Introduction

The Lut Block is a geotectonic unit, in eastern Iran, composed of lithologies that, in general, were not significantly affected by tectonic deformation since the Jurassic times. This unit is surrounded by highly deformed domains of clear oceanic affinity, with ophiolite series and flysch-type rocks, particularly to the north, south and east (Stöcklin 1972). The present eastern border of the Lut Block would have belonged to the active margin of the subducted Neotethys Ocean (Dercourt et al. 2000; Golonka 2004; Bagheri and Stampfli 2008). This ocean closed in eastern Iran, between the Afghan and Lut plates, in the Oligocene–Middle Miocene (Sengör and Natalin 1996). The East Iran ophiolite complex marks the boundary between the Lut and Afghan continental blocks (Fig. 1).

Fig. 1
figure 1

Modified geological sketch map of Iran after Berberian and King (1981). The point indicates the location of Dehsalm intrusives, and the box indicates the location of the Lut Block volcanic–plutonic belt

Full size image

The East Iran volcanic–plutonic belt extends for 1,000 km in the N–S direction, within the Lut Block (Figs. 1, 2). The magmatic activity, mainly with calc-alkaline signatures as shown by Berberian (1983), began in the middle Jurassic (165–162 Ma) and reached its peak in the Tertiary, especially in the middle Eocene. Volcanic and subvolcanic rocks of Tertiary age cover large areas of the Lut Block, attaining a thickness up to 3,000 m, and seem to have resulted from subduction prior to the collision of the Arabian and Eurasian plates (Camp and Griffis 1982; Tirrul et al. 1983; Berberian et al. 1999). According to Eftekharnejad (1981), magmatism in the northern Lut area resulted from subduction beneath the Lut Block. Recently, a two-sided asymmetric subduction model has been proposed to explain the Tertiary magmatic and metallogenic events recorded in the Lut Block (Arjmandzadeh et al. 2011). In this model, west-verging subduction beneath the Lut Block was steeper and faster, favoring the formation of great amounts of calc-alkaline magmas, as recorded within the Lut Block; in contrast, east-verging subduction, under the Afghan block, is testified by stronger tectonic deformation but less important magmatism.

Fig. 2
figure 2

Geological map of the Lut Block volcanic–plutonic belt and the location of Dehsalm intrusives. Adapted from Griffis et al. (1992)

Full size image

Various mineralization types, such as Cu–Mo–Au porphyry-type deposits, epithermal-type ores, Cu–Au–Ag IOCG-type deposits, Cu and Au–Sb–Pb–Zn vein-type deposits, Cu–Au massive sulfide-type deposits, granite-related Sn–W–Au ores and magmatic-skarn Sn deposits, formed during Jurassic to Tertiary stages of magmatism in the Lut Block (e.g., Malekzadeh 2009; Arjmandzadeh et al. 2011).

The Dehsalm porphyritic granitoids belong to the East Iran volcanic–plutonic belt (Fig. 2) and have associated Cu–Mo porphyry-type deposits (Arjmandzadeh et al. 2013). The purpose of this work is to present and discuss geochemical (both elemental and isotopic) and geochronological (Rb–Sr) data from those shallow intrusives, aiming at establishing tighter constraints on the petrogenetic processes and the geodynamic evolution of the Lut Block.

Geological setting and mineralization

The Lut Block is composed of pre-Jurassic metamorphic rocks and Jurassic sediments, intruded by Jurassic and Tertiary plutons, mainly of granitoids, and covered by Tertiary mafic to felsic lava flows and pyroclastic materials. Magmatism in the Lut Block, represented by a variety of lava flows, volcaniclastic rocks and subvolcanic and plutonic bodies, started in the late Jurassic with the intrusion of Shah-Kuh batholith and continued into the Quaternary (Esmaeily 2005). Most of the East Iran mineral deposits are related to the Tertiary magmatism (Arjmandzadeh et al. 2013).

The Dehsalm intrusive complex is located about 55 km west of Nehbandan, in the South Khorasan province. This complex is composed essentially of stocks intrusive into Eocene volcanics, sandstones and siltstones (Fig. 3). The intrusive rocks range from gabbro to granite, with a clear dominance of monzonite and quartz monzonite. Plagioclase is a major rock-forming mineral in most lithologies. K-feldspar is common as phenocrysts as well as in the matrix in the more felsic rock types; it also occurs as a minor phase, interstitial to plagioclase and ferromagnesian minerals in the mafic rocks. Biotite, clinopyroxene and hornblende are the mafic silicates present, in variable proportions, in the studied intrusions (Fig. 4a). Apatite and oxide minerals (magnetite and lesser ilmenite) are common accessory phases, especially in the most mafic rocks.

