plutons of Variscan age that are widespread mostly across Turkish territory. New LA-ICP-MS data support the suggestion of Variscan granitoid magmatism in the studied area but do not confirm the presence of Precambrian rocks. Furthermore, two stages of magmatism are determined in relation to the Variscan metamorphism and deformation. The first one (301.9 ± 1.1 Ma) is represented by strongly deformed metagranites and thus is interpreted as syntectonic, while the second one is relatively younger (293.5 ± 1.7 Ma) and postmetamorphic.

Key words: Strandja, Variscan granitoid magmatism, SE Bulgaria, geochronology

Received: 25.07.2014 Accepted/Published Online: 02.02.2015 Printed: 29.05.2015

Research Article

231

MACHEV et al. / Turkish J Earth Sci

Bulgaria and NW Turkey forms part of a major early Alpine orogeny known as the Strandja Zone (Chatalov, 1990), Strandzhides (Gochev, 1991), Strandja Massif (Okay et al., 2001; Sunal et al., 2006, 2008, 2011; Natal’in et al., 2012; etc.), or Sakar-Strandja Zone (SSZ), constituting part of the internal Balkanides according to the tectonic subdivision of Bulgaria (Ivanov, 1998) (Figure 1a). On the basis of structural, stratigraphic, and metamorphic criteria, Gerdjikov (2005) distinguished three rock complexes in the SSZ. They are, from top to bottom, the Veleka, Strandja, and Sakar units.

The Veleka Unit is the most recently named unit in the Strandja Zone (Dabovski et al., 2002), and it is also the least known. It comprises two stratigraphically and lithologically distinct entities: the Strandja allochthon (the allochthon of the Zabernovo nappe; Chatalov, 1990) and the phyllite-marble complex from the Dervent heights. According to Chatalov (1990), the Zabernovo nappe is built of Strandja-type Triassic metamorphic rocks marking the Paleo-Tethyan suture. This view was unquestioningly used in a number of tectonic models (e.g., Şengör et. al., 1984). It now appears that at least the subduction-accretion complex part of the Zabernovo nappe consists of Paleozoic rocks (Sergeeva et al., 1979; Boncheva and Chatalov, 1998; Lilov and Maliakov, 2001). The continuation of the unit into the territory of Turkey is enigmatic.

The Strandja Unit consists of Variscan crystalline basement and granitoids (Okay et al., 2001; Sunal et al., 2006, 2008, 2011; Natal’in et al., 2012; etc.), outcropping mostly onto Turkish territory. It is overlain by Late Permian to Middle Jurassic sediments and volcanic rocks (Figure 1b). The data about the presence of older Cadomian metagranites (534.5 ± 4.7 and 546.0 ± 3.9 Ma) in the Strandja Massif were reported by Yilmaz Şahin et al. (2014). It is noteworthy that the presence of high-pressure rocks (eclogites) in the Strandja Zone has been only indicated in the Variscan gneisses from the eastern part of the Dervent heights (Dabovski et al., 1993).

The basement includes a wide range of metamorphic and igneous rocks. Evaporation Pb-Pb ages of detrital zircons show that biotite schists constituting the central part of the metamorphic basement (on Turkish territory) were deposited later than 430 Ma and prior to 315 Ma. Biotite schists exposed along the southern boundary of the basement were deposited between 300 and 271 Ma (Sunal et al., 2008) while the Bt-Ms and Hbl-Bt orthogneisses have Carboniferous ages of 314.7 ± 2.6 Ma and 312.3 ± 1.7 Ma, respectively (Sunal et al., 2006).

The metamorphic basement was intruded by Permian granites (245–271 Ma for Kirklareli granites) (Okay et al., 2001; Sunal et al., 2006).

All pre-Cenomanian rocks have undergone one or more episodes of regional metamorphism. Furthermore, the

Paleozoic basement rocks were involved in a pre-Triassic (pre-Permian?) regional metamorphic event but the exact timing of this event is disputed. Okay et al. (2001) suggested an early Permian age synchronous to the emplacement of the Kırklareli granites. Sunal et al. (2006, 2008) inferred that the Late Paleozoic deformation and metamorphism predated the granite emplacement. This metamorphic event affected only the pre-Permian basement, whereas the mid-Mesozoic regional metamorphism affected both the basement and the cover rocks. Both sequences underwent a greenschist to low-amphibolite (epidote-amphibolite) facies metamorphism in a compressional regime in Late Jurassic to Early Cretaceous time. The ages of peak or near-peak metamorphism are clustered between 157.7 ± 1.5 and 162.3 ± 1.6 Ma and subsequent cooling occurred diachronously between 153.9 ± 1.5 Ma and 134.4 ± 1.3 Ma (Sunal et. al., 2011). Ar/Ar amphibole and white mica dating in the neighboring Sakar Zone yields ages ranging between 144 and 136 Ma, constraining the age of the main tectonic event (Neubauer et al., 2010).

Variscan granites in the Strandja unit have two ages: “older” in the range of 308–315 Ma (Carboniferous) and “younger” (Carboniferous - Permian) in the range of 309–257 Ma (Okay et al., 2001; Sunal et al., 2006; Natal’in et al., 2012). Evidence about the presence of younger granites (249.4 ± 1.5) in Strandja was given by Yilmaz Şahin et al. (2012).

The metamorphic basement of the Sakar Unit is divided in two stratigraphic units (Gerdjikov, 2005): the volcanoterrigenous complex (Upper Permian?) and the Topolovgrad Group (Lower-Middle Triassic; Chatalov, 1990). Both units were regarded as a coherent section by Ivanov et al. (2001). In terms of metamorphism and synmetamorphic fabric, the Sakar Unit is more similar to the adjacent Rhodope Zone (for further information, see Gerdjikov, 2005).

From a general tectonic point of view, the position of the Strandja Massif in the Alpine orogen of Bulgaria has been long discussed (Chatalov, 1990; Natal’in et al., 2012 and references there). All concepts can be generalized as follows: 1) the Strandja Massif is a part of the Rhodope-Pontide continental fragment originating from Gondwanaland (Şengör, 1984; Şengör et al., 1984; Sunal et. al., 2008) and collided with Eurasia in the Triassic-Early Jurassic (Cimmerian orogeny) and formed the Paleo-Tethyan suture (Şengör, 1984; Şengör et al., 1984); 2) the Strandja Massif is a part of Eurasia (Ustaömer and Robertson, 1993), i.e. prior to the Late Paleozoic the Rhodope-Pontide fragments belonged to Eurasia. These two initial models imply that the magmatic activity of the Strandja Massif during the Late Paleozoic-Triassic occurred in arc and back-arc tectonic settings. 3) The Strandja Massif is a part of the European Variscan orogen,

232

MACHEV et al. / Turkish J Earth Sci

Figure 1. a- Tectonic subdivision of Bulgaria after Ivanov (1998); and b- simplified geological map of Sakar-Strandja zone after Gerdjikov (2005).

Topo

lovg

rad

42o12

'0"N

42o0'0"

N

41o36

'0"N

26o0'0"

E28

o0'0"

E

010

20km

Edirne

Rhodope

Massif

Has

kovo

Harman

li

Maritza

Elhov

o

Kırk

lareli

Malko

Tarnov

o

Tza

revo

41o48

'0"N

L1

L2

Rhodopemassif

Sakar-Strandja

zone

Moesian

platform

SC

South

Carpathians

Balkanides

Outher

Kraiste

zone

In

te

rn

al

Ba

lk

an

id

es

Serbia

F.Y.R.O.M

Greece

Turkey

BlackSea

Balk

anid

es

I nt e

rm

edi a

t e

Tertiary

sedimentary

andvo

lcanicrocks

LateCretaceousgranito

ids

LateCretaceousvo

lcanicandsedimentary

rocks

Saka

rUnit LateJurassic-L

owCretaceousgranito

ids

Perm

ian-Triassicmetasedim

entsandmetavolcanites

VelekaUnit

Triassicmetasedim

entsandmetavolcanites

Strand

jaUnit

Perm

ian-MiddleJurassicmetasedim

entsandmetavolcanites

Paleoziocgranito

ids

Variscanbasement

Thrustzon

es

Lesovo

fault

Studiedarea

a

b

233

MACHEV et al. / Turkish J Earth Sci

e.g., it represents an eastern continuation of the Central European Variscan belt (Okay et al., 2001) in which the orogeny occurred not in the mid-Carboniferous as in Europe and Bulgaria, but later in the early Permian. This orogeny resulted in the metamorphism and emplacement of widespread Early Permian granites. The most popular opinion is that the Variscan orogeny in the Balkans was related to the Late Carboniferous collision of the Balkan and Moesia continental blocks (Yanev, 2000).

