Post-Mineralization, Cogenetic Magmatism at the Sungun Cu-Mo Porphyry Deposit (Northwest Iran): Protracted Melting and Extraction in an Arc System

: The Sungun porphyry ore deposit is located in Eastern Azarbaijan province, Northwestern Iran. The oldest intrusive pulse in the region is a quartz-monzonite pluton, which hosts the porphyry copper-molybdenum mineralization. The Sungun Copper Mine includes the mineralized Sungun porphyry as well as six groups of cross-cutting and lithologically distinct post-mineralization dykes. The composition of these a dioritic magma. This subsequently migrated to shallower levels to evolve further and feed individual dyke groups into the Sungun porphyry.


Introduction
The Sungun Cu-Mo porphyry is situated in the NW-trending part of the Central Iran post-collisional magmatic arc zone [1] known as Sahand-Bazman ( Figure 1). Porphyry copper deposits

Introduction
The Sungun Cu-Mo porphyry is situated in the NW-trending part of the Central Iran postcollisional magmatic arc zone [1] known as Sahand-Bazman ( Figure 1). Porphyry copper deposits form along convergent plate boundaries in continental magmatic arcs or in island-arc environments [2]. These deposits are associated with subduction-related volcanic centers, although in some examples, they are thought to be associated with post-collisional volcanism [2]. Giant porphyry copper deposits in Iran and Pakistan occur in a region along the suture zone between the Arabian (Afro-Arabian) and Eurasian plates following subduction of the Neo-Tethys oceanic plate [3]. Reference [4] noted that continental arc-style magmatism related to the subduction of the Neo-Tethys oceanic plate produced several porphyry deposits such as the porphyry copper-molybdenum deposit at Sungun (northwestern Iran), the giant gold-rich porphyry copper deposit at Sarcheshmeh, and porphyry copper deposit at Meiduk. The magmatic suites in this area are part of the NW-SE trending Cenozoic magmatic belt of Iran and the porphyries occur as stocks and dykes [5,6] (Figure 1).  [1,7]; (b) geologic map of Northwestern Iran, showing the locations of Mesozoic and Cenozoic igneous rocks and associated porphyry, epithermal, and Carlin-like deposits [7]. The location of the Sungun porphyry Cu-Mo deposit in the Sahand-Bazman belt is shown by a star [8].
Most porphyry copper deposits are associated to shallow-intermediate silicic intrusive complexes composed of small plutons and dykes. Some are associated solely to dykes. Porphyry Cu deposits are centered in porphyry intrusions that range from vertical, pluglike stocks, and are circular to elongate in a map view due to dyke arrays forming small, irregular bodies [9]. The stocks and dykes commonly have diameters and lengths of ≤1 km.
In most porphyry deposits around the world dyke suites have compositions associated to mineralized porphyry and are hence genetically related. In the copper belt of Iran, late and unmineralized dykes genetically related with the mineralized porphyry occur at Sarcheshmeh, Sara,  [1,7]; (b) geologic map of Northwestern Iran, showing the locations of Mesozoic and Cenozoic igneous rocks and associated porphyry, epithermal, and Carlin-like deposits [7]. The location of the Sungun porphyry Cu-Mo deposit in the Sahand-Bazman belt is shown by a star [8].
Most porphyry copper deposits are associated to shallow-intermediate silicic intrusive complexes composed of small plutons and dykes. Some are associated solely to dykes. Porphyry Cu deposits are centered in porphyry intrusions that range from vertical, pluglike stocks, and are circular to elongate in a map view due to dyke arrays forming small, irregular bodies [9]. The stocks and dykes commonly have diameters and lengths of ≤1 km.
In most porphyry deposits around the world dyke suites have compositions associated to mineralized porphyry and are hence genetically related. In the copper belt of Iran, late and unmineralized dykes genetically related with the mineralized porphyry occur at Sarcheshmeh, Sara, Haftcheshmeh and Darrehzar [10][11][12][13]. The majority of studies on the Sungun porphyry Mo-Cu deposit consist of exploration companies reports alongside master and PhD dissertations, and focused on surveying the deposit geometry as well as determining the mechanisms of mineralization [14][15][16][17][18][19][20][21][22][23][24][25][26][27].  between 30 m and 3 km. They are easily recognizable by their coarse and pink orthoclase phenocrysts. DK1b dykes contain enclaves (Figure 5c) that can be divided into two groups: Dioritic enclaves and microgranular enclaves with quartz dioritic composition. The diameter of enclaves varies from a few millimeters to 30 centimeters (Figure 5c).
DK1c is the third and youngest generation compared to DK1a and DK1b, based on cross cutting relationships. The grey color of DK1 is clearly different from all of the other rocks of the units at Sungun, making them easily recognizable in the field.    DK1c is the third and youngest generation compared to DK1a and DK1b, based on cross cutting relationships. The grey color of DK1 is clearly different from all of the other rocks of the units at Sungun, making them easily recognizable in the field.   Gabbroic dykes (DK2) have not been seen inside the Sungun deposit but are found on its Northeastern side. Gabbroic rocks (DK2) are dark grey in colour. This dyke group cuts quartz-diorite dykes and is cut by the diorite dykes (DK3). The diorite dykes (DK3) are younger than the quartzdiorite (DK1) and gabbroic dykes (DK2), and contain abundant hornfels enclaves (Figure 5d). These dykes can be seen in drill cores and outcrop in the Skarn portion. Lamprophyre and microdiorite dykes have formally been reported only from drill core observations ( Figure 6) [29,30]. Small microdiorite dykes are common in the Eastern skarn and are found in drill cores SUS_01 (Figure 6a), 03, 07, 17,41,43,44,46, and 52 ( Figure 2). At Sungun the microdiorite dykes cross cut quartz-diorite dykes DK1a and DK1b. Lamprophyre dykes (LAM) have not been reported at the surface but are found in drill cores SUS_24, 26,38,42, and 64 ( Figure 6b). The samples of the lamprophyre dykes are light green and grey in colour, fine-grained, and with porphyritic textures. . Two Sungun mine drill core logs representative of the diversisty of lithologies. (a) Core SUS01 contains a microdioritic dyke and (b) core SUS64 contains a lamprophyric dyke that have no surface exposure. SP is Sungun porphyry; DK1a, b and c are three phases of quartz dioritic dykes; DK2 is gabbrodiorite; DK3 is diorite; DK4 is dacite; MDI is a microdioritic dyke; LAM is a lamprophyric dyke; SKR is skarn; and HFS is hornfels.