Fig. 3
figure 3

Geological map of Dehsalm area, after Arjmandzadeh et al. (2013)

Full size image
Fig. 4
figure 4

a Late-stage biotite in monzodiorite sample D4-245 (borehole). Besides forming large anhedral grains (that sometimes enclose plagioclase and/or clinopyroxene crystals), biotite also appears as thin rims around the opaque minerals and as patches inside clinopyroxene (PPL-10X). b K-feldspar megacrysts in biotite pyroxene quartz monzonite (XPL-10X). c Plagioclase showing complex zonation in a pyroxene-hornblende monzodiorite (XPL-10X). d Poikilitic texture in pyroxene-hornblende monzonite intrusive body. Pyroxene and plagioclase inclusions distributed throughout the K-feldspar poikilocryst

Full size image

Most of the felsic-intermediate intrusions display a porphyritic texture, due to the occurrence of millimeter-sized phenocrysts of plagioclase and K-feldspar, surrounded by a groundmass formed by crystals (mainly of feldspars and quartz) no larger than tenths of millimeter. In quartz monzonites, the length of K-feldspar phenocrysts may attain almost 1 cm (Fig. 4b). Plagioclase phenocrysts usually display compositional zoning (Fig. 4c), which is most commonly of the normal type, but, especially in the biggest ones, may also be oscillatory.

A pyroxene-hornblende gabbroic diorite displays a poikilitic texture (Fig. 4d), with large irregular and optically continuous K-feldspar grains hosting several small crystals of other minerals, dominantly plagioclase, hornblende and clinopyroxene.

The diorite usually contains variably oriented subhedral plagioclase laths that define a framework whose interspaces are occupied by small grains of mafic minerals, in an intergranular-like texture. In monzodiorite compositions, K-feldspar becomes a more abundant phase, but systematically with an interstitial character.

Crosscutting relations at surface exposures and in the diamond drill boreholes suggest that quartz monzonite stocks were the earliest, while the biotite granite (as small stocks and dikes) was the latest intrusions emplaced in the Dehsalm complex.

The quartz monzonite stocks have been affected by potassic alteration, represented by abundant secondary biotite. The secondary biotite alteration is overprinted by sericite–calcite–quartz alteration and cut by quartz + pyrite + galena + sphalerite + chalcopyrite veinlets (Fig. 5a). The quartz monzonite also hosts several quartz + pyrite + magnetite + molybdenite + chalcopyrite + anhydrite ± gold veins (Fig. 5b).

Fig. 5
figure 5

a Galena and sphalerite replaced by chalcopyrite. This setting belongs to the veins including paragenetic minerals such as quartz, pyrite, galena, sphalerite and chalcopyrite, which formed within the sericite–calcite–quartz alteration zone. b Molybdenite (Mo), chalcopyrite (Ccp) and pyrite (Py). Chalcopyrite seems to replace the other minerals, mainly through the molybdenite cleavages (PPL-10X). This setting belongs to the veins including paragenetic minerals such as quartz, anhydrite, magnetite, molybdenite, chalcopyrite, pyrite and gold

Full size image

Monzonites and diorites show weak or no mineralization potential, despite the fact that propylitization is not uncommon in monzonites.

Granites are variably sericitized and are cut by several types of veins and veinlets: quartz + pyrite + molybdenite; quartz + pyrite + chalcopyrite + arsenopyrite ± gold; and quartz + pyrite + galena + sulfosalts.

Previous studies on alteration, hydrothermal fluids and ore-forming processes indicated the occurrence of a Cu–Mo porphyry-type mineralization in the area (Arjmandzadeh et al. 2013).

Analytical techniques

Major and trace element analysis

After a detailed petrographic study (using transmitted and reflected light microscopy) of a large set of samples collected in various rock units, from both surface exposures and drill cores, fourteen of the least altered samples were selected for whole-rock geochemical elemental analysis. The samples were analyzed for major elements by wavelength-dispersive X-ray fluorescence (XRF) spectrometry of fused disks by a Philips PW 1410 XRF spectrometer at Ferdowsi University of Mashhad, Iran. Eleven of these samples were analyzed for trace elements using inductively coupled plasma-mass spectrometry (ICP-MS), following a lithium metaborate/tetraborate fusion and nitric acid total digestion, in the Acme Laboratories, Vancouver (Canada). Whole-rock analytical results for major element oxides and trace elements are listed in Table 1.