The aim of this paper is to provide new data about the petrology and age of two granitic bodies outcropping in the Bulgarian part of the Strandja Massif close to the Turkish border and to highlight the problem of Variscan magmatism in this part of SE Bulgaria.

2. Geological settingTraditionally in the Bulgarian geological literature, high-grade metamorphic rocks of the Strandja Massif have been correlated with those from the Rhodope massif, i.e. they are considered to be of Precambrian age (Chatalov et al., 1995 and references there). According to this assumption, they were described as different types of migmatites (banded or porphyroblastic), two-mica gneisses, leucocratic gneisses, marbles, and paraamphibolites in irregular alternation. They are shown in Figure 1b as Variscan basement on the basis of published geochronological data on Turkish territory (Okay et al., 2001; Sunal et al., 2006, 2008, 2011; Natal’in et al., 2012; etc.) and our investigations. Without any personal geochronological data, Gerdjikov (2005) assumed a Hercynian age for these rocks, too.

The large granitic intrusions in the massif were interpreted as postmetamorphic (regarding the Precambrian metamorphism) and of Variscan age according to the traditional concept for Precambrian age of the host rocks (Chatalov et. al., 1995).

Our field studies were focused on the high-grade metamorphic rocks (Locality 1 in Figure 1b) and on the adjacent Variscan granites (Locality 2, respectively). In this particular part of the Strandja Massif the metamorphic rocks from Locality 1 are represented by two-mica gneisses, Ms-schists, leucocratic gneisses, and paraamphibolites (Figure 2a). All of the rocks in the locality are strongly weathered; hence, the sampling is difficult. The amphibolites are dark green in color, strongly foliated, and composed of plagioclase, epidote, blue-green amphibole, quartz, and rare biotite without garnet. Ms-schists are fine-grained and foliated. Quartz is the dominant mineral and under a microscope they are seen as strongly elongated bands consisting of fine grains with undulose extinction. The white mica is the next mineral whose grains form bands parallel to the quartz ones, i.e. they form the foliation of the rocks. Less commonly, big

(up to 1 mm) idiomorphic plagioclase grains oriented with the long axes parallel to foliation are also observed. They are replaced by sericite and epidote and they look like remnants from the protolithic rocks.

The two-mica gneisses contain micas (biotite and muscovite), quartz, and rare plagioclase. They are foliated to the same extent as the other metamorphic rocks in the area. On the basis of the equilibrium mineral assemblages, we can conclude that the grade of metamorphism does not exceed low-amphibolite facies (epidote-amphibolite). Our investigations show that the dominant part of the previously described as biotite gneisses (Chatalov et al., 1995) in fact represents deformed and metamorphosed granites (metagranites) (Figure 2b). Because of the high vegetation, contacts between gneisses and metagranites are rarely observed. Therefore, detailed geological mapping in the area is impossible. In the places where contacts are exposed, they are sharp but always concordant, i.e. the foliations in metagranites and in the neighbor rocks are parallel. The whole complex is ubiquitously intruded by younger (Upper Cretaceous) intrusive and volcanic rocks of basic and intermediate composition or by relatively large intrusive bodies: the gabbro intrusion in the area of Zelyazkovo village. For this gabbro without geochronological data, Kamenov (1976) suggested a Paleozoic age according to assigned Precambrian age for the host rocks.

Before the study presented here, metamorphic rocks of the region were not the subjects of any age determinations, regardless of the information about their possible Cambrian age (Savov, 1979).

The Variscan granites (Locality 2) cover large areas in the SE and E of Golyamo Bukovo and Granichar villages and in Turkey (Figure 1b). They are represented by magmatic rocks of mainly granitic composition and were described as Central Strandja batholith (Chatalov et. al., 1995). We named the southern part of the batholith Fakiya granite, which can be correlated with the Kula granites that crop out across a wide area on the Turkish border. The contacts with the metamorphic rocks are sharp and intrusive and the endocontact zones are enriched by xenoliths from the host rocks. The granites are covered by Triassic sediments or Upper Cretaceous volcanosedimentary sequences. The outcrops of the granites are very scarce. They are gray in color with pinkish potassium feldspar and massive structure, and they are commonly crosscut by melanocratic dykes of Late Cretaceous age. Along shear zones the granites are foliated (Figure 2c), but these zones are not common.

Previously there was no dating study on the Fakiya granite, but its equivalent Kula granite was dated as 271 ± 11 Ma (Okay et al., 2011).

234

MACHEV et al. / Turkish J Earth Sci

Figure 2. a- One of the rare outcrops of L1 granites host rocks west of the village of Fakiya, alternation of deep weathered amphibolites and two-mica schists. b- Outcrop of L1 granites (42°09’976”N, 27°03’484”E) with poisonous snake in the front. c- Outcrop of L2 granites close to the shear zone (42°06’40”N, 27°01’26”E).

3. Analytical methods The content of major chemical elements in the studied rocks was determined by wet chemical analysis. The REEs and other elements were measured in glass pellets by LA-ICP-MS technique using results from the wet analysis as the internal standard. The pellets were prepared by melting at 1050 °C of a homogeneous mixture of powdered rocks and Li2B4O7 in the proportion of 1:4.

Zircons from the granites were analyzed by LA-ICP-MS technique using a PerkinElmer ELAN DRC quadrupole ICP-MS connected to the NWR/ESI UP-193FX ArF excimer laser ablation system with a homogenized laser beam on the sample surface. The selected zircon grains were mounted in epoxy, striped to their middle, and finely polished. Cathodoluminescence (CL) and back-scattered electron (BSE) imaging of the selected zircon set was performed prior to the mass spectrometry measurements. The main parameters of the LA-ICP-MS instrument’s set-up are given in Table 1. The machine was calibrated and optimized to the proper U-Pb ratios at high signal sensitivity level by repeated measurements of GJ-1 natural zircon standard (Jackson et al., 2004). Another widely used natural zircon reference material in

LA-ICP-MS geochronology (Plesovice) was adopted for data verification purposes. Each five sampled unknowns were bracketed by two standard measurements from GJ-1, performed under the same analytical conditions. The offline data reduction code ‘Iolite’ (Paton et al., 2003) was implemented for calculation and correction of the measured U-Pb ratios before the final age estimation procedure performed by the Isoplot toolkit (Ludwig et al., 2003).

4. Petrology4.1. Metagranites (L1)The metagranites crop out mainly in the Fakiyska River valley between the villages of Fakiya and Gorno Yabalkovo. They are deformed and foliated rocks consisting of plagioclase, biotite, quartz, scarce potassium feldspar, and accessories. The most deformed parts are converted into typical S-C mylonites. In this case, white mica appears as a second sheet silicate in the rocks.

Plagioclase is the predominant mineral (»50%), being represented by coarse idiomorphic grains twinned after Albite and Carlsbad laws. It shows features typical for deformation during the submagmatic stage in the sense

235

MACHEV et al. / Turkish J Earth Sci

of Bouchez et al. (1992): the crystals are cracked and the cracks are filled with “clear” fine-grained quartz without undulose extinction (Figure 3a). In the weakly deformed domains the plagioclase grains are rounded, rotated, and partly recrystallized due to formation of subgrains (Figure 3b). In the strongly deformed and mylonitized parts, the plagioclase is completely recrystallized and forms the rock matrix together with quartz. Needle-shaped inclusions of apatite are abundant in the plagioclase.

The quartz is completely recrystallized and forms ribbons of fine irregular grains floating around the plagioclase clasts. The big clasts in the mylonitic parts show a chessboard pattern, e.g., evidence for high temperature deformation during the magmatic stage (Kruhl, 1996) (Figure 3c).

The potassium feldspar is rare. It is observed as single idiomorphic microperthitic grains with Carlsbad twins forming the porphyritic texture of the granites. They contain plagioclase and rare biotite inclusions orientated with their long axes parallel to the faces of the host potassium feldspar. In the mylonites the porphyroclasts are orientated parallel to the foliation with well-developed pressure shadows filled with recrystallized fine-grained quartz.

The biotite is brown to yellow and is partly recrystallized. The new grains are small and in the weakly deformed domains they form short bands orientated parallel to the foliation. Sometimes the biotite forms typical folia wrapping around plagioclase and K-feldspar. The primary magmatic biotite is preserved as relatively coarse grains in the space between the big plagioclase clasts. In the strongly deformed domains (S-C mylonite), the biotite forms thin bands and forms the foliation plane together with white mica.