Materials and Methods
A total of 200 rock samples from various lithologies in the Sungun deposit (20-SP, 95-DK1, 5-DK2, 35-DK3, 5-DK4, 20-LAM, and 20-MDI) were collected, of which 100 thin sections were studied by a polarized microscope to identify the mineral assemblages. Twenty nine representative samples and an additional thirty six from [26] were selected for whole-rock chemical analysis. Samples were analyzed for major elements by X-ray fluorescence using standard XRF techniques. Trace and rare-earth elements were analyzed by Lithium borate fusion ICP-MS at the ALS-Chemex Laboratories, Vancouver, Canada. For mineral chemistry 6 sample thin sections including amphiboles, biotites and plagioclase were selected and analyzed using the CAMECA SX100 at the Oklahoma University Microprobe Laboratory. Typical beam operating conditions were 20 kV and 20 nA.
Six samples from the post-mineralization dykes were analysed for Sr and Nd isotopic compositions in the Laboratory of Isotope Geology of the University of Aveiro, Portugal. Powdered samples were dissolved by HF/HNO3 solution in PTFE-lined Parr acid digestion bombs at a temperature of 180 °C for 3 days. After the final solution evaporation, the samples were dissolved in HCl (6.2 M) also in Figure 6. Two Sungun mine drill core logs representative of the diversisty of lithologies. (a) Core SUS01 contains a microdioritic dyke and (b) core SUS64 contains a lamprophyric dyke that have no surface exposure. SP is Sungun porphyry; DK1a, b and c are three phases of quartz dioritic dykes; DK2 is gabbrodiorite; DK3 is diorite; DK4 is dacite; MDI is a microdioritic dyke; LAM is a lamprophyric dyke; SKR is skarn; and HFS is hornfels.
Dacite dykes DK4 are related to Chaldaghi subvolcanic bodies, which have been attributed to the Plio-Quaternary [23]. These dacite dykes are covered by Plio-Quaternary epiclastic and pyroclastic rocks known as Dashdibi volcanics. Dacite dykes are grey to dark grey, fine grained, and containg mafic phenocrysts. Dacite dykes intrude the quartz monzonite of the Sungun porphyry. Dashdibi volcanics, as the latest and youngest episode of magmatism in the study area, cover approximately 2 km 2 of the altered SP quartz monzonite stock and post-mineralization dykes in the Southwest and West of Sungun. Dashdibi lavas are basaltic trachyandesite [23]. A total of 200 rock samples from various lithologies in the Sungun deposit (20-SP, 95-DK1, 5-DK2,  35-DK3, 5-DK4, 20-LAM, and 20-MDI) were collected, of which 100 thin sections were studied by a polarized microscope to identify the mineral assemblages. Twenty nine representative samples and an additional thirty six from [26] were selected for whole-rock chemical analysis. Samples were analyzed for major elements by X-ray fluorescence using standard XRF techniques. Trace and rare-earth elements were analyzed by Lithium borate fusion ICP-MS at the ALS-Chemex Laboratories, Vancouver, Canada. For mineral chemistry 6 sample thin sections including amphiboles, biotites and plagioclase were selected and analyzed using the CAMECA SX100 at the Oklahoma University Microprobe Laboratory. Typical beam operating conditions were 20 kV and 20 nA.