Table 1 Whole-rock major and trace element compositions of the studied intrusive rocks of Dehsalm
Full size table

Rb–Sr and Sm–Nd isotopic analysis

Sr and Nd isotopic compositions were determined for seven whole-rock samples and two mineral separates (plagioclase and biotite) of the Dehsalm granitoids at the Laboratório de Geologia Isotópica da Universidade de Aveiro, Portugal. Plagioclase and biotite were separated from sample D3-227 using magnetic separation procedures and purified by handpicking under a binocular microscope. The mineral separates were rinsed using double-distilled water and crushed in several steps to remove inclusions and then powdered in agate mortar. The selected powdered samples were dissolved with HF/HNO3 in Teflon Parr acid digestion bombs at 200 °C for 3 days. After evaporation of the final solution, the samples were dissolved with HCl (6 N) and dried. The target elements were purified using conventional ion chromatography technique in two stages: (1) separation of Sr and REE elements in ion exchange column with AG8 50 W Bio-Rad cation exchange resin and (2) purification of Nd from other lanthanide elements in columns with Ln Resin (ElChrom Technologies) cation exchange resin. All reagents used in the preparation of the samples were sub-boiling distilled, and the water was produced by a Milli-Q Element (Millipore) apparatus. Sr was loaded on a single Ta filament with H3PO4, whereas Nd was loaded on a Ta outer-side filament with HCl in a triple-filament arrangement. 87Sr/86Sr and 143Nd/144Nd isotopic ratios were determined using a Multi-Collector Thermal Ionization Mass Spectrometer (TIMS) VG Sector 54. Data were acquired in dynamic mode with peak measurements at 1–2 V for 88Sr and 0.8–1.5 V for 144Nd. Sr and Nd isotopic ratios were corrected for mass fractionation relative to 88Sr/86Sr = 0.1194 and 146Nd/144Nd = 0.7219. During this study, the SRM-987 standard gave an average value of 87Sr/86Sr = 0.710256(16) (N = 12; conf. lim = 95 %) and 143Nd/144Nd = 0.5121057(61) (N = 13; conf. lim = 95 %) for the JNdi-1 standard (143Nd/144Nd data are normalized to the La Jolla standard). The concentrations of Rb, Sr, Sm and Nd in the mineral separates and in two whole-rock samples (D3-227 and De-7) were determined by isotope dilution mass spectrometry method (IDMS), using a 87Rb/84Sr and 150Nd/149Sm double spike. The Rb–Sr and Sm–Nd isotopic compositions are listed in Table 2.

Table 2 Rb–Sr and Sm–Nd isotopic data from seven whole-rock samples, one plagioclase separate and one biotite separate of the Dehsalm granitoids
Full size table

Geochemistry

Major element geochemistry

The Dehsalm intrusive bodies have SiO2 contents from 52 to 69 wt% and plot mainly in the gabbroic diorite, diorite, monzodiorite, quartz monzonite and granite domains on the Middlemost (1985) diagram (Fig. 6). The samples plot in the fields of high-K calc-alkaline and shoshonitic series on the K2O versus SiO2 discrimination diagram proposed by Peccerillo and Taylor (1976) (Fig. 7), showing a strong potassium enrichment (1.57–5.87 K2O wt%) from the most mafic to the most felsic compositions. Since the Na2O (2.32–3.65 wt%) trend does not show any obvious correlations with silica enrichment, the K2O/Na2O ratios increase from 0.57 to 1.68 toward the more evolved compositions. On the other hand, MgO, FeOt, CaO, P2O5 and TiO2 decrease with increasing SiO2 (Fig. 8).

Fig. 6
figure 6

Na2O+K2O versus SiO2 diagram. Fields after Middlemost (1985)

Full size image
Fig. 7
figure 7

K2O versus SiO2 diagram. Fields after Peccerillo and Taylor (1976)

Full size image
Fig. 8
figure 8

Harker diagrams for the intrusive rocks of Dehsalm

Full size image

As a whole, the major element variation diagrams point to a differentiation mechanism controlled mostly by fractionation of clinopyroxene, plagioclase and hornblende, in agreement with the order of crystallization that can be inferred from textural criteria. The expected increasing Na contents in fractionating plagioclase with differentiation, precluding a clear Na2O enrichment in the evolved magmas, can explain the variation of K2O/Na2O ratio. Fractionation of apatite, and to some extent oxide minerals (Fe–Ti oxides), should also have played a role in magma differentiation, as testified by the constant and regular decreases in phosphorus, iron and titanium with increasing SiO2 contents. Similar trends have been reported for several porphyry copper deposits elsewhere (Mason and McDonald 1978; Eastoe and Eadington 1986; Dilles 1987).

The Al2O3/(CaO + Na2O + K2O) molar ratios are always below 1.1, showing that the Dehsalm intrusions are metaluminous or only slightly peraluminous, as is expected in both M- and I-type granitoids but not in S-type granitoids (White and Chappell 1983; Chappell and White 1992).

The Dehsalm intrusions have MgO contents from 0.92 to 4.6 wt%, and the magnesium numbers (Mg# = 100 * Mg/[Mg + Fe], using atomic proportions) are moderately high, ranging from 40.1 to 55.6.