The accessories except zircon comprise opaque, large (up to 3 mm) allanite and apatite.4.2. Fakiya granites (L2)The granites from L2 are coarse-grained to porphyritic, having massive structure. The rocks show poor indications of ductile deformation.

Plagioclase is the predominant mineral, forming relatively big idiomorphic grains with albite-type twins. They are almost completely replaced by fine-grained white mica (sericite) and rare epidote grains (Figure 3d). The alteration of plagioclase has also affected grains included in potassium feldspar.

The potassium feldspar (microcline) is observed as two structural types of crystals. The first is represented by large idiomorphic phenocrysts with perthitic texture rich in inclusions of plagioclase, biotite, hornblende, and sphene (Figure 3e). The second occurs as small xenomorphic grains in the rock matrix. The microcline lattice is better developed in those grains.

Biotite is the dominant mafic mineral, forming large idiomorphic grains, which are fully replaced by chlorite, ore mineral, epidote, and white mica.

Hornblende is the second mafic mineral in subordinate amounts relative to the biotite. It occurs as idiomorphic grains in the rock matrix or inclusions in potassium feldspar megacrysts. The hornblende crystals are partially replaced by chlorite.

The quartz is xenomorphic, showing undulose extinction and a chessboard pattern (Figure 3f).

Epidote is the most abundant secondary mineral, being a replacement product after biotite or forming single grains in the rock matrix as short bands and lenses.

Table 1. The main parameters of the used LA-ICP-MS facility.

ICP-MS Laser system

Type Quadrupole in first zone Type ArF excimer

Mode Standard Wavelength 193 nm

Scanning regime Peak jumping Pulse duration <5 ns

RF power 1450 W Energy density 10 J/cm2

Plasma gas flow rate 15 L/min Output laser energy 6.1 mJ

Aux. gas flow rate 0.84 L/min Focus conditions At sample surface

Neb. gas flow rate 0.86 L/min Repetition rate 5 Hz

Abl. gas flow rate (He) 0.92 L/min Spot size 35 µm

Dwell times: 208Pb, 232Th 235U, 238U, 206Pb, 207Pb

10 ms 20 ms 30 ms

Cell volume 20 cm3

236

MACHEV et al. / Turkish J Earth Sci

Figure 3. Microscopic features of L1 and L2 granites. a- Submagmatic microfracture in plagioclase (Pl) of L1 granites filled by magmatic quartz (Qtz1), surrounded by recrystallized quartz (Qtz2). CPL. b- Rounded and rotated partially recrystallized plagioclase clasts (Pl) with formation of subgrains (black arrows) between them. CPL. c- Chessboard pattern in the large clast of quartz (Qtz) in the mylonitic parts of L1 granites. CPL. d- Large plagioclase crystal (Pl) completely replaced by fine-grained white mica (sericite) and rare epidote grains. CPL. e- Perthitic potassium feldspar (Kfs) enriched by inclusions of plagioclase (Pl) and hornblende (Amp). CPL (L2 granites). f- Xenomorphic quartz grain (Qtz) with chessboard pattern. CPL (L2 granites).

The accessories include sphene, apatite, and ore mineral.4.3. Zircon morphologyThe zircons in L1 granites are observed as two different morphological types. The first is represented by idiomorphic elongated grains with well-developed

dipyramids with length/width ratio reaching 4:1. They have “simple” zonation in the CL images: oscillatory zoned core (central part of the crystal, not inherited core) and relatively homogeneous rim. The second type of zircons is partly “rounded” and the length/width ratio of the crystals is less than 2:1. They have more complicated CL zonation:

237

MACHEV et al. / Turkish J Earth Sci

the core and rim are oscillatory zoned and in many cases the zones in the core are oblique to those in the rim (Figure 4). No inherited cores are observed. Scarce apatite inclusions occur in both of the two zircon types. On the other hand, zircon is observed as inclusions in the rims of K-feldspar grains.

The zircons in L2 granites form elongated idiomorphic grains with length/width ratio of about 4:2. They are clear and pinkish-brownish in color. No inherited cores are observed in the BSE and CL images. The zircons have relatively large homogeneous cores (central part of the crystal) and rims with concentric compositional zoning represented by thin dark and light bands (Figure 5). The inclusions comprise idiomorphic apatite and drop-like quartz.

5. GeochemistryFor geochemical characterization of the rocks, three wet chemical and LA-ICP-MS analyses from each granite type were performed. Despite the limited number of analyses, some trends can be established as follows.5.1. Metagranites (L1)The rocks from this locality contain SiO2 in a narrow range between 66% and 67% and on the basis of their normative composition they correspond to monzogranites (Table 2). The Al2O3 (16.2%–16.4%) and CaO (4.8%–5.0%) contents determine the presence of normative diopside in the range

of 0.5–0.76. The rocks are calc-alkaline and metaluminous with A/CNK of <1.0. Patterns of incompatible elements in the L1 granites on the spider diagram (normalized to primitive mantle according to values presented by Lyubetskaya and Korenaga, 2007) show a regular decrease of the enrichment factor with increasing compatibility of the elements. They are also characterized by negative anomalies of Th, Nb, Sm, and Ti and positive anomalies of Sr, Eu, and Yb (Figure 6). The REE distribution pattern in general is characteristic for acid magmatic rocks with LREE enrichment and LaN/LuN ratios of 16.47–30.84. A specific feature of the rocks from this locality is the positive Eu anomaly related to the higher CaO content in the rocks in comparison with Fakiya granites. The distribution of HREEs is peculiar. After Eu the distribution trend has a negative slope down to Er, and after that it becomes positive (Figure 7).5.2. Fakiya granites (L2)The L2 granites have similar geochemical features compared to the rocks from L1 but they show some differences as well. Despite the lower Al2O3 content (15.38%–15.68%), they are normative corundum-bearing due to the higher K2O and lower CaO contents. The rocks are calc-alkaline and peraluminous with A/CNK of >1.0, and they have higher concentrations of some rare elements compared to the L1 granites (Table 2). Nevertheless, the distribution patterns of these elements are similar with

Figure 4. Representative selection of cathodoluminescence images of analyzed zircon grains from L1-type granite. The ablation points are marked by circles with solid lines and corresponding 206Pb/238U ages were used for dating. Dashed-line circles represent element measurements.

238

MACHEV et al. / Turkish J Earth Sci

Figure 5. Representative selection of cathodoluminescence images of analyzed zircon grains from L2-type granite. The ablation points are marked by circles with solid lines and corresponding 206Pb/238U ages were used for dating. Dashed-line circles represent element measurements.

0.5

1

10

100

300

Rock

/Prim

itive

Man

tle

Figure 6. Spider diagram (rock/primitive mantle) for L1 and L2 granites. The data are normalized to primitive mantle according to values presented by Lyubetskaya and Korenaga (2007).

0.1

1

10

100

200

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Rock

/Cho

ndrit

e

Figure 7. Chondrite-normalized REE pattern for the investigated granites. Normalizing data after Haskin et al. (1968).

small differences when compared to the L1 granites due to positive Sr anomaly (Figure 6). The REE distribution is characteristic for acid magmatic rocks with LREE enrichment and (La/Lu)N ratios of 17.85–30.86. The L2 granites have flat distribution of the HREEs without Eu anomaly (Figure 7).

6. Zircon chemistry, thermometry, and geochronology The REEs and other measured element contents in zircons were determinate by LA-ICP-MS method. In this case,

NIST-610 (Jochum et al., 2011) standard reference glass was used for calibration. The preliminary Si concentration in the studied zircon grains obtained by SEM (EDX) was implemented as the internal standard. The Iolite code (Paton et al., 2003) was implemented for calculation of the element contents (Table 3).

The zircons from L1 granites contain HfO2 in the range of 1.12%–1.62% and the Zr/Hf ratio is 34–52. They have Pb concentrations of 24–286 ppm, Th concentrations varying from 16 to 192 ppm, low Th/Pb ratios from 0.44 to 0.68,

L1 GranitesL1 Granites

L2 Granites

L2 Granites

239

MACHEV et al. / Turkish J Earth Sci

Table 2. Chemical composition of the L1 and L2 granites. The geographical coordinates of the geochronological samples are provided.