Materials and Methods
Six samples from the post-mineralization dykes were analysed for Sr and Nd isotopic compositions in the Laboratory of Isotope Geology of the University of Aveiro, Portugal. Powdered samples were dissolved by HF/HNO 3 solution in PTFE-lined Parr acid digestion bombs at a temperature of 180 • C for 3 days. After the final solution evaporation, the samples were dissolved in HCl (6.2 M) also in acid digestion bombs, and dried again. The elements analyzed were purified in two steps using conventional ion chromatography technique: Sr and REE elements separation in ion exchange columns was done using AG8 50 W cation exchange resin (Bio-Rad, Hercules, CA, USA); Nd separation from the other lanthanides was done with Ln cation exchange resin (Eichrom Technologies, Lisle, IL, USA). All reagents used in sample preparation were sub-boiling distilled and the water was produced by a Milli-Q Element System apparatus (Merck Millipore, Darmstadt, Germany). Strontium was loaded in a single Ta filament with H 3 PO 4 whereas Nd was loaded on a Ta outer side filament with HCl in a triple filament arrangement. 87 Sr/ 86 Sr and 143 Nd/ 144 Nd isotopic ratios were determined using a multi-collector thermal ionization mass spectrometer (TIMS) VG Sector 54 device. Data were acquired in dynamic mode with peak measurements at 1-2 V for 88 Sr and 0.5-1.0 V for 144 Nd. typical runs consisted of acquisition of 60 isotopic ratios. The isotopic ratios of Sr and Nd were corrected for mass fractionation relative to 88 Sr/ 86 Sr = 0.1194 and 146 Nd/ 144 Nd = 0.7219. During this study, the SRM-987 standard gave an average value of 87 Sr/ 86 Sr = 0.710249 ± 12 (conf. lim 95%, N = 12) and the JNdi-1 standard gave an average value of 143 Nd/ 144 Nd = 0.5121019 ± 58 (conf. lim 95%, N = 15).

DK1 (Quartz Diorite)
These dykes are dark gray to dark green in hand specimen and have massive texture. Post mineralization quartz diorite dykes in this region are subdivided into three categories based on alteration and relative emplacement age (DK1a, DK1b and DK1c).

DK1b
This rock is porphyritic with a microlithic porphyre textures groundmass and mainly consist of plagioclase, amphibole and biotite (Figure 7c). Plagioclase crystals mostly show alteration to sericite and epidote. Amphibole phenocrysts are partially altered to chlorite, but to a lesser degree than biotite, which is mostly altered to chlorite.

DK1b
This rock is porphyritic with a microlithic porphyre textures groundmass and mainly consist of plagioclase, amphibole and biotite (Figure 7c). Plagioclase crystals mostly show alteration to sericite and epidote. Amphibole phenocrysts are partially altered to chlorite, but to a lesser degree than biotite, which is mostly altered to chlorite.

DK1c
These dykes have dark green color. General textures are porphyritic with a microlithic groundmass. The quartz diorite consists of plagioclase, amphibole (Figure 7d), and biotite. Accessory minerals include quartz, titanite, and apatite.

DK3 (Diorite)
These dykes are dark green and commonly show a porphyritic texture with phenocrysts consisting of plagioclase, amphibole, K-feldspar, and accessory phases including quartz, apatite, and titanite. Most plagioclases are altered to epidote, chlorite, and calcite, and K-feldspar is altered to sericite and clay minerals (Figure 7e). Amphibole is altered to chlorite and calcite.

DK4 (Dacite)
These are dark grey in color and have porphyritic textures consisting of plagioclase and sanidine. Amphibole is an accessory mineral.

MDI (Microdiorite)
Micro diorite is dark grey in color and commonly consists of phenocryst, groundmass, and cognate inclusions (Figure 7f). Phenocrysts mainly consist of feldspar and quartz.

LAM (Lamprophyre)
In hand sample the color is gray to bright green. The rock is porphyritic with a microlitic groundmass. The main minerals include biotite (Figure 7g,h), amphibole, plagioclase, alkali feldspar, and pyroxene. Accessory minerals include apatite (Figure 7h), zircon, and quartz.

Plagioclase
Representative compositions of plagioclase in Sungun deposit post-mineralization dykes, are reported in Table 1 and illustrated in Figure 8. Plagioclase in quartz diorite dykes (DK1) ranges from andesine to oligoclase with some nearly pure albite. Dacite dykes have the most anorthitic felsdspars, ranging from labradorite to andesine. Lamprophyre dykes contain oligoclase to albite, whereas microditorite dykes contain nearly pure albite.

Amphibole
Representative compositions of amphiboles are presented in Table 2. Based on the Mg/(Mg + Fe 2+ ) vs. Si classification diagram, amphiboles from quartz diorite, diorite and lamprophyre dyke samples, plot in the fields of magnesiohornblende, and tschermakite ( Figure 9). All amphiboles have between 6 and 7 Si atoms per formula unit (apfu), values which are typical of igneous amphiboles [32].

Biotite
Representative biotite compositions are reported in Table 3. The biotites of quartz diorite and lamprohyre dykes generally do not show differences in chemical composition and are classified as Mg-rich biotite and phlogopite with XFe = Fe 2+ /Fe 2+ + Mg) = 0.28-0.49 ( Figure 10).

Whole Rock Geochemistry
The results of major and trace elements analysis of the Sungun porphyry and of post-mineralization dykes are presented in Tables 4-6. The amount of SiO 2 vary from 48.18 to 69.24, and of MgO from 1.47 to 8.27. On the SiO 2 versus Zr/TiO 2 diagram [35] the sample plot in the andesite, dacite, and rhyodacite fields, which in plutonic terminology are equivalent to diorite, quartz diorite and granodiorite respectively ( Figure 11a). Lamprophyre dykes plot in the field of alkaline basalt. The Ta/Yb-Th/Yb diagram [36] shows the shoshonitic nature of the main intrusive suite at Sungun except for the microdiorite samples that have calk-alkaline affinities (Figure 11b).

