3

GEOLOGICA BALCANICA, 40. 1–3, Sofia, Dec. 2011, p. 3–12.

Potassic syenite from Shipka, Central Balkan Mts, Bulgaria: characterization and insight into the source

Momchil DyulgerovSofia University “St Kliment Ohridski”, Faculty of Geology and Geography, 15 Tzar Osvoboditel, 1504 Sofia, Bulgaria;e-mail: momchil@gea.uni-sofia.bg(Accepted in revised form: 10.03.2011)

Abstract. Petrographic, mineralogical and whole-rock chemistry data from a new outcrop of peralkaline po-tassic syenite of presumed Variscan age near the town of Shipka, are presented here. The syenites exhibit peralkaline chemistry and very potassic character which markedly differs from nearby situated potassic mon-zonites, known as Shipka pluton. In spite of the close spatial association of these two bodies, genetic link between them can hardly be supposed. The syenites are composed of K-feldspar, diopside, mica and late interstitial amphibole. Mineral composition reflects the agpaitic conditions of crystallization with formation of sodic amphibole and K-feldspars with important Sr and Ba contents. These rocks have peralkaline whole-rock chemistry, very high potassic content and extreme enrichment in LILE, LREE, Th and U. Their trace element signature and isotope characteristics are in favour of derivation from metasomaticaly enriched mantle source.

Dyulgerov, M. 2011. Potassic syenite from Shipka, Central Balkan Mts, Bulgaria: characterization and insight into the source. Geologica Balcanica 40(1–3), 3–12.

Key words: potassic syenites, sodic amphibole, enriched source, Central Balkan Mountains.

IntroductIon

Variscan potassic syenites and monzonites occur rarely in Western and Central Balkan Mts, Bulgaria (Fig. 1), as they always outcrop as small, isolated bodies. The com-position of these rocks is essentially intermediate, paren-tal mafic rocks usually lack, and the most differentiated lithologies have acid composition (quartzsyenite and granite). They intrude low-grade metamorphic rocks, at-testing for hypabyssal level of emplacement (Dimitrov, 1935, 1937; Kostov, 1950; Dyulgerov, 2005). Spatial relationships between the potassic rocks and the main Variscan calc-alkaline granitoids have not been ob-served, the distances between these two rock types be-ing 20 to 50 km. The existing age data suggest that sy-enites and granites are contemporaneous, formed within the time interval 330–305 Ma (Kamenov et al., 2002; Carrigan et al., 2005; Peycheva et al., 2006; Dyulgerov et al., 2010).

The subject of this study is a new outcrop of po-tassic syenite in a recently opened quarry (N 42°40'41"; E 25°17'06") near Golyama Varovita River, situated 3–4 km west of the town of Shipka. Isotopic data were recently reported in Dyulgerov et al. (2007). This paper presents a comprehensive characterization (mineralogi-

cal, petrographic and geochemical) of the studied rocks. The visible part of the outcrop has small dimensions – 30×50 m, surface exposures not being found so far. It has plug- or dyke-like form and most probably repre-sents an upper part of more deeply situated body.

This syenite belongs to Sliven-Shipka Unit of the Srednogorie Zone. It is intruded into greenschist-facies metapelites dated as Devonian by Kalvacheva and Prokop (1988). The contacts are discordant and sharp, with thermal aureole and recrystallization in the country rocks. The intrusion of the body within the green-schist metamorphic rocks limits its emplacement depth up to 2–2.5 kbar (7–8 km).

PetrograPhy and MIneralogy

The rocks are fresh, greyish and pinkish, with massive structure. The texture is fine to medium grained and porphyroid on clinopyroxene. Clinopyroxene (7%) and mica (20%) are the earliest crystallized phases, always euhedral and slightly larger than other minerals. The potassium feldspar (65%) forms platy subhedral crystals; the sodic amphibole (6%) is interstitial, anhedral, being the last crystallized rock-forming mineral. The most

4

Fig. 1. Position of potassic-alkaline magmatism in the Variscan edifice in Bulgaria