Trace element geochemistry

Primitive mantle-normalized trace element spider diagrams display strong enrichments in large-ion lithophile elements (LILE) and those incompatible elements that behave similarly to LILE (Th and U) (Fig. 9). The most characteristic high-field strength elements (HFSE)—e.g., Nb, Zr, Y, Ti and HREE—have, compared to LILE, clearly lower normalized values; Nb and Ti, in particular, display negative anomalies (Fig. 9). These features are typical of subduction-related magmas, such as the calc-alkaline volcanic arcs of continental active margins (e.g., Gill 1981; Pearce 1983; Wilson 1989; Walker et al. 2001). High Sr and low Nb, Ta and Ti contents, as in the Dehsalm intrusions, are thought to be due to the absence of plagioclase and presence of Fe–Ti oxides in the residue in the source area of the parental magmas (Martin 1999); Nb and Ta impoverishment has also been attributed to earlier depletion events in the mantle source rocks (Woodhead et al. 1993; Gust et al. 1997). In the case of Ti, and taking into account the geochemical and petrographic evidence discussed above, its negative anomalies are also related to the fractionation of oxides. The phosphorus negative anomalies in the studied samples can be explained by fractionation of apatite.

Fig. 9
figure 9

Primitive mantle-normalized trace element spider diagram (Sun and McDonough 1989) for Dehsalm intrusives

Full size image

Rare-earth element patterns of the Dehsalm intrusives in chondrite-normalized plots display high degrees of REE fractionation, with strong enrichment in LREE (Fig. 10), as testified by the range of LaN/YbN values between 14.5 and 22.6. Their strong resemblance to each other suggests a common magma source and a similar trend of evolution.

Fig. 10
figure 10

Chondrite-normalized diagram (Boynton 1984), showing significant LREE enrichments and high degrees of REE fractionation for Dehsalm intrusives

Full size image

Most of the studied rocks have Eu/Eu* ratios from 0.88 to 0.99 (Table 1). Normally, a negative Eu anomaly develops with magma differentiation due to fractional crystallization of early, calcium-rich, plagioclase (Henderson 1984). However, at high fO2 conditions, Eu will be present mainly as Eu3+ and, therefore, only small amounts of Eu2+ will be available for incorporation in plagioclase (Drake and Weill 1975). This may be the explanation for the lack of distinct negative Eu anomalies in the Dehsalm intrusions. The occurrence of high fO2 conditions during magma differentiation is further supported by petrographic evidence, since oxides (especially magnetite) are common minerals in the most mafic compositions, and also by Harker diagrams for FeOt and TiO2 (Fig. 8), showing clear negative slopes as is typical of magma suites where the oxide minerals’ fractionation has a significant role since the early stages of differentiation (e.g., Miyashiro 1974; Miyashiro and Shido 1975; McBirney 1993).

In agreement with the metaluminous and high-K calc-alkaline characteristics of the Dehsalm granitoids, almost all samples plot in the volcanic arc granites domain in the diagrams proposed by Pearce et al. (1984), with a tendency toward the syn-collision granites (Fig. 11). Low Rb/Sr ratios, with the mean value of 0.15, also fit into the described geochemical signature.

Fig. 11
figure 11

Plot of the compositions of the Dehsalm intrusives on the geotectonic setting discrimination diagrams of Pearce et al. (1984) for granitoid rocks. WPG within-plate granites, VAG volcanic arc granites, ORG ocean ridge granites, syn-COLG syn-collisional granites

Full size image

The Sr/Y and La/Yb ratios are high (31.6–72.2 and 21.5–33.5, respectively) and overlap the values reported for adakites (Kepezhinskas et al. 1997; Castillo et al. 1999). However, when the compositions are plotted in the Sr/Y–Y and La/Yb–Yb discrimination diagrams (Fig. 12a, b), Y and Yb contents are generally higher than expected in typical adakites. More importantly, two of the most typical features of adakites, as shown by Defant and Drummond (1990), are high Na2O contents (3.5–7.5 wt%) and low K2O/Na2O ratio (~0.42), which clearly contrast with the K-rich compositions of the Dehsalm intrusives. The hypothesis that K enrichment could be mainly an effect of hydrothermal alteration is not supported by immobile trace element information, since the Dehsalm samples plot, in the Th–Co diagram (Hastie et al. 2007), in the high-K calc-alkaline and shoshonitic series fields (Fig. 13).