Whole Rock Geochemistry
The results of major and trace elements analysis of the Sungun porphyry and of postmineralization dykes are presented in Tables 4-6. The amount of SiO2 vary from 48.18 to 69.24, and of MgO from 1.47 to 8.27. On the SiO2 versus Zr/TiO2 diagram [35] the sample plot in the andesite, dacite, and rhyodacite fields, which in plutonic terminology are equivalent to diorite, quartz diorite and granodiorite respectively (Figure 11a). Lamprophyre dykes plot in the field of alkaline basalt. The Ta/Yb-Th/Yb diagram [36] shows the shoshonitic nature of the main intrusive suite at Sungun except for the microdiorite samples that have calk-alkaline affinities (Figure 11b).

Trace Element Geochemistry
The chondrite-normalized rare earth element [39] patterns of the Sungun stock and postmineralization dykes show a steep slope from light rare earth elements (LREE) (LaN/SmN = 6.80) to the heavy rare earth elements (HREE) (GdN/YbN = 2.49) (Figure 13a), but do not have conspicuous Eu anomalies. Their primitive-mantle normalized trace element diagrams display enrichment in large ion lithophile elements (LILE) and depletion of high-field-strength elements (HFSE), with positive K,

Trace Element Geochemistry
The chondrite-normalized rare earth element [39] patterns of the Sungun stock and post-mineralization dykes show a steep slope from light rare earth elements (LREE) (LaN/SmN = 6.80) to the heavy rare earth elements (HREE) (Gd N /Yb N = 2.49) (Figure 13a), but do not have conspicuous Eu anomalies. Their primitive-mantle normalized trace element diagrams display enrichment in large ion lithophile elements (LILE) and depletion of high-field-strength elements (HFSE), with positive K, U, Pb, Nd, and Cs and negative Nb, P, Zr, Pr, and Ti anomalies, consistent with a subduction-related setting (Figure 13b). Their Nb, Ta, and Ti depletions are thought to be due to the absence of plagioclase and presence of Fe-Ti oxides (e.g., rutile) as residue in the source area of the parental magmas [40]; Nb and Ta depletion may also result from previous depletion events in the mantle source rocks [41,42]. Phosphorous also shows negative anomalies in the studied samples, which may be related to apatite fractionation. These chemical properties are analogous to those of the Sungun Porphyry. to each other and are therefore likely to have a common origin. These three samples together with LAM and DK3 define a trend (red arrow on Figure 14). The microdiorite dyke (MDI) has 87 Sr/ 86 Sr initial ratio of 0.706464, which is higher compared to the other samples (0.704554 to 0.705139) although its epsilon Nd value overlaps with those of the quartz-diorite (DK1) samples.
Rare earth element (REE) chondrite-normalized patterns for microdiorite dykes are displayed in Figure 13c. Microdiorite dykes have LREE enrichment with weak negative Eu anomalies (Eu/Eu* = 0.93-0.97) and shallow HREE slope (Figure 13d). The minor negative Eu anomaly is likely related to the abundance and fractionation of plagioclase in the rocks. On a multi-element primitive mantle normalized diagram [43], microdiorite dykes show enrichments in most large ion lithophile elements (LILE), and a fairly regular depletion of high field strength elements (HFSE).

Nd-Sr Isotopes
Sm-Nd and Rb-Sr isotopic data for six whole-rock samples of the mafic-intermediate dykes are provided in Table 7 and the initial epsilon values (20 Ma); calculations made with the 87 Rb decay constant recommended by Reference [44] are illustrated in Figure 14. The 87 Sr/ 86 Sr ratios range from 0.704617 to 0.706464 and 143 Nd/ 144 Nd ratios range from 0.512648 to 0.512773. The similarity in trace element characteristics between the Sungun porphyry and diorite and quartz-diorite dykes within the area of the mine suggests that the dykes are comagmatic with the quartz-monzonite. The quartz diorite dykes are also isotopically associated to the Sungun Porphyry. Within the dyke suite a trend is seen from lower to higher 87 Sr/ 86 Sr. This could have two most plausible explanations: (1) The samples would represent rocks whose parental magmas came from a progressively more enriched mantle compared to DK1c to DK3; or (2) assuming a single parental melt (in which case, DK1c is the sample that more closely represents its isotopic signature) the isotopic variation represent minor degrees of contamination, assimilation or mixing with crustal materials.   The variations in the 87 Sr/ 86 Sr and 143 Nd/ 144 Nd ratios of the Sungun dykes DK1, DK3, MDI, and LAM is shown and compared with bulk Earth and the mantle array ( Figure 14). Five samples from the quartz diorite, dacite, and lamprophyre dykes plot just to the right of the mantle array with a potential minor contribution from subduction derived fluid. The three DK1 samples plot very close to each other and are therefore likely to have a common origin. These three samples together with LAM and DK3 define a trend (red arrow on Figure 14).
The microdiorite dyke (MDI) has 87 Sr/ 86 Sr initial ratio of 0.706464, which is higher compared to the other samples (0.704554 to 0.705139) although its epsilon Nd value overlaps with those of the quartz-diorite (DK1) samples.
According to the study by reference [27], the Sungun porphyry stock and the post-mineralization dykes have unrelated origins. We show here, however, that based on trace elements and Sm-Nd isotopic results [27], the dykes intruding the Sungun stock are genetically related to the porphyry.
The similarity in trace element characteristics between the Sungun porphyry and diorite and quartz-diorite dykes within the area of the mine suggests that the dykes are comagmatic with the quartz-monzonite. The quartz diorite dykes are also isotopically associated to the Sungun Porphyry. Within the dyke suite a trend is seen from lower to higher 87 Sr/ 86 Sr. This could have two most plausible explanations: (1) The samples would represent rocks whose parental magmas came from a progressively more enriched mantle compared to DK1c to DK3; or (2) assuming a single parental melt (in which case, DK1c is the sample that more closely represents its isotopic signature) the isotopic variation represent minor degrees of contamination, assimilation or mixing with crustal materials.