SiO2 53.17 52.67 54.33 53.35 53.71 53.84 53.66 53.52 53.65TiO2 0.13 0.17 0.16 0.23 0.17 0.20 0.12 0.15 0.13Al2O3 2.38 2.47 2.37 3.18 2.11 0.64 0.48 0.55 0.62Cr2O3 0.00 0.03 0.04 0.34FeO 3.72 4.03 4.25 10.02 5.34 5.21 4.78 4.78 4.28MnO 0.15 0.16 0.14 0.21 0.20 0.12 0.13 0.16 0.11MgO 17.98 16.48 17.35 12.31 16.27 14.86 15.65 15.51 15.39CaO 21.68 23.25 22.39 20.45 22.16 23.82 23.87 23.73 24.74Na2O 0.00 0.00 0.00 1.12 0.00 0.41 0.49 0.66 0.43K2O 0.05 0.00 0.00 0.00 0.05 0.01 0.00 0.00 0.01total 99.26 99.23 100.99 100.87 100.01 99.11 99.21 99.09 99.71Si 1.944 1.938 1.962 1.971 1.971 2.000 1.982 1.977 1.974IVAl 0.056 0.062 0.038 0.029 0.029 0.000 0.018 0.023 0.026t 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000Ti 0.004 0.005 0.004 0.006 0.005 0.005 0.003 0.004 0.004VIAl 0.047 0.046 0.063 0.110 0.062 0.028 0.002 0.001 0.001Cr 0.000 0.000 0.000 0.000 0.000 0.000 0.001 0.001 0.010Fe3+ 0.005 0.007 0.000 0.000 0.000 0.000 0.044 0.060 0.038Fe2+ 0.109 0.117 0.128 0.310 0.164 0.162 0.104 0.088 0.093Mn 0.005 0.005 0.004 0.007 0.006 0.004 0.004 0.005 0.004Mg 0.980 0.904 0.934 0.678 0.890 0.823 0.862 0.854 0.844Ca 0.849 0.917 0.866 0.810 0.871 0.948 0.945 0.939 0.975Na 0.000 0.000 0.000 0.080 0.000 0.029 0.035 0.047 0.030K 0.002 0.000 0.000 0.000 0.002 0.000 0.000 0.000 0.000M1+M2 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000Wo 0.400 0.433 0.416 0.394 0.423 0.474 0.470 0.467 0.481En 0.490 0.452 0.467 0.339 0.445 0.411 0.431 0.427 0.422Fe 0.002 0.002 0.002 0.003 0.003 0.002 0.002 0.002 0.002

Note: blank boxes – not analysed.

Table 1Pyroxene analyses. Structural formula based on 4 cations, Fe2+ and Fe3+ according to the charge balance

5

abundant accessory phases are apatite, zircon and titanite; less common are iron oxide and magmatic carbonate. Among the accessories apatite and carbonate are the most interesting. Apatite occurs as prismatic to acicular, euhedral crystals. Carbonate is the last crystallizing mineral; it is subhedral to anhedral and shows lamellar structure. According to the modal composition and mineral proportions the rock is classified as syenite.

Mineral composition is determined by electron mi-croprobe JEOL–733 Superprobe–EDS at 25 kV accel-erating voltage at EUROTEST-CONTROL Plc, Sofia, and electron microprobe JEOL JXA 8500F, with 5 WDS detectors and 20 kV acceleration voltage at Humboldt University, Berlin.

The clinopyroxene belongs to the group of Ca-pyro-xenes. It shows distinct compositional zoning with XMg varying between 0.91 and 0.69 (Table 1). This significant magnesian character should be attributed to the tempera-ture of crystallization: pyroxene, together with mica, are among the first crystallized phases and hence acquired most elevated Mg/Fe ratio. The composition of pyro-xene is diopside to augite and Wo–En–Fs end-members dominate. Al2O3 is between 0.12 and 3.2 wt.% (0.02 and 0.14 apfu – atoms per formula unit) and small amount of tschermakite type (Al-based) end-members also present. In spite of the peralkaline character of the rock, Na con-tent is low. This fact is in favour of multistage crystal-lization of the rock and supports the observation that the

clinopyroxene was formed before the transition toward peralkaline conditions.

Mica forms platy, hexagonal, always euhedral crys-tals without inclusions. It shows distinct zoning with pale cores and thin band of dark brown rims. Mica composi-tion is magnesian with limited range in XMg: 0.88–0.75 decreasing from the centre to the periphery (Table 2). Ti content is moderate (TiO2 = 0.91–2.89 wt.%) and cor-relates negatively with Mg in the micas. Most magnesian species have highest Cr content – up to 0.44 wt.% Cr2O3. Al2O3 is relatively low (12.67–14.11 wt.%) and 2.16 to 2.58 apfu, reflecting the low Al2O3 content of the rock.