Fig. 12
figure 12

a Plot of Dehsalm intrusives on Y versus Sr/Y diagram. Fields after Defant and Drummond (1990). b Plot of Dehsalm intrusives on Yb versus La/Yb diagram. Fields after Defant and Drummond (1990)

Full size image
Fig. 13
figure 13

Plot of Deshalm intrusives in the Th–Co diagram. Fields after Hastie et al. (2007). Subhorizontal boundaries separate fields of magma series typical of subduction-related settings. Subvertical boundaries separate fields of volcanic rocks in those settings

Full size image

Therefore, a suitable petrogenetic model for the Dehsalm granitoids must reconcile both their shoshonitic nature, as revealed by petrography and most of the geochemical data, and the “adakitic” affinity, suggested by some trace element features. A recently defined (Xiao and Clemens 2007) category of adakites (C-type) displays K-rich compositions; however, as will be discussed below, the features of the Dehsalm granitoids show that these rocks are also distinct from C-type adakites.

Rb–Sr and Sm–Nd isotope geology

One of the least altered samples—D3-227—was selected for Rb–Sr geochronology. Biotite and plagioclase concentrates were obtained, and together with the whole-rock analysis, their Rb–Sr compositions gave an age of 33 ± 1 Ma (Fig. 14). Since the plagioclase and whole-rock data plot close to each other, the result is strongly dependent on the Sr isotopic composition of biotite and, accordingly, it must be viewed mainly as a biotite Rb–Sr age. An identical age within error (34 ± 1 Ma) was obtained (Arjmandzadeh et al. 2011) for a shallow felsic intrusive from Chah-Shaljami (~85 km to the northwest of Dehsalm, again in the Lut Block; Fig. 2), which belongs to a magmatic suite displaying geochemical features similar to those shown by the granitoids studied in the present work.

Fig. 14
figure 14

Plot of the whole rock–plagioclase–biotite isochron of sample D3-227

Full size image

Considering that the Dehsalm granitoids are subvolcanic, their post-emplacement cooling should have been fast and, therefore, the 33 Ma age may be considered as dating the magmatic event. As such, initial isotopic ratios and ε values were calculated for 33 Ma.

Sr and Nd isotopic compositions were determined for seven whole-rock samples. Initial 87Sr/86Sr and єNd values are tightly clustered in the ranges from 0.70470 to 0.70508 and from +1.5 to +2.5, respectively. In the єNdi versus (87Sr/86Sr)i diagram (Fig. 15), this cluster plots to the right of the so-called mantle array and overlaps the field of island-arc basalts. These isotopic compositions also overlap almost perfectly the isotopic data obtained for Chah-Shaljami samples (Arjmandzadeh et al. 2011).

Fig. 15
figure 15

єNdi-(87Sr/86Sr)i diagram for the Dehsalm intrusive rocks. The field of Cenozoic subducted oceanic crust-derived adakites was defined after Defant et al. (1992), Kay et al. (1993), Sajona et al. (2000) and Aguillón-Robles et al. (2001). The data for adakitic rocks directly derived from a thick lower crust are after Atherton and Petford (1993), Muir et al. (1995) and Petford and Atherton (1996). MORB mid-ocean ridge basalts, DM depleted mantle, OIB ocean-island basalts, IAB island-arc basalts. Initial ratios calculated for 33 Ma

Full size image

The very similar initial Sr and Nd isotopic compositions in the seven-sample cluster suggest that the Dehsalm intrusions are co-genetic, deriving from the same parental magmas by magmatic differentiation processes. Taking into account the IAB-like isotopic compositions of the studied rocks, the parental magmas may have been formed by partial melting in a supra-subduction mantle wedge (Stolz et al. 1996). The occurrence of gabbroic rocks in the Dehsalm suite provides additional evidence in favor of an origin of the parental magmas by melting of mantle peridotites, rather than by melting of mafic crust.

Discussion

Origin of the parental magmas

Some relevant geochemical features of Dehsalm intrusives—such as the high Sr/Y and La/Yb ratios, and the low HREE contents—are similar to those exhibited by adakites. These characteristics could be a consequence of melting of garnet amphibolite or eclogite facies rocks that may be found in subducted oceanic crust (Defant and Drummond 1990). However, other sources of adakitic parental magmas have been proposed, such as hydrous mantle peridotite (Stern and Hanson 1991), mafic rocks at the base of thickened lower crust (Zhang et al. 2001; Chung et al. 2003; Xiong et al. 2003; Hou et al. 2004; Wang et al. 2005; Guo et al. 2006) or delaminated mafic lower crust (e.g., Kay and Kay 1993; Defant et al. 2002; Gao et al. 2004; Guo et al. 2006; Lai et al. 2007; Liu et al. 2008a, b). Some authors (e.g., Castillo et al. 1999; Macpherson et al. 2006) consider that assimilation-fractional crystallization (AFC) processes must be taken into account to explain the genesis of the adakitic rocks.