Interpretations and Discussion
Tectonic Setting of the Sungun Stock and Post-Mineralization Dykes Potassic rocks are found in several tectonic environments, such as cratons, post-collisional setting, and active orogenic belts and, to a lesser extent, oceanic intraplate settings [46] and have moderate abundances of LREE, LILE, and HFSE [47]. Shoshonitic rocks in particular are associated with many continental and oceanic arcs [48]. The tectonic affinities of ancient and altered/mineralized rocks can be investigated with the use of immobile trace elements. A significant feature of volcanic arc magmatism is the depletion in HFSE relative to other incompatible elements as well as their LILE enrichment (e.g., [49,50]). At Sungun the average Ba/Nb and Rb/Nb ratios are 37.5 and 10.5 in felsic dykes and 62.5 and 7.6 in the intermediate-mafic dykes, respectively, which is a feature of arc environments [51]. The Th/Yb versus Ta/Yb diagram [52] highlights the Shoshonitic character of the Sungun rocks with the Th enrichment indicative of subduction fluids input (Figure 15a). Based on the diagram of [53], high Th/Yb ratios correlated with high values for La/Yb are consistent with continental arc magmas (Figure 15b), in agreement with the Y vs. Zr plot (Figure 15c). Based on the Ce/P 2 O 5 vs. Zr/TiO 2 diagrams of [47] the studied samples plot in the post-collisional arc related fields (Figure 15d).
In summary, the combination of field, petrographic and geochemical data of the Sungun stock and post-mineralization dykes shows that they formed during a post-collisional stage. This is in agreement with previous reports on the petrogenesis of igneous rocks in the Urumieh-Dokhtar magmatic belt [54][55][56]. Indeed, recent tectono-magmatic reviews have suggested that the Eocene magmatic event may have been a consequence of the geometrically complex subduction of the Neotethys oceanic crust underneath Central Iran, with the Sungun post-mineralization dykes intruding in the late stages of continental collision. The negative Eu anomalies can be generated by melting of a source rich in residual feldspar and/or plagioclase fractionation during ascent and emplacement of these intrusions [57]. The steep REE slope on the other hand is indicative of residual garnet in in the mantle source [58,59], hence precluding the presence of plagioclase as the aluminous phase in the mantle. The generally steep REE slope of the multi-element patterns is also typical of lamprophyres generated by small degrees of partial melting. The primitive mantle-normalized patterns of lamprophyre dykes (Figure 13f) show the characteristic enrichments in Th, Rb, K, Pb, Nd, Sm, and Dy, and depletion in Ba, Ce, Nb, Ta, Sr, Zr, and Ti, which are compatible with a geodynamic environment involving subduction zone magmatism [60]. This enrichment in incompatible elements implies that the melt source from which the magmas were derived was likely to be metasomatized lithospheric mantle enriched in K and incompatible elements [61]. with many continental and oceanic arcs [48]. The tectonic affinities of ancient and altered/mineralized rocks can be investigated with the use of immobile trace elements. A significant feature of volcanic arc magmatism is the depletion in HFSE relative to other incompatible elements as well as their LILE enrichment (e.g., [49,50]). At Sungun the average Ba/Nb and Rb/Nb ratios are 37.5 and 10.5 in felsic dykes and 62.5 and 7.6 in the intermediate-mafic dykes, respectively, which is a feature of arc environments [51]. The Th/Yb versus Ta/Yb diagram [52] highlights the Shoshonitic character of the Sungun rocks with the Th enrichment indicative of subduction fluids input (Figure 15a). Based on the diagram of [53], high Th/Yb ratios correlated with high values for La/Yb are consistent with continental arc magmas (Figure 15b), in agreement with the Y vs. Zr plot (Figure 15c). Based on the Ce/P2O5 vs. Zr/TiO2 diagrams of [47] the studied samples plot in the post-collisional arc related fields (Figure 15d). Figure 15. (a) Th/Yb vs. Ta/Yb diagram [52] for the Sungun post mineralization dykes. Tho is tholeiitic, CA is calc-alkaline, Sho is shoshonitic; (b) La/Yb vs. Th/Yb geotectonic discrimination diagram after [53]. The sungun stock and late dykes plot in the Continental Margin Arc fields; (c) Y vs. Zr geotectonic discrimination diagrams of [47], for within-plate and from arc related and (d) Ce/P2O5 vs. Zr/TiO2 geotectonic discrimination diagrams of Reference [47], samples plot mostly in the PAP field. PAP is post collision arc and CAP is active continental margin.
In summary, the combination of field, petrographic and geochemical data of the Sungun stock and post-mineralization dykes shows that they formed during a post-collisional stage. This is in agreement with previous reports on the petrogenesis of igneous rocks in the Urumieh-Dokhtar magmatic belt [54][55][56]. Indeed, recent tectono-magmatic reviews have suggested that the Eocene magmatic event may have been a consequence of the geometrically complex subduction of the Neotethys oceanic crust underneath Central Iran, with the Sungun post-mineralization dykes Figure 15. (a) Th/Yb vs. Ta/Yb diagram [52] for the Sungun post mineralization dykes. Tho is tholeiitic, CA is calc-alkaline, Sho is shoshonitic; (b) La/Yb vs. Th/Yb geotectonic discrimination diagram after [53]. The sungun stock and late dykes plot in the Continental Margin Arc fields; (c) Y vs. Zr geotectonic discrimination diagrams of [47], for within-plate and from arc related and (d) Ce/P 2 O 5 vs. Zr/TiO 2 geotectonic discrimination diagrams of Reference [47], samples plot mostly in the PAP field. PAP is post collision arc and CAP is active continental margin.