Potassium feldspar is subhedral, fresh, not affected by alteration processes. Its composition is homogenous with dominating orthoclase end-member. Na content is negligible – up to 0.65 wt.% Na2O, forming 6 mol.% al-bite (Table 3). Ba and Sr contents are important and may reach up to 4.41 wt.% BaO and up to 2.24 wt.% SrO. They form the celsian component up to 13.87 mol.%. High Ba and Sr values (when analyzed) correlate very well with Al and Fe content in the feldspar suggesting charge balance role of these elements when they substi-tute Si. There is no clear tendency for Ba and Sr partition in the crystals of K-feldspar from core to rim and it can be suggested that these two elements are randomly dis-tributed. This fact is probably related to the emplacement at hypabyssal level. The magma cooled fast and during its crystallization the accommodation of cations – K, Na,

SiO2 41.13 39.56 41.91 39.59 39.54 40.35 38.63 40.34 39.76 39.91 38.91 39.15 39.37TiO2 1.27 2.89 0.91 1.57 2.95 2.30 2.41 1.84 1.90 2.21 2.37 2.43 2.43Al2O3 13.22 12.67 12.88 12.70 13.98 14.11 15.04 13.28 13.71 13.81 13.54 13.82 13.41FeO 6.85 11.10 5.52 7.98 9.94 9.22 8.51 8.37 9.93 10.11 9.66 9.67 9.34MnO 0.00 0.16 0.19 0.15 0.12 0.15 0.10 0.11 0.10 0.10 0.09 0.13 0.08MgO 23.26 19.74 23.64 24.05 17.08 17.80 19.21 19.85 18.69 17.50 17.50 17.90 18.51CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.01 0.01Na2O 0.58 0.00 0.00 0.00 0.07 0.06 0.12 0.08 0.11 0.04 0.11 0.30 0.16K2O 9.55 9.75 9.92 9.40 10.86 10.05 11.47 11.41 11.37 11.16 10.28 11.23 10.35Cr203 0.09 0.14 0.16 0.20 0.14 0.00 0.08 0.44 0.18total 95.86 95.87 94.97 95.44 94.64 94.18 95.64 95.47 95.70 94.86 92.54 95.07 93.82Si 5.853 5.765 5.975 5.706 5.828 5.912 5.630 5.869 5.815 5.876 5.849 5.769 5.830Ti 0.136 0.317 0.098 0.170 0.327 0.254 0.264 0.202 0.209 0.245 0.268 0.269 0.271Al 2.217 2.176 2.164 2.157 2.429 2.436 2.584 2.277 2.364 2.397 2.398 2.400 2.340Fe2+ 0.815 1.353 0.658 0.962 1.225 1.130 1.037 1.018 1.215 1.244 1.214 1.192 1.157Mn 0.000 0.020 0.023 0.018 0.015 0.019 0.012 0.013 0.012 0.013 0.012 0.016 0.010Mg 4.933 4.287 5.023 5.166 3.751 3.888 4.173 4.304 4.074 3.841 3.921 3.931 4.084Ba 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Ca 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.004 0.000 0.001 0.001Na 0.160 0.000 0.000 0.000 0.019 0.017 0.033 0.023 0.031 0.011 0.031 0.086 0.046K 1.734 1.813 1.804 1.728 2.043 1.879 2.134 2.117 2.122 2.097 1.972 2.111 1.955Cr 0.000 0.000 0.000 0.000 0.011 0.016 0.018 0.023 0.016 0.000 0.009 0.052 0.021cations 15.852 15.740 15.750 15.912 15.663 15.563 15.893 15.854 15.868 15.741 15.687 15.840 15.725XMg 0.86 0.76 0.88 0.84 0.75 0.77 0.80 0.81 0.77 0.76 0.76 0.77 0.78

Note: blank boxes – not analysed.

Table 2Mica analyses. Structural formula based on 22 oxygens

6

Ba, Sr was to some extent fortuitous. It should be noted that the magma is very rich in LIL elements and this fea-ture contributes to the crystallization of such Ba and Sr rich feldspar.

Amphiboles are late and restricted to the groundmass. They form tiny, prismatic or needle-like crystals, with pleochroism in green to violet colours. The composition of amphiboles is sodic; the presented species are arf-vedsonite, magnesioarfvedsonite and rarely eckermman-ite (Table 4). They have low Al and Ca contents; XMg var-ies from 0.61 to 0.32. Amphiboles are slightly enriched in Ti – up to 3.15 wt.% TiO2 (0.365 apfu), suggesting certain substitution of OH– groups by O (Hawthorne et al., 2000). [A] – site is filled by K and Na in variable proportions, with weak dominance of Na. The [A] – site occupancy is significant – from 0.479 to 1.013 admitting elevated temperature of crystallization of these amphi-boles (Ernst, 1968).