Sample

L1 Granites L2 GranitesC-2142°09’976’’N,27°03’484’’E

C-24 C-32 C-542°06’40’’N,27°1’26’’E

C-53 C-62

SiO2 66.86 66.95 67.03 67.05 67.85 67.29TiO2 0.45 0.41 0.39 0.56 0.46 0.54Al2O3 16.41 16.4 16.25 15.68 15.38 15.54Fe2O3 0.82 0.88 0.81 1.34 1.29 1.39FeO 1.85 1.95 1.98 1.36 1.48 1.35MnO 0.04 0.02 0.03 0.04 0.03 0.05MgO 1.55 1.5 1.56 1.59 1.46 1.63CaO 4.97 4.95 4.88 2.84 2.77 2.91Na2O 4.28 4.23 4.22 4.43 4.35 4.4K2O 1.5 1.39 1.38 2.65 2.67 2.61P2O5 0.3 0.3 0.3 0.32 0.28 0.4LOI 0.58 0.55 0.67 1.77 1.66 1.55Total 99.61 99.53 99.5 99.63 99.68 99.66Ba 726.08 726.08 753.46 1199.72 1226 1148.88Rb 31.99 31.99 32.21 64.35 65.24 62.03Sr 912.03 919 937.61 752.43 754.44 714.92Ta 0.14 0.11 0.16 0.56 0.49 0.61Nb 2.97 2.96 2.98 7.49 7.79 7.38Hf 2.01 2.01 1.78 3.89 3.66 3.12Zr 81.36 81.35 73.92 147.12 149.04 135.54Y 2.28 2.28 2.14 9.46 9.53 9.67La 11.06 10.81 11.55 29.04 30.22 30.84Ce 19.04 18.91 19.93 55.09 59.43 57.48Pr 2.04 2.06 2.09 6.09 6.42 6.3Nd 7.05 7.28 6.89 24.15 24.68 25.44Sm 1.15 1.09 1.02 4.45 3.99 4.5Eu 0.91 0.58 0.65 1.12 1.12 1.07Gd 0.51 0.88 1.02 3.33 2.46 3.69Tb 0.067 0.08 0.066 0.3 0.29 0.39Dy 0.45 0.64 0.62 1.86 2.05 1.58Ho 0.101 0.07 0.07 0.34 0.35 0.4Er 0.27 0.26 0.21 0.92 0.81 1.03Tm 0.07 0.07 0.07 0.11 0.16 0.14Yb 0.56 0.57 0.34 0.97 0.9 0.92Lu 0.07 0.07 0.04 0.1 0.18 0.16CIPW normsQ 23.74 24.47 24.67 23.61 24.71 24.46Ab 36.53 36.12 36.09 38.35 37.66 38.1An 21.3 21.84 21.54 12.51 12.4 12.39Or 8.96 8.31 8.26 16.05 16.18 15.82C 1.06 0.91 1Di 0.76 0.5 0.5

240

MACHEV et al. / Turkish J Earth Sci

and a narrow Nb concentration range from 1.0 to 2.76 ppm (Table 3). There are no significant differences in these parameters between the dark and light zones in the CL images (Figure 4). The REE distribution pattern is typical for zircons of magmatic origin (Hoskin and Schaltegger, 2003): positive Ce anomaly, weakly negative Eu anomaly (Eu/Eu* = 0.07–0.19), and strong enrichment in HREEs (Figure 8). The (Lu)N values obtained from these zircons (where N denotes the normalized to chondrite values of Haskin et al., 1968) range from 300 to 1900. Figure 8 shows that the dark zones are enriched in REEs (400–700 ppm vs. 90–264 ppm in the light zones) in general and especially in LREEs. Therefore, in these zones, La was dominantly analyzed in concentrations above the detection limit of the LA-ICP-MS facility.

The zircons from L2 granites contain variable Pb (135–696 ppm), Th (89–419 ppm), and U (276–591 ppm)

and a relatively narrow range of Th/Pb ratios (0.45–0.79), with Nb amount of 1.37–1.68 ppm and Zr/Hf ratio of 38–46 (Table 3). There are no considerable differences in these values between the cores (homogeneous part of the crystals in the CL image) and the rims. The cores have higher concentrations of REEs (300–644 ppm) compared to those in the rims (271–573 ppm) as shown in Table 3 but the rims are relatively enriched in LREEs (Figure 9). Nevertheless, La is in amounts below the detection limit in both core and rim zones. The chondrite-normalized REE patterns are similar to these of the L1 granites. The zircons from the L2 granites also have positive Ce anomaly, negative Eu anomaly (Eu/Eu* = 0.09–0.26), (Lu)N in the range of 758–1570, and Y of 304–782 ppm, and they resemble the zircons from other granitoids described in the literature (Hoskin and Ireland, 2000).

The chemical characteristics of the analyzed zircons

Table 3. Representative analyses of the zircons from L1 and L2 granites.

Sample

L1 GranitesC-21-1 C-21-1 C-21-3 C-21-3 C-21-5 C-21-5 C-21-7 C-21-7 C-21-9 C-21-12 C-21-12

Light zone Dark zone Light zone Light zone Light zone Light zone Dark zone Dark zone Dark zone Dark zone Dark zone

SiO2 32.81 32.89 31.43 32.04 31.63 33.27 32.63 32.87 32.28 32.06 32.43ZrO2 64.17 64.17 62.78 63.14 63.93 63.63 63.63 63.91 66.09 64.17 66.78HfO2 1.22 1.18 1.35 1.32 1.18 1.19 1.51 1.62 1.19 1.27 1.12Total 98.2 98.24 95.56 96.5 96.74 98.09 97.77 98.4 99.56 97.5 100.33Ti 12.8 7.3 4.3 2.5 4.2 4.2 3.2 3 4.4 4.4 12.5Ta 0.58 0.44 0.99 0.305 0.24 0.334 3.09 2.28 0.43 0.77 0.67Nb 1.201 1.201 1.61 1.041 1.23 1.23 2.76 2.25 1.28 1.57 1.35Hf 10,240 9880 11,440 11,300 9950 10,260 12,730 13,790 10,100 10,780 9590Y 349.4 803 324 197.9 145.6 110.8 883 485 778 313 1073Th 68.1 132.2 79 25.6 16.1 16.73 152 49.7 128.4 122 192.4U 137.1 154.5 196.5 72.4 44.1 49.3 567 348 169 242 245.8Pb 112 200 96 46 24.4 24.7 274 72.6 286 186 286La b.d.l. b.d.l. 0.08 b.d.l. b.d.l. b.d.l. 0.03 b.d.l. b.d.l. b.d.l. 0.05Ce 14.26 17.12 14 6.91 6.59 0.848 20.5 10.88 17.11 17.54 18.57Pr 0.025 0.192 0.026 0.02 0.02 0.02 0.058 0.02 0.33 0.013 1.21Nd 0.55 3.27 0.46 0.17 0.025 0.13 0.48 0.1 2.65 0.17 5.67Sm 1.05 5.83 0.84 0.43 0.47 0.33 1.64 0.5 5.03 1.26 7.6Eu 0.306 0.91 0.311 0.199 0.171 0.104 0.308 0.181 0.94 0.189 1.5Gd 6.1 21.3 4.35 2.38 2.28 1.42 10.8 5.07 20.5 5.4 26.8Tb 2.38 6.16 1.78 0.91 0.78 0.542 4.32 2.27 6.01 1.82 7.88Dy 29.2 72.7 22.3 13.7 10.54 7.58 67.1 35.5 71.5 24.9 99.1Ho 11.18 25.69 9.57 5.64 4.31 3.19 27 14.79 25.78 9.65 35.45Er 50 112.5 47.5 30.1 21.1 15.9 138.4 76.5 112.9 48.8 156Tm 11.81 24.5 11.84 8.04 5.33 4.27 32.8 19.8 12.55 11.14 32.77Yb 110.3 224.8 126.4 90.8 59.7 45.9 331 202.5 224.2 112.1 308.5Lu 23.37 44.3 25.5 20.49 12.97 10.3 65.9 40.4 43.6 23.8 60.6Zr/Hf 45.94 47.5 40.51 41.62 47.27 46.58 36.82 34.35 48.53 44.14 52.03

b.d.l. - Below detection limit.

241

MACHEV et al. / Turkish J Earth Sci

(relatively low Pb, U, and Th contents and negative Eu anomaly) are very similar to the zircons obtained from I-type granites (Wang et al., 2012). Their significant similarities make impossible any explicit conclusions about differences in the magma sources for both L1 and L2 granites. On the other hand, the estimated chondrite-normalized REE distribution patterns in the studied zircons greatly differ from the REE distribution patterns in zircons from granitoids of orogenic origin obtained by Belousova et al. (2002).