Petrogenesis of Quartz Monzonite Stock, Quartz Diorite and Diorite Dykes
The 87 Sr/ 86 Sr and 143 Nd/ 144 Nd vs. SiO 2 diagrams of the Sungun dykes are used to examine any roles of crustal contamination superposed to the principal fractional crystallization control of magmatic evolution. The isotope ratios do not show any obvious correlation with SiO 2 contents (Figure 16a,b) and incompatible elements abundance, such as Th (Figure 16c). Hence, we suggest that crustal assimilation did not play an important role during the magmatic evolution of the quartz diorite and diorite magmas, whereas the high 87 Sr/ 86 Sr of the microdiorite dykes suggests that it may have been affected by such processes. roles of crustal contamination superposed to the principal fractional crystallization control of magmatic evolution. The isotope ratios do not show any obvious correlation with SiO2 contents (Figure 16a,b) and incompatible elements abundance, such as Th (Figure 16c). Hence, we suggest that crustal assimilation did not play an important role during the magmatic evolution of the quartz diorite and diorite magmas, whereas the high 87 Sr/ 86 Sr of the microdiorite dykes suggests that it may have been affected by such processes. To determining the degree of enrichment or depletion in the samples from the study area we used the ratios of incompatible elements Y vs Zr [63] (e.g., Figure 15c). These elements are useful to interpret the petrogenesis of igneous rocks, even in highly altered rocks, due to their very low mobility [64]. Based on their incompatible element characteristics ( Figure 15) the magma of Sungun porphyry and post-mineralization dykes originated from mantle enriched by subduction processes. To determining the degree of enrichment or depletion in the samples from the study area we used the ratios of incompatible elements Y vs. Zr [63] (e.g., Figure 15c). These elements are useful to interpret the petrogenesis of igneous rocks, even in highly altered rocks, due to their very low mobility [64]. Based on their incompatible element characteristics ( Figure 15) the magma of Sungun porphyry and post-mineralization dykes originated from mantle enriched by subduction processes.
Previous studies on the genesis and evolution of Neotethys show the different opening ages. Different ages are due to uncertainty in the transition from extensional tectonics to the onset of contraction. Moreover, different researchers considered the time for oceanic closing and collision events to range from upper Cretaceous-Paleocene (Laramie phase) [55,65,66], Oligocene [67], and late Miocene-Pliocene [68][69][70][71]. Due to these complexities in Neotethys evolution more data are needed to clarify its tectonic history and particularly its closing stages.
It can be said, in summary, that some researchers have related the Neotethys closing to the upper Cretaceous, emphasizing the oceanic extension from the Kermanshah to the Esphandeqeh and Neyriz [65]. They relate the Eocene and Oligocene magmatic activities of Central Iran and the Alborz-Azerbaijan crust to the tectonic and petrogenetic evolution after the collision [65].
On the other hand, other researchers do not consider the origin of the Upper Cretaceous Ophiolite mélanges as resulting from collision of the Arabia plate with Iran. They argue this because Ophiolite mélanges are also present in Makran, where there is continent collision and the subduction is still ongoing. S-type granites, such as the Alvand Granite, adjacent to the ophiolites strongly support collision occurrence and end of subduction in this area, whereas no S-type granites occur in the Makran area. Considering these evidences it can be argued that this extended basin with NW-SE trend, closed during the Upper Cretaceous and Laramie phase of the Arabia Plate and Iran block collision. After that, during the Pyrenean phase, various fractures have been created in Central Iran, Alborz, and Azerbaijan while subduction fluids influenced the mantle wedge [6] imparting a clear subduction geochemical signature on it. In the last stages of Arabian Plate and Iran block convergence, the Pyrenean phase induced a strong compressive regime, which caused early formed mafic melts to pond in the crust, providing favourable conditions for melting the thick continental crust and generating hybrid magmas [65].
The sialic back arc basins contracted after the closing of the main oceanic basin, causing the generation of thrusts and inverse faults [65]. Subsequently these basement fractures were reactivated in Central Iran during tensional events. This caused localized upwelling in the metasomatized mantle wedge, leading to partial melting that formed the Tertiary magmatism of the Urmia-Dokhtar and Alborz-Azerbaijan zones. Due to the subduction, Neotethys Ocean closing and the post collision processes, several porphyry copper deposits were formed [65]. The important Iran copper belt (Urmia-Dokhtar), contains volcanic and plutonic rocks and extends in a NW-SE direction. Its SE and NW extensions are known as the Kerman copper belt and Arasbaran copper belt, respectively.
The Sungun porphyry stock and associated dykes were derived from partial melting in the lithospheric mantle metasomatised and enriched by the release of fluids from the Neothetys oceanic crust and subducted pelagic sediments, causing enrichment of LILE and LREE and depletion of HFSE. Sr and Nd isotopic data, as well as trace elements, indicated magmas feeding the post-mineralization dykes were generated in the LILE and LREEe enriched lithospheric mantle. Magma generation and ascent was therefore likely promoted by tectonic tensions after continental collision.
Based on the overall trace element and isotopic similarities, we consider that the Sungun Porphyry and dioritic dykes (DK1 and DK3) all evolved from a common parent melt. Thermo-barometric models [28] of the dioritic parent suggest the presence of a magma chamber at a maximum of 19 kbar (approximately from 57 to 60 km). Secondary magma chambers existed up to depth of 12-14 km (6 kbar), where magma underwent gravity settling and differentiation [28].
The chemical composition of younger dykes is more mafic than older ones. The more felsic member of the magmatic system is represented by the Sungun quartz-monzonite (SP) porphyry, and was likely the more evolved and higher portion of a layered magma chamber ( Figure 17). The more mafic member, DK3, was being stored at the base of the chamber. The DK1 magma represent intermediate levels within the layered magma chamber (Figures 18 and 19), which were extracted after the emplacement of the Sungun porphyry, and prior to the emplacement of the lower mafic portion as DK3 dykes.