The high Sr/Y and La/Yb ratios could be attributed to the retention of Y and HREE in residual garnet and hornblende (Defant and Drummond 1990). The strong LREE/HREE fractionation in adakites is classically interpreted as reflecting the presence of garnet and amphibole in the residue resulting from the partial melting of their source, whereas those minerals are not residual phases during the genesis of typical of the most common calc-alkaline magmas (Martin 1986). In this case, although not truly adakitic, the Dehsalm suite could be related to a garnet- and/or amphibole-bearing magma source.

The Ti–Nb–Ta negative anomalies are typical of all types of calc-alkaline magmas, and they may be explained by residual hornblende and/or Fe–Ti oxides (rutile, ilmenite) in the source of the parental magmas (Pearce and Norry 1979). However, since Nb and Ta are both highly incompatible in typical mantle assemblages and immobile during metasomatic events, their anomalies can, alternatively, be explained by the addition of slab components to the mantle wedge causing increase in several incompatible elements (namely LILE), but not in Nb and Ta (e.g., Turner et al. 2003; Wang et al. 2006; Tamura et al. 2011).

The plot of the Dehsalm samples on the єNdi—(87Sr/86Sr)i diagram (Fig. 15) shows that their compositions do not fit into an origin of the parental magmas by melting of thick lower crust or Cenozoic subducted oceanic crust as proposed for typical adakites. In contrast, they have Sr and Nd isotopic composition very similar to those of normal island-arc basalts, pointing to melting in a mantle wedge followed by magmatic differentiation.

Experimental studies demonstrate that Mg# is a useful index to discriminate melts purely derived from the crust from those coming from the mantle. Adakitic magmas, whether derived directly from partial melting of the subducted oceanic slab (MORB) or from lower crustal mafic rocks, usually show low Mg# (<40), regardless of melting degrees (Rapp and Watson 1995), while the studied intrusives have moderately high Mg#, varying from 40.1 to 55.6, thus providing additional evidence for the involvement of a mantle source in the origin of the parental melts. The occurrence, at Dehsalm, of mafic lithologies, with gabbro-dioritic compositions, also supports the hypothesis of a peridotitic mantle source.

The studied rocks, although displaying some resemblances with adakites, are markedly enriched in some elements, such as K and Rb, revealing a high-K calc-alkaline to shoshonitic signature.

The high potassium contents can be explained by decomposition of a K-rich phase (probably phlogopite) during the partial melting of a previously metasomatized mantle peridotite (Conceição and Green 2004).

Ascent of magmas through thickened continental crust could have been the cause of crustal contamination resulting in higher Rb/Sr and LILE/HFSE ratios and increase in K2O and Th contents due to assimilation and fractional crystallization (AFC) processes (Esperanca et al. 1992). However, if such mechanisms had extensively occurred, significant variation in Sr–Nd isotopic composition would become evident and correlations of isotopic ratios with SiO2 should be expected (Castillo et al. 1999). In addition, the very restricted range of both (87Sr/86Sr)i and єNdi precludes assimilation and fractional crystallization (AFC) as a major process in the generation of the diverse magma compositions of the Dehsalm suite. As such, high Rb/Sr and K2O values are most likely attributed to the source geochemistry. Ionov and Hofmann (1995) have shown from mantle xenoliths that amphiboles can have high K and very low Rb concentration while coexisting phlogopite is rich in both K and Rb. Thus, a strong participation of phlogopite decomposition (but not necessarily its complete melting) in the generation of the parental magmas would account for the potassium-rich nature and high Rb/Sr ratios displayed by the Dehsalm intrusives. Metasomatism of mantle peridotite by slab melts produces orthopyroxene, clinopyroxene, garnet, phlogopite, and richterite or pargasite (Sen and Dunn 1994; Rapp et al. 1999; Prouteau et al. 2001).

Hypotheses on the processes in the mantle source suffer from the fact that, as is common in studies on subduction-related magmatism (e.g., Turner et al. 2003, 2011), none of the Dehsalm samples (Mg# ≤ 55.6) represents directly a primary magma and, consequently, magma differentiation has also played a role even in the most mafic compositions. However, taking the complete set of geochemical evidence into account, probably the parental magmas originated by partial melting of metasomatized mantle peridotite. Contribution of phlogopite breakdown to the primitive melts would cause the high potassium contents, responsible for the shoshonitic signature of the studied rocks. Additionally, residual garnet and amphibole may have enhanced the LREE/HREE fractionation. The amphibole contribution could have taken place also as low-P fractionation, and therefore, its presence in the mantle source is not required. The role of garnet as a residual mantle phase is also debatable, taking into account that the high LREE/HREE ratios are accompanied by only small HREE fractionation (Fig. 10). In fact, Lin et al. (1989) have shown that, in some cases, melting processes in spinel peridotite sources may produce magmas with LREE enrichment but flat HREE. Turner et al. (2003) used the Tb/Yb ratio as an indicator of the participation of garnet in residual assemblages, and according to their modelling for the genesis of parental magmas of K-rich suites in a volcanic arc setting, Tb/Yb values around 0.4 (such as those obtained in the Dehsalm rocks) fit into a scenario of small amounts (~3 %) of residual garnet.