To determine the temperature of crystallization of the studied granites, two independent methods were used: zircon saturation thermometry (TZr) and Ti-in-zircon thermometry. TZr (Watson and Harrison, 1983) defines a hypothetical stage in the evolution of a given melt at which zirconium saturation is achieved and zircon crystallization begins. Implicit to this approach is that whole-rock major and trace element concentrations accurately approximate the melt composition at the time of zircon crystallization, which is difficult to demonstrate

Table 3. (Continued).

Sample

L2 Granites

C-51-1 C-51-1 C-51-5 C-51-5 C-51-8 C-51-8 C-51-11 C-51-11 C-51-15 C-51-15 C-51-20

Core Rim Core Rim Core Rim Core Rim Core Rim Core

SiO2 33.03 33.06 33.7 33.39 31.69 33.24 32.56 33.04 31.97 32.71 33.04

ZrO2 66.14 65.76 65.79 63.9 62.82 63.49 65.65 65.65 63.36 63.9 65.52

HfO2 1.33 1.35 1.43 1.38 1.22 1.36 1.49 1.44 1.25 1.27 1.29

Total 100.5 100.17 100.92 98.67 95.73 98.09 99.7 100.13 96.58 97.88 99.85

Ti 3.57 3.51 3.02 3.48 19.1 5.71 3.04 4.24 4.27 3.56 3.38

Ta 0.59 0.59 0.63 0.369 0.483 0.401 0.37 0.403 0.3 0.362 0.49

Nb 1.626 1.68 1.643 1.371 1.436 1.568 1.478 1.404 1.411 1.457 1.462

Hf 11,300 11,460 12,170 11,720 10,400 11,550 12,700 12,280 10,630 10,830 11,000

Y 782 647 584 304.5 491.7 404.3 329 320 4902 382.9 400

Th 166 257.2 394 183.8 178 254 146.1 190.6 127.3 137.4 89.2

U 334 452 591 365 316.2 392.3 367 405 242.2 312.9 195.8

Pb 368 484 654 323 281 413 252 313 197 222 135

La b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l. b.d.l.

Ce 32.9 42.2 40.7 26.8 31.6 79.5 19.1 30.9 31.3 26.58 19.49

Pr 0.041 0.019 0.02 0.056 0.039 0.55 0.016 0.128 0.15 0.1 0.107

Nd 1.37 0.76 0.7 1.4 1.06 14.8 0.59 3.35 2.88 0.42 0.87

Sm 4.41 2.12 2.43 1.95 2.11 5.31 0.7 1.73 2.62 1.11 1.64

Eu 1.31 0.77 0.89 0.45 0.41 0.83 0.49 0.73 0.98 0.52 0.55

Gd 20.4 11.7 11.1 5.94 8.1 9.38 4.83 5.81 11.17 5.66 7.08

Tb 5.98 4.04 3.61 1.67 3 2.68 1.8 2 3.68 2.27 2.68

Dy 75.7 54.1 51.5 24.2 42.1 37.1 26.2 27.1 42.3 29.1 32.2

Ho 26.59 21 18.28 8.51 15.34 12.46 9.63 9.44 17.47 12.19 13.63

Er 125 104.8 97.2 51.3 84.6 68.7 59.6 53.4 77.3 57.7 67

Tm 27.3 24.7 22.23 11.17 19.39 14.79 13.17 11.5 18.06 14.9 14.45

Yb 256 254 225.5 118.5 182.7 142.2 132 119.5 182.6 157.4 161

Lu 53.4 53.1 49.8 26.8 38.2 31.3 32.2 29.2 40.6 36.5 36

Zr/Hf 43.31 42.46 40 40.34 44.69 40.68 38.25 39.56 44.11 43.66 44.07

b.d.l. - Below detection limit.

242

MACHEV et al. / Turkish J Earth Sci

except under favorable circumstances (Kemp et al., 2005). Ti-in-zircon thermometry, on the other hand, provides a more direct approach to quantify ambient temperatures of zircon crystallization (Watson et al., 2006b). To apply the TZr method to both granites, we used Zr concentrations in the zircon and in the rocks obtained by LA-ICP-MS. Zr concentrations vary from 73.92 to 81.36 ppm with an average value of 78.83 ppm in the L1 metagranites and from 135.54 to 149.04 ppm (average: 143.9 ppm) in the L2 granites. In any case, we used analyses of the cores (homogeneous in the CL images of the central part of the crystals). Zircon saturation temperatures (Watson and Harrison, 1983) calculated from bulk rock compositions for the L1 samples ranged from 730 to 745 °C and in the L2 samples they revealed slightly higher temperatures (788–811 °C). If the melts produced for crystallization of both granite types were undersaturated in zirconium, as indicated by the general lack of zircon inheritance, these temperatures should be the minimum temperature that prevailed during crystal growth. The lack of older inherited cores and small zircon crystals in both types of studied granites (L1 and L2) could be a result of dissolution of smaller crystals in a melt. Dissolution of zircon would increase zircon saturation of the melt and promote formation of overgrowth on larger grains via Oswald ripening (Nemchin et al., 2001).

Ti content in the zircon is insensitive to pressure but it has a strong dependence on temperature (Watson et al., 2006b) and therefore can be used as a geothermometer. To apply this method, we used the Ti concentration in the zircons obtained by LA-ICP-MS. In the zircons from L1 granites, Ti content ranges from 2.5 to 12.8 ppm without

significant variations in concentrations between the dark and light zones (Table 3). The L2 zircons are depleted in Ti at 3.02–5.08 ppm (once again, no differences between homogeneous cores and rims). The temperatures calculated by Ti-in-zircon thermometry (Watson et al., 2006a, 2006b) are lower than those obtained by TZr (713 ± 30 °C for L1 granites and 675 ± 8 °C for L2). Despite the higher range of temperature values for the L1 granites, they are comparable to those for the L2 granites. Surprisingly lower Ti-in-zircon temperatures in magmatic and metamorphic rocks have been observed by many authors (Fu et al., 2008; Bolhar et al., 2012; Ewing et al., 2013, etc.) but no simple explanation has been provided for this phenomenon. One of the explanations is related to the accuracy of the LA-ICP-MS analyses since the Ti concentration in zircon is close to the detection limit of the facility. Zircon in a cooling magma grows over a wide range of temperatures that begin at TiO2 activity of >0.5, and it is estimated that the Ti-in-zircon thermometer underestimates temperatures at more than 70 °C if aTiO2 is unknown (Ferry and Watson, 2007). For this reason, we assume that the Zr saturation temperatures represent the starting temperatures of zircon crystallization, and the Ti-in-zircon values are due to later stages of this process.

Isotopic values and calculated ages are presented in Tables 4 and 5. Eighteen zircon grains from L1 granite were analyzed (both core and rim) and they give a precise Concordia age of 301.9 ± 1.1 Ma (Figure 10a). The obtained 206Pb/238U ages are in the range of 293.5 ± 6.9 to 310 ± 6.1 Ma (weighted mean: 302.5 ± 2.1; Figure 10b). The L2 granites (19 zircon grains) give a slightly younger Concordia age of 293.5 ± 1.7 Ma (Figure 11a)

0.0

0.1

1

10

100

1000

3000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Zirc

on/C

hond

rite

Zirc

on/C

hond

rite

3000

1000

100

10

1

0.1 L a Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Figure 8. Chondrite normalized REE pattern for zircons from L1 granites. Normalizing data after Haskin et al. (1968).

Figure 9. Chondrite-normalized REE pattern for zircons from L2 granites. Normalizing data after Haskin et al. (1968).

light zones

dark zones

homogeneous corerim

243

MACHEV et al. / Turkish J Earth Sci

and range from 282.8 ± 5.5 to 305.1 ± 7.2 Ma 206Pb/238U ages (weighted mean: 294.2 ± 3.7; Figure 11b). There are no age differences between cores and rims. Considering the high isotopic closure temperature of zircon (above 900 °C) for U and Pb (Cherniak and Watson, 2001), the zircon ages from both granite types should represent magmatic crystallization ages of the plutons and indicate that the granites have intruded the gneiss basement during the Late Carboniferous-Early Permian.

7. Discussion and conclusionsOur study does not confirm the presence of Precambrian rocks in the study area as was stated by Chatalov et al. (1995). The age of the L1 granites is close to the age of Variscan metamorphism and deformation (Okay et al., 2001) since they are strongly affected by that. This age (301.9 ± 1.1 Ma) is younger that those obtained by Sunal et al. (2006) for the host orthogneisses (314.7 ± 2.6 and 312.3 ± 1.7 Ma). Thus, we can interpret the L1 granite as

Table 4. Estimated ages of L1 granites (metagranites). *: Obtained age is presented in Figure 10.