Petrogenesis of Lamprophyre Dykes
Lamprophyre dykes cross cut the Sungun copper deposit and various other rock units. According to the results of borehole logging in this mine and geothermometry results [29], these dykes were intruded at least after the Oligo-Miocene. In order to evaluate the source of magma, and the degree of partial melting of the mantle, we can use a diagram of La/Sm vs. La [72] (Figure 18). All the samples have a high La/Sm ratio, which is consistent with an enriched mantle source, and less than 5% partial melting of garnet-lherzolite (Figure 18a). Trace elements can also give us information on the influence of hydrous mineral phases on the melting process, as elements such as Ba and Rb are compatible in phlogopite, whereas Ba, Sr, Rb, are only moderately compatible in amphibole [59]. As a consequence, melts in equilibrium with phlogopite will have high values of the Rb/Sr and lower values of Ba/Rb compared to melts in equilibrium with amphibole ( Figure 18b). The data illustrate how the lamprophyre magma was likely formed from mantle with residual phlogopite.
The lamprophyre dykes are homogeneously characterized by high abundances of incompatible trace elements (LILE and LREE) and they are distinguished by significant negative Ti, Zr, Sr, Ta, Nb, and Ba anomalies (Figure 13e,f), indicating that they were not derived from normal MORB or OIB source mantle (e.g., [73,74]). Moreover, the average Ce/Pb ratios (~11) of the most primitive samples are considerably different from those of oceanic basalts (~25; [75]), suggesting that these rocks are not derived from normal asthenospheric mantle. than 5% partial melting of garnet-lherzolite (Figure 18a). Trace elements can also give us information on the influence of hydrous mineral phases on the melting process, as elements such as Ba and Rb are compatible in phlogopite, whereas Ba, Sr, Rb, are only moderately compatible in amphibole [59]. As a consequence, melts in equilibrium with phlogopite will have high values of the Rb/Sr and lower values of Ba/Rb compared to melts in equilibrium with amphibole ( Figure 18b). The data illustrate how the lamprophyre magma was likely formed from mantle with residual phlogopite. The lamprophyre dykes are homogeneously characterized by high abundances of incompatible trace elements (LILE and LREE) and they are distinguished by significant negative Ti, Zr, Sr, Ta, Nb, and Ba anomalies (Figure 13e,f), indicating that they were not derived from normal MORB or OIB source mantle (e.g., [73,74]). Moreover, the average Ce/Pb ratios (~11) of the most primitive samples are considerably different from those of oceanic basalts (~25; [75]), suggesting that these rocks are not derived from normal asthenospheric mantle.
Furthermore, reference [76] as well as reference [77] showed that HFSEs such as Nb and Ta are depleted in the lithospheric mantle compared to LREEs. For instance, the ratio of Nb/La > 1, indicates an OIB-like asthenospheric mantle source for basaltic magmas, whereas a ratio lower than 0.5 indicate a lithospheric mantle source ( Figure 19). In the Sungun calc-alkaline Lamprophyres Nb/La is between 0.49 and 0.20, confirming their origin from the lithospheric mantle. On the other hand, the low amount of Nb/La may be a consequence of mantle metasomatism by subduction fluids, introducing more La with respect to Nb due to its greater mobility in subduction fluids. Ba/Rb ratios that may constrain source mineralogy [59]. Furthermore, reference [76] as well as reference [77] showed that HFSEs such as Nb and Ta are depleted in the lithospheric mantle compared to LREEs. For instance, the ratio of Nb/La > 1, indicates an OIB-like asthenospheric mantle source for basaltic magmas, whereas a ratio lower than 0.5 indicate a lithospheric mantle source ( Figure 19). In the Sungun calc-alkaline Lamprophyres Nb/La is between 0.49 and 0.20, confirming their origin from the lithospheric mantle. On the other hand, the low amount of Nb/La may be a consequence of mantle metasomatism by subduction fluids, introducing more La with respect to Nb due to its greater mobility in subduction fluids. . Nb/La versus La/Yb variation diagram. Average OIB is after [51] and average lower crust is after Reference [78]. Dashed lines separating fields of the asthenospheric, lithospheric and mixed mantle are plotted based on data given in Reference [77]. The HIMU-OIB area is reported in Reference [79].
Overall the major and trace elements and Sr-Nd isotopic characteristics of lamprophyre dykes are similar to previously intruded SP and post-mineralization dykes, suggesting they derived from a similar source. Our data suggests that the lamprophyres dyke originated from a low degree of partial melting of an enriched lithospheric mantle source and were emplaced in a post collisional tectonic setting. The fluxing of the mantle wedge by the subduction fluids caused the formation of phlogopite, which was likely a residual phase during melting to form the lamprophyre. Figure 19. Nb/La versus La/Yb variation diagram. Average OIB is after [51] and average lower crust is after Reference [78]. Dashed lines separating fields of the asthenospheric, lithospheric and mixed mantle are plotted based on data given in Reference [77]. The HIMU-OIB area is reported in Reference [79].