Several stages of crystallization can be distinguished in the rock. The clinopyroxene and mica were formed first, followed by potassium feldspar. Amphibole and carbonate are restricted to the groundmass and mark the final stage, before complete solidification. The presence of calcic pyroxene and mica as early phases shows that there was a significant time gap between the formation of porphyries (pyroxene and mica) and late, interstitial amphibole, because these two minerals – cal-

cic pyroxene and mica (mineral rich in Al) are not in equilibrium with sodic amphibole. Sodic amphibole and carbonate testify to a change of crystallization condi-tions. The generation of mafic alkaline phase indicates peralkaline character of the residual magma. The pres-ence of carbonate evidences increasing activity of CO2 during the final stages of crystallization. Clinopyroxene is strongly altered due to circulation of residual fluids in the magma/rock.

geocheMIstry

Chemical composition of the rock was determined by one wet chemical silicate analysis for the major oxides at the Department of Mineralogy, Geochemistry and Petrology, Sofia University. Trace elements of the sample were ana-lysed by LA-ICP MS at ETH, Zurich.

The studied syenite belongs to the group of basic rocks (Table 5, Fig. 2). MgO and FeO are 5.73 and 3.22 wt.% respectively, as the rock shows important magnesian number: XMg is 0.76. It is characterized by low Al and very high alkaline content: ∑K2O +Na2O=13.88 wt.%. This fact predetermines the strong peralkalinity of the rock, ag-paicity index being 1.38 (A.I. – molecular (K2O +Na2O)/Al2O3). The potassic character is pronounced, K2O reaches 9.61 wt.% and dominates over Na2O (4.27 wt.%).

SiO2 62.19 64.73 64.20 60.32 61.79 62.31 63.14 62.61 63.14 62.88 63.51 62.75Al2O3 18.08 17.40 17.38 18.05 17.77 17.29 20.43 19.69 19.47 19.97 19.52 19.62FeO 0.36 0.19 0.26 0.61 0.36 0.33 0.06 0.11 0.14 0.28 0.33 0.59CaO 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.19 0.05 0.01 0.00Na2O 0.65 0.52 0.93 0.64 0.00 0.29 0.25 0.37 0.39 0.25 0.20 0.48K2O 13.56 15.61 15.06 13.08 14.61 14.70 15.63 15.70 15.70 15.99 15.85 15.62SrO 1.22 1.30 1.96 1.83 2.24 1.88BaO 4.41 0.57 0.25 4.66 3.01 2.45total 100.5 100.3 100.0 99.2 99.8 99.3 99.5 98.5 99.0 99.4 99.4 99.1Si 2.952 3.012 3.000 2.922 2.956 2.979 2.924 2.937 2.946 2.926 2.950 2.930Al 1.012 0.954 0.957 1.030 1.002 0.974 1.115 1.089 1.071 1.095 1.068 1.080Fe3+ 0.014 0.007 0.010 0.025 0.014 0.013 0.002 0.004 0.005 0.011 0.013 0.023Ca 0.000 0.000 0.000 0.000 0.000 0.000 0.002 0.000 0.009 0.002 0.000 0.000Na 0.060 0.047 0.084 0.060 0.000 0.027 0.022 0.034 0.035 0.022 0.018 0.043K 0.821 0.927 0.898 0.808 0.892 0.897 0.923 0.940 0.935 0.949 0.939 0.930Sr 0.034 0.035 0.053 0.051 0.062 0.052 0.000 0.000 0.000 0.000 0.000 0.000Ba 0.082 0.010 0.005 0.088 0.056 0.046 0.000 0.000 0.000 0.000 0.000 0.000total 4.975 4.993 5.007 4.985 4.982 4.989 4.989 5.003 5.001 5.006 4.988 5.006Or 82.39 90.93 86.35 80.17 88.26 87.77 97.47 96.51 95.46 97.50 98.13 95.56Ab 6.00 4.61 8.11 5.96 0.00 2.63 2.34 3.49 3.58 2.27 1.84 4.44An 0.00 0.00 0.00 0.00 0.00 0.00 0.19 0.00 0.97 0.23 0.03 0.00Cn 11.60 4.46 5.55 13.87 11.74 9.60 0.00 0.00 0.00 0.00 0.00 0.00total 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00

Note: blank boxes – not analysed.