Analyses 207Pb/235 U 2σ [%] 206Pb/238U 2σ [%] Rho 206Pb/238U [age] 2σ [age]

Gr-1-c* 0.33646 11.2 0.04691 3.1 0.28 301.7 9.1

Gr-4-c 0.346 4.2 0.04812 1.9 0.45 304.5 5.7

Gr-4-r 0.34676 3.6 0.04746 1.7 0.47 300.7 5

Gr-6-r 0.3515 4.1 0.04792 2 0.49 299.7 5.9

Gr-6-c 0.34164 6 0.04691 2.1 0.35 299.6 6.2

Gr-7-c 0.34346 10.5 0.04796 2.9 0.28 304.2 8.6

Gr-7-r 0.35147 12.4 0.04921 2.9 0.23 305.9 8.7

Gr-8-r 0.37297 5.9 0.04866 2 0.34 310.9 6.1

Gr-10-c* 0.35007 4 0.04826 1.6 0.4 300.6 4.7

Gr-10-r* 0.34513 3.1 0.0485 1.4 0.45 304.8 4.2

Gr-13-c* 0.34198 8.6 0.04895 2.3 0.27 310.5 7

Gr-13-r* 0.35166 13.8 0.04717 3.3 0.24 299.6 9.7

Gr-14-r* 0.34539 6.2 0.04787 1.8 0.29 301.2 5.3

Gr-14-c* 0.3438 7.9 0.04754 2 0.25 304 5.9

Gr-15-c 0.35007 8.5 0.04651 2.4 0.28 293.5 6.9

G r-15-r 0.34085 5.2 0.04724 1.6 0.31 295.9 4.6

Gr-16-c 0.37572 5.5 0.04841 1.9 0.35 303.2 5.6

Gr-17-c 0.34633 8.9 0.0475 2.4 0.27 303.3 7.1

Gr-17-r 0.34781 10.5 0.04826 2.5 0.24 301.3 7.4

Gr-21-r 0.36486 11.6 0.04686 3.4 0.29 296.4 9.9

Gr-22-c 0.352 3.4 0.0485 1 0.1 305.3 2.8

Gr-23-c 0.35365 4.1 0.04855 1.6 0.39 306.2 4.8

Gr-23-r 0.35616 3.8 0.04718 1.7 0.45 299.7 5

Gr-25-c 0.33914 8.8 0.04742 2.6 0.3 300.4 7.6

Gr-25-r 0.34732 9.2 0.04712 3.2 0.35 300.2 9.4

Gr-27-r* 0.34391 6.3 0.04832 2.3 0.37 305.8 6.9

Gr-27-c* 0.369 7.9 0.04731 2 0.1 298.2 5.8

Gr-31-c* 0.36395 13.6 0.04838 4.3 0.32 306.3 12.9

Gr-32-c 0.34686 6.2 0.04835 3.1 0.5 308.5 9.3

Gr-32-r 0.34 5.3 0.0471 2.8 0.4 296.5 7.7

244

MACHEV et al. / Turkish J Earth Sci

syntectonic. Their crystallization was accompanied by high-temperature liquid-state deformation followed by low-temperature solid-state deformation. According to the type of oscillatory zoning in the zircon (Erdman et al., 2013) and the lack of pegmatite and aplite veins, the magma most likely was volatile-undersaturated. The features of L1 granites (negative anomalies of Th, Nb, and Sm) and the absence of inherited zircons suggest a subduction-related magmatic arc origin for the granites.

The melting, magma generation, and granite emplacement have continued after the peak of metamorphism and deformation causing the formation of L2 granites. The second magma pulse had different composition; the magma was volatile-saturated in the sense of Erdman et al. (2013), resulting in crystallization of large homogeneous cores of the zircon. The L2 granites have intruded in the later stage of metamorphism and

deformation of the host rocks; they are partly and poorly deformed. The obtained data differ from the results of Okay et al. (2001) since they show older ages of the granitic magmatism in the Strandja Massif. The differences in ages obtained for the Kula granite (271 ± 2 Ma by Okay et al., 2001 vs. 293.5 ± 1.7 Ma in this study) might be due to the inhomogeneity of the pluton. It was probably formed by several different magma pulses that differ in age but are close to the age of metamorphism of the host rocks. This assumption requires further petrological and isotopic investigations to be confirmed. In this sense, the L2 granite could be rather interpreted as postmetamorphic.

The Late Jurassic metamorphism under green schist-facies conditions (Chatalov, 1988, 1990; Okay et al., 2001) is manifested by the above described alterations in the L2 granites.

Unlike in the Strandja Massif, many granitic intrusions

Table 5. Estimated ages of Fakiya (L2) granites. *: Obtained age is presented in Figure 11.

Analyses 207Pb/235U 2σ [%] 206Pb/238U 2σ [%] Rho 206Pb/238U [age] 2σ [age]

Gr-1-r 0.30926 12.1 0.04626 2.8 0.23 288.7 7.9

Gr-2-r 0.3785 10.4 0.04658 2.5 0.24 296.8 7.3

Gr-3-r* 0.34031 15.2 0.04559 2.8 0.18 290.7 8

Gr-3-c* 0.40291 14.8 0.04699 2.3 0.16 293.4 6.6

Gr-4-c 0.35361 12.1 0.04527 2.4 0.2 289.5 6.8

Gr-4-r 0.34829 9.9 0.04666 2.2 0.22 296.6 6.4

Gr-6-r 0.39881 15.8 0.04717 2.6 0.16 298.4 7.6

Gr-9-r 0.33169 19.3 0.04726 2.4 0.12 305.1 7.2

Gr-14-c 0.31585 14.9 0.04697 2.2 0.15 300.5 6.5

Gr-16-r 0.36316 9.7 0.04675 2 0.21 291.7 5.7

Gr-19-r* 0.35919 12.6 0.04795 1.9 0.15 300 5.6

Gr-19-c* 0.35025 13.9 0.04491 2 0.14 282.8 5.5

Gr-20-r* 0.31903 9.6 0.04692 1.8 0.19 296.9 5.2

Gr-20-c* 0.35488 7.2 0.04686 1.8 0.25 295.8 5.2

Gr-21-c 0.32965 17.9 0.0457 2.3 0.13 291.2 6.6

Gr-21-r 0.33423 12.3 0.04668 2.2 0.18 292.6 6.3

Gr-22-r* 0.39363 18.3 0.04683 2.2 0.12 296.8 6.4

Gr-22-c* 0.34882 11.7 0.04705 2.2 0.19 296.5 6.4

Gr-46-r* 0.34087 10.1 0.04646 2.2 0.22 292.5 6.3

Gr-44-c* 0.3677 14.9 0.04709 2.5 0.17 296.9 7.3

Gr-44-r* 0.33855 13.1 0.04572 2.6 0.2 291.8 7.4

Gr-27-r 0.33769 15 0.04599 2.3 0.15 292.4 6.6

Gr-41-r 0.36747 15.7 0.04593 2.6 0.17 290.7 7.4

Gr-40-r 0.32624 10.2 0.04619 2.4 0.24 294.2 6.9

Gr-35-r 0.34339 13.3 0.04691 2.9 0.22 293.2 8.3

Gr-37-c 0.35579 12.1 0.04801 2.6 0.21 299 7.6

245

MACHEV et al. / Turkish J Earth Sci

of this age in Bulgaria are not metamorphosed and crosscut the metamorphic fabric of the country rocks. There, the age of the regional metamorphism in Sredna Gora Zone is constrained to be 336 Ma (Carrigan et al., 2006), but in Strandja it postdated the emplacements of metagranites that are 312 Ma old (Okay et al. 2001). This difference in ages is difficult to reconcile with the Variscan collision common to the European Hercynides and thus we support the assumption of Natal’in and Şengör (2005)

and Natal’in et al. (2012) that the Strandja Massif, together with the Balkan Zone in the west and the Istanbul Zone in the east, represents a segment of the Variscan belt in Europe and shares all aspects of its Precambrian-Paleozoic history. As a part of the Pontides (Okay 2008), affected by numerous terrane amalgamations from the Variscan time until the Cretaceous-Paleocene opening of the Black Sea basin, the Strandja Massif bears evidence for Variscan and Cimmeride orogenies. The Strandja Massif

360

320

280

240

0.032

0.036

0.040

0.044

0.048

0.052

0.056

0.060

0.064

0.22 0.26 0.30 0.34 0.38 0.42 0.46207 Pb/ 235 U

206

Pb/23

8U

Concordia Age = 301.9 ±1.1 Ma(95% confidence, decay-const. errs included)

MSWD (of concordance) = 1.18,Probability (of concordance) = 0.28

data-point error ellipses are 2 σ

280

290

300

310

320

206

Pb/23

8U

, Age

[Ma]

Mean = 302.5±2.1 [0.70%] 95% conf.MSWD = 0.44, probability = 0.996

data-point error symbols are 2 σ

320

260

0.038

0.042

0.046

0.050

0.054

0.2 0.3 0.4 0.5207Pb/ 235U

206 Pb

/238 U

Concordia Age = 293.5 ±1.7 Ma(95% confidence, decay-const. errs included)

MSWD (of concordance) = 7.5,Probability (of concordance) = 0.10

data-point error ellipses are 2 σ

270

280

290

300

310

320

206

Pb/23

8U

, Age

[Ma]

Mean = 294.7±1.3 [0.45%] 95% conf.Wtd by data-pt errs only, 2 of 26 rej.