Petrogenesis of Microdirite Dykes
Overall the major and trace elements and Sr-Nd isotopic characteristics of lamprophyre dykes are similar to previously intruded SP and post-mineralization dykes, suggesting they derived from a similar source. Our data suggests that the lamprophyres dyke originated from a low degree of partial melting of an enriched lithospheric mantle source and were emplaced in a post collisional tectonic setting. The fluxing of the mantle wedge by the subduction fluids caused the formation of phlogopite, which was likely a residual phase during melting to form the lamprophyre.

Petrogenesis of Microdirite Dykes
Isotope data of the microdiorite (MDI) sample are typical of rocks that underwent hydrothermal alteration, which may affect the Sr isotopic signature without changing the Nd isotopic ratio. Therefore, the MDI parental melt could have derived from the same mantle source as the quartz diorite (DK1) samples, but the rock could have been subjected to hydrothermal alteration, causing an enrichment in crustal Sr (with higher 87 Sr/ 86 Sr) and the shift to the right on the Nd-Sr isotopic diagram (blue arrow in Figure 14).
The chemical composition of magma can be modified through assimilation of crustal materials during ascent and storage leading to the modification of selected trace element ratios diagnostic of arc and subduction environments. The microdiorite (MDI) dykes cross cutting the Sungun stock have high Ba/Nb (>28), which is a common feature of arc magmas (e.g., Reference [51]), and high La/Nb ratios (1.9-2.30) indicating crustal involvement [80]. Furthermore, prevalent low Ce/Pb (1.64-2.1), Rb/Ba, Nb/U, Nb/Y, and high La/Nb and La/Ta ratios also suggest lithospheric contributions. On primitive mantle-normalized multi-element diagrams, however, MDI dykes have similar signatures to those of the Sungun stock and other late dykes ( Figure 13). Menzies et al. [81] have argued that Zr/Ba values can be used to distinguish lithospheric (0.3-0.5) from asthenospheric sources (Zr/Ba 0.5-0.7). The microdirite dykes have low Nb/La (0.43-0.51) and high Ba/Nb (172-269) values, but their Zr/Ba ratios (0.03-0.04) suggest generation from the lithospheric sources. Therefore, these observed differences may instead result from heterogeneous chemical composition in the source region, or changes in melting degree or post-crystallization alteration, rather than the direct crustal involvement.
Other chemical parameters can be used to assess the levels of crustal contamination. For example, basaltic rocks have been affected by crustal contamination have ratios La/Ta > 22 and La/Nb > 1.5 [82]. These ratios in the studied microdioritic dykes are 34-41 and 1.90-2.30, respectively, suggesting the influence of crustal contamination.

Conclusions
The mineralized porphyry and post-mineralization dykes at Sungun have a composition of quartz-monzonite, quartz diorite, gabbrodiorite, microdiorite, granodiorite, and lamprophyre. Generally, the post-mineralization dykes of the Sungun copper mine are subdivided into eight groups based on relative time of intrusion and into six groups according to their composition. The dykes trend sub-parallel to joints within the mine area, indicating intrusion and emplacement of dykes within previous fractures and crustal weaknesses.
Our data allowed us to relate all the igneous rocks occurring within the Sungun Cu-Mo mine area. The Sungun quartz-monzonite can be related through fractional crystallization processes to the quartz diorite and diorite dykes as they all have analogous chemical characteristics, such as light rare earth elements enrichment and HFSE depletion, and have similar Sr-Nd isotopic ratios. We propose a scenario where these intrusions belong to a single magmatic system that evolved in the middle crust. The mineralized Sungun porphyry was the first to be emplaced in the upper crust, with associated fluids forming Cu-Mo mineralization. Further portions of the evolving magmatic system were subsequently injected as dykes, becoming progressively more mafic with time due to the tapping of a zoned magma chamber.
Microdiorite and lamprophyric dykes are also related to the sample geodynamic environment as they share similar subduction signatures, such as depletion of Nb, Ta, and Ti. These dykes were intruded at a later stage. The microdiorite magma had a greater influence of crustal contamination compared to the other Sungun suite. The lamprophyre magma was likely derived from residual low-degree melts in equilibrium with mantle phlogopite, following the main subduction-related pulse.