Table 3Feldspar analyses. Structural formula based on 8 oxygens

7

SiO2 50.10 51.12 51.17 48.70 50.31 49.96 50.78 51.49TiO2 2.78 2.33 2.37 2.81 3.15 3.07 3.05 2.68Al2O3 2.69 2.02 2.39 1.23 1.84 2.03 1.55 1.74FeOt 23.37 19.41 21.17 24.67 21.14 17.20 19.00 18.89MnO 0.49 0.69 0.49 0.53 0.54 0.58 0.53 0.52MgO 7.91 9.70 8.30 5.50 7.50 9.48 8.86 8.16Cr2O3 0.14 0.01 0.02 0.05 0.10CaO 1.36 1.46 1.29 1.71 1.35 1.17 1.54 1.29Na2O 6.82 6.67 7.39 7.09 8.05 8.26 7.89 7.45K2O 2.28 1.57 1.79 1.76 1.76 1.67 1.77 1.65total 97.80 94.97 96.36 94.13 95.63 93.44 95.03 93.96Si 7.477 7.680 7.712 7.754 7.771 7.762 7.813 7.967IVAl 0.473 0.320 0.288 0.231 0.229 0.238 0.187 0.033Ti 0.050 0.000 0.000 0.015 0.000 0.000 0.000 0.000t 8.000 8.000 8.000 8.000 8.000 8.000 8.000 8.000VIAl 0.000 0.038 0.136 0.000 0.106 0.134 0.093 0.283Ti 0.262 0.263 0.269 0.322 0.365 0.358 0.353 0.312Fe3+ 1.107 1.041 0.694 0.461 0.186 0.180 0.176 0.138Fe2+ 1.810 1.398 1.974 2.825 2.545 2.054 2.269 2.306Mn 0.062 0.088 0.063 0.071 0.070 0.077 0.069 0.068Mg 1.759 2.172 1.864 1.305 1.726 2.194 2.032 1.881Cr 0.000 0.000 0.000 0.017 0.001 0.003 0.006 0.012c 5.000 5.000 5.000 5.000 5.000 5.000 5.000 5.000Mg 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000Ca 0.217 0.235 0.208 0.291 0.224 0.195 0.254 0.214Na 1.783 1.765 1.792 1.709 1.776 1.805 1.746 1.786B 2.000 2.000 2.000 2.000 2.000 2.000 2.000 2.000K 0.434 0.301 0.344 0.357 0.346 0.331 0.347 0.325NaA 0.191 0.178 0.368 0.480 0.636 0.683 0.609 0.450[a] 0.625 0.479 0.712 0.837 0.982 1.013 0.957 0.775XMg 0.49 0.61 0.49 0.32 0.40 0.52 0.47 0.45

Note: blank boxes – not analysed.

Table 4Amphibole analyses. Structural formula based on 23 oxygens

wt. % ppm ppm ppmSiO2 51.76 La 96.33 V 129 Cs 8TiO2 1.20 Ce 205.33 Cr 246 Rb 397Al2O3 12.67 Pr 27.60 Co 23 Ba 11000Fe2O3 1.95 Nd 120.67 Ni 116 Sr 1793FeO 3.22 Sm 20.63 Sc 18 Th 137MnO 0.10 Eu 3.91 Cu 89 U 22MgO 5.73 Gd 12.27 Zn 381 Ga 27CaO 5.48 Tb 1.18 Zr 692Na2O 4.27 Dy 5.20 Hf 19K2O 9.61 Ho 0.82 Nb 36P2O5 0.34 Er 2.08 Ta 1H2O- 0.46 Tm 0.26 Pb 111LOI 2.97 Yb 1.49 Y 23Total 99.76 Lu 0.18 Mo 2A.I. 1.38 ΣREE 498 W 1

Table 5Whole-rock analysis

8

Trace elements show variable contents and several groups can be distinguished (Table 5, Figs 3 and 4). Specific feature of the rock is the high incompatible ele-ments enrichment which is well marked on the spiderdia-gram (Fig. 3).

Transitional metals demonstrate typical values for ba-sic rocks: V (129 ppm), Co (23 ppm) and Sc (18 ppm); Ni (116 ppm) and Cr (246 ppm) are most remarkable and are indicative for the primitive character of the liquid.

LIL elements present extreme enrichment; among them most outstanding are Ba – 11000 ppm and Sr – 1793 ppm. HFS elements are enriched in different de-

gree in the sample: very elevated are Th (132 ppm), U (22 ppm), Nb (36 ppm) and Zr (692 ppm), whereas Mo (2 ppm) and W (1 ppm) have moderate content. The rock is characterized by high ΣREE – 498 ppm. LaN/LuN ratio is 50, predetermining strongly fractionated trend of distribution (Fig. 4). The negative europium anomaly is slight (Eu/Eu* is 0.75), which suggests pyroxene and/or feldspar fractionation at deeper level, prior to final in-sertion of the magma at hypabyssal level.