MSWD = 1.00, probability = 0.46

data-point error symbols are 2 σ

Figure 10. a- Concordia plot showing the U-Pb dating results of zircons from L1-type granites indicating an age of 301.9 ± 1.1 Ma (2σ) obtained from 30 laser ablation analyses in 18 zircon grains. b- Diagram of individual 206Pb/238U ages obtained in zircons from L1-type granites corresponding to calculated weighted mean age of 302.5 ± 2.1 Ma (2σ).

Figure 11. a- Concordia plot showing the U-Pb dating results of zircons from L2-type granite indicating an age of 293.5 ± 1.7 Ma (2σ) obtained from 26 laser ablation analyses in 19 zircon grains presented in Table 5. b- Diagram of individual 206Pb/238U ages obtained in zircons from L2-type granites corresponding to calculated weighted mean age of 294.7 ± 1.3 Ma (2σ). Two 206Pb/238U ages (dashed lines) are not included (Gr-9-r and Gr-19-c in Table 1).

246

MACHEV et al. / Turkish J Earth Sci

shows remarkable similarities with the late Paleozoic-early Mesozoic Silk Road arc evolving on the southern margin of Eurasia due to the northward subduction of the Paleo-Tethys, i.e. the Strandja Massif is a fragment of the long-lived Ordovician-Triassic magmatic arc, which evolved on the northern side of the Paleo-Tethys (Natal’in et al., 2012).

Acknowledgments This study was financially supported by the Scientific Fund of St. Kliment Ohridski University, Grants 9/2011, 155/2012, and 97/2013, and the Synthesys Program, Project DE-TAF-763. We are grateful to the reviewers who critically read the manuscript and made numerous suggestions for it improvement.

Belousova E, Griffin W, O’Reilly S, Fisher N (2002). Igneous zircon: trace element composition as an indicator of source rock type. Contrib Mineral Petr 143: 602–622.

Bolhar R, Ring U, Kemp A, Whitehouse M, Weaver S, Woodhead J, Uysal I, Turnbull R (2012). An integrated zircon geochronological and geochemical investigation into the Miocene plutonic evolution of the Cyclades, Aegean Sea, Greece: Part 2 –geochemistry. Contrib Mineral Petr 164: 915–933.

Boncheva I, Chatalov G (1998). Paleozoic conodonts from the Dervent Heights and the Strandzha Mountain - SE Bulgaria. Cr Acad Bulg Sci 51: 45–48.

Bouchez JL, Delas C, Gleizes G, Nėdėlec A Coney M (1992). Submagmatic microfractures in granites. Geology 20: 35–38.

Carrigan C, Mukasa S, Haydoutov I, Kolcheva K (2005). Age of Variscan magmatism from the Balkan sector of the orogen, central Bulgaria. Lithos 82: 125–147.

Carrigan C, Mukasa S, Haydoutov I, Kolcheva K (2006). Neoproterozoic magmatism and carboniferous high-grade metamorphism in Sredna Gora Zone, Bulgaria: an extension of the Gondwana-derived Anatolia-Cadomian belt? Gondwana Res 147: 404–416.

Chatalov G (1980). Two facies types of Triassic in Strandza Mountain, SE Bulgaria. Riv Ital Paleontol S 85: 1029–1046.

Chatalov G (1988). Recent developments in the geology of the Strandza zone in Bulgaria. B Tech Univ Istanbul 41: 433–466.

Chatalov G (1990). Geology of the Strandja Zone in Bulgaria. Sofia, Bulgaria: Publishing House of the Bulgarian Academy of Science (in Bulgarian with abstract in English).

Chatalov G (1991). Triassic in Bulgaria – a review. B Tech Univ Istanbul 44: 103–135.

Chatalov G, Dabovski C, Savov S, Philipov L, Nikolov G (1995). Geological Map of Bulgaria in Scale 1:100 000, Map Sheet “Zelyazkovo”. Sofia, Bulgaria: BAS Publishing House.

Cherniak D, Watson E (2001). Pb diffusion in zircon. Chem Geol 172: 5–24.

Dabovski C, Boyanov I, Khrischev K, Nikolov T, Sapunov I, Yanev Y, Zagorchev I (2002). Structure and Alpine evolution of Bulgaria. Geol Balc 32: 9–15.

Dabovski C, Savov S (1988). Structural studies in the nappes of Southeast Strandza. Geol Balc 18: 19–36.

Dabovski C, Savov S, Chatalov G, Shilyafov G (1993). Geological Map of Bulgaria in Scale 1:100 000. Map Sheet “Edirne”. Sofia, Bulgaria: BAS Publishing House.

Dimitrov S (1946). The metamorphic and magmatic rocks of Bulgaria. Ann Comm Geol Inv 4: 61–93 (in Bulgarian).

Erdman S, Wodicka N, Jackson SE, Corrigan D (2013). Zircon textures and composition: refractory recorders of magmatic volatile evolution. Contrib Mineral Petr 165: 45–71.

Ewing T, Hermann J, Rubatto D (2013). The robustness of the Zr-in-rutile and Ti-in-zircon thermometers during high-temperature metamorphism (Ivrea-Verbano Zone, Northern Italy). Contrib Mineral Petr 165: 757–779.

Ferry J, Watson E (2007). New thermodynamic models and revised calibrations for the Ti-in-zircon and Zr-in-rutile thermometers. Contrib Mineral Petr 154: 429–437.

Fu B, Page FZ, Cavosie AJ, Fournelle J, Kita NT, Lackey JS, Wilde SA, Valley JW (2008). Ti-in-zircon thermometry: applications and limitations. Contrib Mineral Petr 156: 197–215.

Gerdjikov I (2005). Alpine metamorphism and granitoid magmatism in the Strandja Zone: new data from the Sakar Unit, SE Bulgaria. Turk J Earth Sci 14: 167–183.

Gochev P (1985). Strandzhides. Geotect Tectonophys Geodynam 18: 28–52 (in Bulgarian with abstract in English).

Gochev P (1991). Some problems of the nappe tectonics of the Strandzhides in Bulgaria and Turkey. B Tech Univ Istanbul 44: 137–144.

Haskin IA, Frey FA, Wildeman TR (1968). Relative and absolute terrestrial abundances of the rare earth. In: Ahrens LH, editor. Origin and Distribution of the Elements. Oxford, UK: Pergamon, pp. 899–912.

Hoskin PWO, Ireland TR (2000). Rare earth element chemistry of zircon and its use as a provenance indicator. Geology 28: 627–630.

Hoskin PWO, Schaltegger U (2003). The composition of zircon and igneous and metamorphic petrogenesis. Rev Mineral Geochem 53: 27–62.

Ivanov Z (1998). Tectonic of Bulgaria. Habilitation thesis, Sofia University, Sofia, Bulgaria (in Bulgarian).

References

247

MACHEV et al. / Turkish J Earth Sci

Ivanov Z, Gerdjikov I, Kounov A (2001). New data and considerations about structure and tectonic evolution of Sakar region, SE Bulgaria. Ann Univ Sofia Geology 91: 35–80 (in Bulgarian with abstract in English).

Jackson SE, Pearson NJ, Griffin WL, Belousova EA (2004). The application of laser ablation-inductively coupled plasma-mass spectrometry to in situ U–Pb zircon geochronology. Chem Geol 211: 47–69.

Jaranoff D (1960). La tectonique de la Bulgarie. Sofia, Bulgaria: Techica Sofia (in Bulgarian with abstract in French).

Jochum KP, Weis U, Stoll B, Kuzmin D, Yang Q, Raczek I, Jacob DE, Stracke A, Birbaum K, Frick DA et al. (2011). Determination of reference values for NIST SRM 610-617 glasses following ISO guidelines. Geostand Geoanal Res 35: 397–429.