C-type adakites (Xiao and Clemens 2007), which have some geochemical resemblance to the studied rocks, have been interpreted as post-collisional granitoids resulting from melting of K-rich (meta-)basaltic, dioritic or tonalitic rocks at the base of overthickened crust and under a very strong geothermal gradient. Examples studied by Xiao and Clemens (2007) correspond to silicic magmas, with low Mg# and an isotope signature suggesting a source with a long crustal residence period (єNdi = −18.7; 87Sr/86Sri = 0.708048). In contrast, Dehsalm rocks include lithologies more mafic than typical C-type adakites and have relatively high Mg#, positive єNdi values and low 87Sr/86Sri ratios. Moreover, several lines of evidence suggest that during the Oligocene, the Lut Block was at the Neothethysian margin of the central Iran microcontinent (Shafiei et al. 2009) and not in a post-collisional setting.

Tectonomagmatic and metallogenic implications

Arjmandzadeh et al. (2011) recently proposed a two-sided asymmetric subduction model to explain the tectonomagmatic setting of the Lut Block. This model relates the voluminous Tertiary magmatism within the Lut Block to fast west-directed subduction and the abundant structural evidence in the Afghan Block to slower eastward subduction, in agreement with the correlation between convergence rate and volume of magmatism along subduction zones that Tatsumi and Eggins (1995) have shown to exist. The larger volumes of subduction predicted along west-directed slabs should favor the formation of greater amounts of arc-related magmas, as reported within the Lut Block.

A period of important magmatism and mineralization took place from middle Eocene to early Oligocene in the Lut Block. The location of major Tertiary mineralization occurrences within the Lut Block is shown in Fig. 16. Tarkian et al. (1983) ascribed an island-arc signature to late Eocene (42 Ma) to mid-Oligocene (31.4 Ma) volcanic rocks of Khur and Shurab from the Ferdows and Mud areas. K-rich calc-alkaline to shoshonitic andesitic rocks from Qaleh-Zari Cu–Au–Ag IOCG were dated at 40.5 ± 2 Ma by Kluyver et al. (1978). The Hired intrusion-hosted Au deposit is reported by Eshraghi et al. (2010) to be related to a post-Eocene quartz diorite porphyry stock intruded into Eocene andesitic volcanic, pyroclastic and sedimentary rocks. Malekzadeh (2009) inferred an island-arc tectonomagmatic setting for the middle Eocene (39 Ma) intermediate subvolcanic rocks of the Maherabad and Khopik Cu–Au porphyry deposit.

Fig. 16
figure 16

Major Tertiary mineralization occurrences associated with the Eocene–Miocene magmatism within the Lut Block. 1 Dehsalm Cu–Mo porphyry (Arjmandzadeh et al. 2013); 2 Chah-Shaljami Cu–Mo porphyry (Arjmandzadeh et al. 2011); 3 Qlaleh zari Cu–Au–Ag IOCG (Kluyver et al. 1978); 4 Hired Au–Sn associated with reduced granitoids (Eshraghi et al. 2010); 5 Khopik Cu–Au porphyry (Malekzadeh 2009); 6 Maherabad Cu–Au porphyry (Malekzadeh 2009); 7 Khur Cu–Pb–Zn–Sb vein-type mineralization (Tarkian et al. 1983); 8 Shurab Cu–Pb–Zn–Sb vein-type mineralization (Tarkian et al. 1983)

Full size image

Chah-Shaljami porphyritic granitoids were dated by Arjmandzadeh et al. (2011), using the Rb–Sr isotopic systematics of minerals and whole rock, at 33.5 ± 1 Ma. Richards et al. (2012) obtained an identical age (33.72 ± 0.08 Ma), within error, using the 40Ar/39Ar method in a sample from a quartz monzonite intrusion from the same area. The Chah-Shaljami rocks constitute a suite with high-K calc-alkaline features, although some trace element characteristics reveal an adakitic affinity. These granitoids plot almost completely in the field of the volcanic arc granites; however, they also straddle the boundary to the syn-collision granites. (87Sr/86Sr)i and εNdi isotopic ratios of Chah-Shaljami intrusives range from 0.70470 to 0.70506 and from +1.9 to +2.7, respectively, which fits into a supra-subduction mantle wedge source for the parental melts. The gathered data on alteration, mineralization and hydrothermal fluids together with field evidence indicate a deep Cu-Mo porphyry system in the Chah-Shaljami area.