On the spiderdiagram the syenite shows enrichment levels of left side incompatible elements exceeding the less incompatible right side ones (Fig. 3). The peralkaline

Fig. 2. Studied syenite and the rocks from other Variscan potassic-alkaline plutons (Shipka, Svidnya and Buhovo-Seslavtsi) plotted on the whole-rock TAS diagram (Le Bas et al., 1986)

Fig. 3. Spidergram for the studied rock. Normalization on Primitive Mantle (Sun and McDonough, 1989)

9

character of the magma is decisive for the HFS elements: Th, U, Nb, and Zr.

Preliminary isotopic results (Dyulgerov et al., 2007) show that syenite has enriched isotopic signature with 87/86Sri of 0.7077 and 143/144Ndi of 0.51194 (320 Ma age corrected). The rock plots in the enriched IVth quadrant of the isotopic systematic, approaching the field of Enriched Mantle II (Fig. 5). The studied syenite shows isotopic characteristics similar to other potassic and ultrapotassic rocks: lamproites from Wyoming or potassic minettes from the Variscan Bohemian Massif.

geodynaMIc settIng and source

The syenite has MgO, XMg, FeO, Ni, Cr, Sc, V and C contents which are unambiguous evidence for its man-tle origin. These values are not high enough to accept

direct mantle origin without further differentiation and the magma was probably not in equilibrium with man-tle peridotite (Wilson, 1989). It can be accepted that the liquid was slightly evolved, as Ni and Cr contents above 100 ppm and 200 ppm, respectively, 5.73 wt.% MgO and 0.76 XMg, preclude extensive fractionation. For the scope of source considerations we admit that the magma is rep-resentative of nearly primary liquid; it has experienced limited evolution and its chemical composition is repre-sentative for the source.

Whole-rock chemistry is in favour of orogenic char-acter of magmatism. Negative Nb and Ti anomalies on the primitive mantle-normalized diagram are typical for rocks formed in subduction or collision related tectonic settings. The striking geochemistry of the magmas – important transitional metals (plus MgO) and richness in incompatible elements raises the question about the magma source. The extreme incompatible elements en-

Fig. 4. REE distribution for the studied rock. Normaliza -tion on chondrite C1 (Sun and McDonough, 1989)

Fig. 5. Sri–Ndi plot for the studied syenite com-pared to other potassic rocks: Bohemian rocks (Janoušek et al., 1995), Spanish lamproites (Nelson et al., 1986), Leucite hills lamproites (Vollmer et al., 1984)

10

richment such as Cs, Rb, Ba, Sr, Th, U and LREE evi-dences participation of the crust in the petrogenesis. A direct derivation of such magmas from a crustal source can not, however, be supposed, since the observed val-ues of incompatible elements exceed those established in all crustal reservoirs (Taylor and McLennon, 1988; Rudnick and Gao, 2003). For elements such as Th, U and Ba this exceeding is 7–8 times, for REE, Zr and Nb is 3–4 times. Possible assimilation of crustal materi-als will therefore ‘dilute’/decrease the contents of such elements in the magma. The basic chemical composi-tion – SiO2 ~ 52 wt.%, high content of MgO and tran-sitional metals, precludes direct crustal origin, as there is no process which can generate basic to intermediate alkaline magmas from crustal precursor and in crustal setting (Rapp and Watson, 1995; Sisson et al., 2005). A viable explanation of such features is the enriched and metasomatically modified mantle source. It is supposed that involvement of crustal components in the source region was mainly via metasomatizing fluids added to mantle peridotites. Derivation of similar fluids is attrib-uted to slab dehydration during the previous episodes of subduction or collision events. Evidence of such a process in the source is the extreme enrichment of ele-ments which are easily transported by fluids (Fig. 6). The comparison of the enrichment level of two incom-patible elements – one hygromagmatofile (Th) and an-other conservative (Ta), shows that the mobile element (Th) has much greater concentration than the immobile incompatible element (Ta). The distribution of elements on the spiderdiagram supports this observation as LILE and LREE present higher enrichment levels comparing to all incompatible elements (Figs 3 and 4).

The proposed mechanism and source explains also the isotopic signature of the rock. Involvement of materials with high Rb/Sr and low Sm/Nd ratios can be assigned to

the introduced crustal materials into the source. The more incompatible elements in both couples – Rb and Nd are preferably enriched in the crust in respect to the mantle. They are also more easily extracted from the descending slab and hence they will be added to the mantle through metasomatizing agents (fluids). Consequently, with time the source will generate magma of high 87/86Sri and low 143/144Ndi isotopic characteristic. Extremely high LaN/LuN ratio can be interpreted as evidence of low melting rate which is in accordance with the other incompatible el-ements content. Low melting rate (1–3%) can explain the very fractionated character of chondrite normalized distribution of REE (Fig. 4) and the distribution on the spiderdiagram, where left side incompatible elements (LILE, Th, U) present greater enrichment levels (Fig. 3).