Kamenov B (1976). Petrology of the Zhelyazkovo pluton. Ann Univ Sofia Geology 67: 180–212 (in Bulgarian with abstract in English).

Kamenov B, von Quadt A, Peytcheva I (2002). New insight into petrology, geochemistry and dating of the Vejen Pluton, Bulgaria. Geochem Mineral Petr 39: 3–26.

Kemp A, Whitehouse M, Hawkesworth C, Alarcon M (2005). A zircon U-Pb study of metaluminous (I-type) granites of the Lachlan Fold Belt, southeastern Australia: implications for the high/low temperature classification and magma differentiation processes. Contrib Mineral Petr 150: 230–249.

Kruhl J (1996). Prism- and basal-plane parallel subgrain boundaries in quartz: a microstructural geothermobarometer. J Metamorph Geol 14: 581–589.

Lilov P, Maliakov Y (2001). Donnees de geochronology isotopique sur des metadiabases du Strandja. Cr Acad Bulg Sci 54: 67–70 (in French).

Lilov P, Maliakov Y, Balogh K (2004). K-Ar dating of metamorphic rocks from Strandja Massif, SE Bulgaria. Geochem Mineral Petr 41: 107–119.

Ludwig KR (2003). Isoplot/Ex Version 3.00: A Geochronological Toolkit for Microsoft Excel. Berkeley, CA, USA: Berkeley Geochronology Center.

Lyubetskaya T, Korenaga J (2007). Chemical composition of Earth’s primitive mantle and its variance: 1. Method and results. J Geophys Res 112: BO3211.

Natal’in BA, Şengör A (2005). Late Palaeozoic to Triassic evolution of the Turan and Scythian platforms: the pre-history of the Palaeo-Tethyan closure: Tectonophysics 404: 175–202.

Natal’in BA, Sunal G, Satir M, Toraman E (2012). Tectonics of the Strandja massif, NW Turkey: History of a long-lived arc at the Northern margin of Palaeo-Tethys. Turkish J Earth Sci 21: 755–798.

Nemchin AA, Giannini LM, Bodorkos S, Oliver NHS (2001). Ostwald ripening as a possible mechanism for zircon overgrowth formation during anatexis: theoretical constraints, a numerical model, and its application to pelitic migmatites of the Tickalara metamorphics, northwestern Australia. Geochim Cosmochim Ac 65: 2771–2788.

Neubauer F, Bilyarski S, Genser J, Ivanov Z, Peycheva I, von Quadt A (2010). Jurassic and cretaceous tectonic evolution of the Sakar and Srednogorie zones, Bulgaria: 40Ar/39Ar mineral ages and structures. Geol Balc 39:273–274.

Okay AI (2008). Geology of Turkey: a synopsis. Anschnitt 21: 19–42.

Okay AI, Satir M, Tüysüz O, Akyüz S, Chen F (2001). The tectonics of the Strandja Massif: late-Variscan and mid-Mesozoic deformation and metamorphism in the northern Aegean. Int J Earth Sci 90: 217–233.

Paton C, Hellstrom J, Paul B, Woodhead J, Hergt J (2011). Iolite: Freeware for the visualisation and processing of mass spectrometric data. J Anal Spectrom 26: 2508–2518.

Peytcheva I (1997). The Alpine metamorphism in Eastrn Rhodopes – Rb-Sr data. Rev Bulg Geol Soc 58: 157–165.

Peytcheva I, Bibikova E, Makarov V (1992). U-Pb dating of zircons from two type gneisses (SE Rhodopes, Bulgaria). Cr Acad Bulg Sci 45: 71–74.

Peytcheva I, Ovtcharova M, Sarov S, Kostitsin Y (1998). Age and metamorphic evolution of metagranitoids from Kesebir reka region, Eastern Rhodopes – Rb-Sr isotope data. In: XVI Congress CBGA Proceedings, p. 471.

Peytcheva I, Sarov S, Kostitsin Y (1995). Rb-Sr isotope-geochronological study of metagranites from the region of Belopoltsy-Zhelezino-Pelevun in the Eastern Rhodopes. Cr Acad Bulg Sci 48: 11–12.

Peytcheva I, von Quadt A, Ovtcharova M, Handler R, Neubauer F, Salnikova E, Kostitsyn Y, Sarov S, Kolcheva K (2004). Metagranitoids from the eastern part of the Central Rhodopean Dome (Bulgaria): U–Pb, Rb–Sr and 40Ar/39Ar timing of emplacement and exhumation and isotope-geochemical features. Mineral Petr 82: 1–31.

Savov S (1979). Lower Cambrian trace fossil from high-grade metamorphic schists, Strandža Mountain, Southeast Bulgaria. Paleont Stratigr Lithol 8: 19–22.

Şengör A (1984). The Cimmeride Orogenic System and the Tectonics of Eurasia. GSA Special Papers 195: 1–74.

Şengör A, Yılmaz Y, Sungurlu O (1984). Tectonics of the Mediterranean Cimmerides: nature and evolution of the western termination of Palaeo-Tethys. In: Dixon, JE Robertson, AHF, editors. Geological Evolution of the Eastern Mediterranean. London, UK: Geological Society of London Special Publications, pp. 77–112.

Sergeeva L, Nachev I, Maliakov I (1979). On the Palaeozoic age of the metamorphic rocks in Strandja. Rev Bulg Geol Soc 40: 10–17 (in Bulgarian).

248

MACHEV et al. / Turkish J Earth Sci

Sunal G, Natal’in BA, Satir M, Toraman E (2006). Paleozoic magmatic events in the Strandja massif, NW Turkey. Geodin Acta 19: 281–298.

Sunal G, Satir M, Natal’in BA, Topuz G, Vonderschmidt O (2011). Metamorphism and diachronous cooling in a contractional orogen: the Strandja Massif, NW Turkey. Geol Mag 148: 580–596.

Sunal G, Satir M, Natal’in BA, Toraman E (2008). Paleotectonic position of the Strandja massif and surrounding continental blocks based on zircon Pb-Pb age studies. Int Geol Rev 50: 519–545.

Ustaömer T, Robertson AHF (1993). Late Palaeozoic–Early Mesozoic marginal basins along the active southern continental margin of Eurasia: evidence from the Central Pontides (Turkey) and adjacent regions. Geol J 28: 219–238.

Wang Q, Zhu D, Zhao Z, Guan Q, Zhang X, Sui Q, Hu Z, Moa X (2012). Magmatic zircons from I-, S- and A-type granitoids in Tibet: trace element characteristics and their application to detrital zircon provenance study. J Asian Earth Sci 53: 59–66.

Watson E, Harrison T (1983). Zircon saturation revisited: temperature and composition effects in a variety of crustal magma types. Earth Planet Sc Lett 64: 295–304.

Watson E, Hayden L, Wark D, Cherniak D, Thomas J, Manchester J (2006a). New crystallization thermometers for zircon, rutile and sphene: calibration, diffusion considerations, and applications. In: 41st Annual Meeting of the Geological Society of America, NE Section, Paper 2-7.

Watson E, Wark D, Thomas J (2006b) Crystallization thermometers for zircon and rutile. Contrib Mineral Petr 151: 413–433.

Yanev S (2000). Palaeozoic terranes of the Balkan Peninsula in the framework of Pangea assembly. Palaeogeogr Palaeocl 161: 151–177.

Yanev S, Göncüoğlu C, Gedik İ, Lakova I, Boncheva I, Sachanski V, Okuyucu C, Özgül N, Timur E, Maliakov Y et al. (2006). Stratigraphy, correlations and palaeogeography of Palaeozoic terranes of Bulgaria and NW Turkey: a review of recent data. Geol Soc London Spec Publ 260: 51–67.

Yilmaz Şahin S, Aysal N, Güngör Y (2012). Petrology, geochemistry and geochronology of the Late Variscan granitoids (Late Permian-Early Triassic) in Western Pontides, Turkey. In: Proceedings of the 34th International Geological Congress, Abstract Book.

Yilmaz Şahin S, Aysal N, Güngör Y, Peycheva I, Heubauer F (2014). Geochemistry and U–Pb zircon geochronology of metagranites in Istranca (Strandja) Zone, NW Pontides, Turkey: implications for the geodynamic evolution of Cadomian orogeny. Gondwana Res 26: 755–771.

Zagorchev I, Dabovski H, Chunev D (1973). On the tectonics of the western part of the Sredna Gora crystalline block (Sredna Gora proper). Rev Bulg Geol Soc 34: 1–10.