The data presented in this work reveal that the Dehsalm subvolcanics are not only contemporaneous (33 ± 1 Ma) but also have very similar geochemical and isotope signatures compared to Chah-Shaljami granitoids, revealing that the intrusions of the two areas are testimonies of the same type of magmatic processes. Additionally, studies on the ore-forming processes in Dehsalm (Arjmandzadeh et al. 2013) concluded that a Cu–Mo porphyry-type mineralization system also existed in this area.

A spatial and temporal relationship between tectono-magmatic cycles in arc mineralization processes has long been recognized, with porphyry Cu deposits typically occurring in subduction-related settings, especially in continental arcs (e.g., Sillitoe 1988; Sillitoe and Bonham 1984), in relation to oxidized I-type granitoids. The relative importance of Cu and Mo in those deposits seems to be controlled by the water content of the initial magma, the water saturation level in each situation and the degree of crystal fractionation required to achieve water saturation (Candela and Holland 1984, 1986; Strong 1988; Candela 1992). Cu–Au porphyry deposits are usually considered more typical of relatively immature arcs (Cooke et al. 1998; Laznicka 2010), although they are also found in the Andean arc, as in the Maricunga belt, Chile (Vila and Sillitoe 1991).

The plot of geochemical data obtained for igneous rocks of the Lut Block on the Rb/Zr–Nb diagram (Brown et al. 1984; Fig. 17) agrees with that general picture. In fact, rocks with associated Cu-Au deposits have mid-Eocene age and lie in the island-arc field, while the Oligocene granitoids with Mo-bearing deposits (this work and Arjmandzadeh et al. 2011) display features of a more mature arc setting (Fig. 17), probably in relation to crustal thickening accompanying the beginning of the collision of the Afghan and Lut plates.

Fig. 17
figure 17

Plot of Rb/Zr–Nb for Dehsalm intrusive rocks. Fields after Brown et al. (1984). The field of Cu–Au porphyry was drawn mainly on the basis of Maher-abad and Khoopic prospects (after Malekzadeh 2009). The data for Chah-Shaljami are after Arjmandzadeh et al. (2011)

Full size image

Sillitoe (1998) remarked that crustal thickening associated with compressive tectonism was synchronous with the formation of giant porphyry copper systems in central and northern Chile, southwest Arizona, Irian Jaya and Iran. The existence of a thick crust (40–48 km) in the Lut Block was suggested by Dehghani and Makris (1983). More recently, Hatzfeld and Molnar (2010) after comparing structural evidence from the Himalaya and the adjacent Tibetan Plateau, on the one hand, and from Zagros and the Iranian Plateau, on the other hand, concluded that crustal thickening occurred beneath short ranges that link strike‐slip faults in the region surrounding the Lut Block. Therefore, the geodynamic setting of the Oligocene subvolcanic intrusives of the Lut Block seems to fit into the conditions favorable to the genesis of important porphyry copper deposits.

Conclusions

A Rb–Sr biotite date yields an intrusion age of 33 ± 1 Ma for the Dehsalm granitoids in the Lut Block volcanic–plutonic belt. These granitoids display trace element features typical of the magmatism related to a subduction zone, such as LILE enrichment and marked Nb, Ta and Ti negative anomalies. Geochemical evidence shows that the Dehsalm intrusives are high-K calc-alkaline to shoshonitic. They also belong to the magnetite series, with mineral potential for Cu–Mo (–Au–Pb–Zn), as detected by geochemical exploration surveys (Arjmandzadeh et al. 2013). Some geochemical resemblance, namely the high LREE/HREE ratios, between Dehsalm granitoids and adakitic rocks can be attributed to the presence of residual garnet and amphibole in a mantle source. The relatively high Mg# values discard a crustal origin (subducted slab or lower crust) for the parental magmas. Isotope geochemistry shows that the studied rocks are co-genetic and should be related to each other mainly by magmatic differentiation processes, such as fractional crystallization. Therefore, the high K2O contents should result from the mantle source geochemistry, rather than from important assimilation of crustal materials. Despite the fact that no primitive melt is directly represented by any of the studied rocks, some hypotheses on the processes involving the mantle source may be put forward: The parental magmas probably derived from partial melting of metasomatized peridotite in a supra-subduction mantle wedge; during the melting event, phlogopite breakdown should have contributed to some of the most important geochemical fingerprints of the suite; garnet and amphibole possibly remained as residual phases in the source. This study provides new evidence for subduction beneath the Lut Block during the Tertiary. A spatial and temporal relationship between tectono-magmatic cycles in the eastern Iran arc and porphyry Cu–(Mo–Au) formation has been recognized. Cu–Au porphyry deposits of the Lut Block seem to be related to an immature arc geotectonic setting during the middle Eocene, while Mo-bearing porphyry-type deposits correspond to a more advanced stage of arc evolution and probably to crustal thickening as a result of the beginning of Afghan and Lut plate collision during Oligocene.