The studied syenite closely correlates with the rocks of Svidnya pluton (Dyulgerov, 2005). These two plu-tons share many common features: they are emplaced as small plugs or dykes at hypabyssal level. Both plutons have identical petrographic compositions: clinopyroxene (diop side-augite), potassium feldspar, mica and sodic am-phibole, without plagioclases. Common accessories are apatite, zircon, titanite and oxides. The minor differences are the presence of magmatic carbonate in the studied sy-enite and the thin rim of aegirine-augite in the pyroxenes from Svidnya. The composition of both magmas is ma-fic to intermediate and presents slightly evolved, near primary liquids. They are peralkaline (A.I.≥1), showing very potassic character with K2O>Na2O and fraction-ated trend of REE. Isotopic compositions are identical and place the rocks in the enriched quadrant of Sr-Nd systematics (Dyulgerov et al., 2007). These geochemi-cal particularities imply on common source of the rocks from both plutons.

The rocks from calc-alkaline association and the studied potassic syenite were formed within a short

Fig. 6. Studied syenite (star) plotted on Ta/Yb– Th/Yb diagram (Pearce, 1983). Outlined for comparison are the fields of the other Variscan potassic-alkaline plutons (Shipka, Svidnya and Buhovo-Seslavtsi)

11

time interval and are consequently products of the same orogenic event. Both rock types also associate spatially occurring at distance of several tens of kilometres, thus building up one orogenic edifice. They have, however, completely different source characteristics: the calc-alkaline granitoids (both I and S types) are product of melting in crustal setting (in terms of T and P), whereas potassic syenite was generated in the mantle. The calc-al-kaline granitoids and potassic rocks (the studied syenite, Svidnya and Shipka plutons) are not genetically related which suggests that the magmas were emplaced in the crust via different structures. All these facts imply that the Variscan orogeny created a complex edifice within a narrow time span, and variable sources can be distin-guished. Generation of each magma batch was an inde-pendent event which took place at different depth.

The potassic rocks form small outcrops comparing to calc-alkaline granitoids (the studied syenite is < 0.5 km3). This can be explained in terms of small percentage of melting rate (1–3% as mentioned above) over limited region where the mantle was locally metasomatized (such amount of magma was generated in a source re-gion around 15–50 km3. The potassic rocks are dispersed as discrete bodies which can be interpreted as evidence that the mantle in the Variscan orogenic edifice had a patchy structure: metasomatized zones appeared as iso-lated regions in the mantle rather than forming a con-tinuous layer.

conclusIons

The studied potassic syenite shares similarities with the other alkaline plutonic rocks from the Balkan Mountains – Svidnya, Buhovo-Seslavtsi and Shipka. They intrude low-grade metamorphic rocks and the intermediate rock types are most voluminous – monzonite and syenite. The rocks demonstrate pronounced alkaline character as K2O dominates over Na2O. All rocks are charac-terized by strong incompatible elements enrichment, mostly LILE, REE, Th and U. This potassic syenite is very close petrographically to the melanocratic syenites from Svidnya pluton both consisting of potassium feld-spar (without plagioclase), abundant mica and clinopy-roxene and late sodic amphibole. Geochemically both rock types are near primary liquid and on the basis of their trace element signature an enriched mantle source can be supposed.

Acknowledgements

The author is indebted to A. von Quadt (ETH, Zurich) for performed LAICP MS analysis and to Peter Czaja (Humboldt University, Berlin) for part of the microprobe analyses. Careful revision of the manuscript by I. Seghedi substantially improved the quality of the text. Scientific discussions with P. Marchev and editorial handling by I. Lakova are much appreciated.

reFerences

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

Dimitrov, S. 1935. Die eruptivgesteine oberhalb von Seslavtci und Buhovo (nordöstlich von Sofia). Annual of Sofia University, Faculty of Mathematics and Physics 3(33), 189–246 (in Bulgarian with German abstract).

Dimitrov, S. 1937. Beitrag zur Kenntnis der kalireichen Alkaligesteine Bulgariens. Die Eruptivegesteine von Svidnja (im Westbalkan). Annual of Sofia University, Faculty of Mathematics and Physics 3(33), 291–343 (in Bulgarian with German abstract).

Dyulgerov, M. 2005. Le plutonisme de tendance alcalin potassique de Stara planina, Bulgarie: etude petrologique des complexes de Buhovo-Seslavtzi, Svidnya et Shipka. PhD thesis, Université Paris XI, France, 310 pp.

Dyulgerov, M., Platevoet, B., Schaltegger, U. 2007. Variscan potassic-alkaline magmatism in Stara Planina, Bulgaria – composition, source and geodynamic significance. Conference Abstract Goldschmidt, Cologne, A247.

Dyulgerov, M., Ovtcharova, M., Schaltegger, U. 2010. Unraveling the time of formation of potassic-alkaline rocks in the Variscan edifice in Stara planina, Bulgaria: ID – TIMS and LA-ICP-MS study. Abstracts of XIX Congress of CBGA, Thessaloniki, Greece, 106–107.

Ernst, W.G. 1968. Amphiboles. Crystal chemistry, phase relations and occurrence. Springer-Verlag, New York, 128 pp.

Hawthorne, F. C., Cooper, M. A., Grice, J. D., Ottolini, L. 2000. A new anhydrous amphibole from the Eifel region,

Germany: description and crystal structure of obertiite, NaNa2(Mg3Fe3+Ti4+)Si8O22. American Mineralogist 85, 236–241.

Janoušek, V., Rogers, G., Bowes, D. 1995. Sr-Nd isotopic con-straints on the petrogenesisof the Central Bohemian pluton, Czech Republic. Geologische Rundschau 84, 520–534.

Kalvacheva, R, Prokop, R. 1988. Fossil evidence (Devonian) crinoides for the age of metamorphic rocks in Shipka Balkan Mountains, Bulgaria. Comptes rendus de l’Academie bul-gare des Science 41(2), 91–94.

Kamenov, B., von Quadt, A., Peycheva., I. 2002. New insight into petrology, geochemistry and dating of the Vejen pluton, Bulgaria. Geochemistry, Mineralogy and Petrology 39, 3–26.

Kostov, I. 1950. Mineral and chemical compositions of the alkalic igneous rocks of the Shipka Mountain. Annual of Sofia University 46(3), 1–37 (in Bulgarian with English abstract).

Le Bas, M. J., Le Maître, R. W., Streckeisen, A., Zanetin, B. 1986. A. Chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27, 745–750.

Nelson, G., McCulloch, M., Sun, S-S. 1986. The origins of ul-trapotassic rocks as inferred from Sr, Nd and Pb isotopes. Geochimica and Cosmochimica Acta 50, 231–245.

Pearce, J. A. 1983. Role of sub-continental lithosphere in mag-ma genesis at active continental margins. In: Hawkesworth, C. J. and Norry, M. J. (Eds.) Continental basalts and mantle xenoliths. Shiva Publ. Ltd., Nantwich, Cheshire, 230–249.

12

Peytcheva, I., von Quadt, A., Malinov, O., Tacheva, E. 2006. Petrochan and Klissura plutons in Western Balkan: Relationships, precise U-Pb dating and isotope tracing. Proceedings of Annual Scientific Conference of the Bulgarian Geological Society “Geosciences” 2006, 221–224.

Rapp, R. P., Watson, E. B. 1995 Dehydration melting of metaba-salts at 8–32 kbar: implications for continental growth and crust-mantle recycling. Journal of Petrology 36, 891–931.

Rudnick, R. L., Gao, S. 2003. Composition of the continental crust. In: Rudnick, R. L. (Ed.) Continental crust. Treatise on Geochemistry (Holland, H. D. and Turekian, K. K. Eds.), Elsevier Pergamon, Oxford, 3, 1–64.

Sisson, T. W., Ratajski, K., Hankins, W. B., Glazner, A. F. 2005. Voluminous granitic magmas from common basaltic source.

Contributions to Mineralogy and Petrology 148, 635–661.Sun, S.-s., McDonough, W. F. 1989 Chemical and isotopic sys-

tematics of oceanic basalts: implications for mantle com-position and processes. In: Saunders, A. D., Norry, M. J. (Eds). Magmatism in the ocean basins. Geological Society, London, Special Publications 42, 313–345.

Taylor, S. R., McLennon, S. M. 1988. The continental crust: its composition and evolution. Mir, Moscow, 384 pp (in Russian).

Vollmer, R., Ogden, P., Schilling, J.-C., Kingsley, R., Waggoner, D. 1984. Nd and Sr isotopes in ultrapotassic volcanic rocks from Leucite Hills, Wyoming. Contributions to Mineralogy and Petrology 87, 359–368.

Wilson, M. 1989. Igneous petrogenesis: A global tectonic ap-proach. Spinger, Dodrecht, 466 pp.