Barren Miocene granitoids in the Central Andean metallogenic belt, Chile: Geochemistry and Nd-Hf and U-Pb isotope systematics.

Katja Deckart; Estanislao Godoy; Alfredo Bertens; Ayesha Saeed

Andean Geology 37 (1): 1-31. January, 2010
Andean Geology formerly Revista Geológica de Chile
www.scielo.cl/andgeol.htm

Barren Miocene granitoids in the Central Andean metallogenic belt, Chile: Geochemistry and Nd-Hf and U-Pb isotope systematics

Granitoides estériles del Mioceno en la franja metalogénica de los Andes Centrales, Chile: geoquímica e isotopía de Nd-Hf y U-Pb.

Katja Deckart¹, Estanislao Godoy², Alfredo Bertens³, Daniela Jerez⁴, Ayesha Saeed⁵

¹ Departamento de Geología, Universidad de Chile, Plaza Ercilla 803, Casilla 13518, Santiago, Chile. kdeckart@cec.uchile.cl
² V. Subercaseanx 4100, Pirque, Chile. egodoyster@gmail.com
³ Department of Geology, Roy al Holloway University of London, Egham, Surrey TW20 OEX, United Kingdom. a.bertens@gl.rhul.ac.uk
⁴ Rosario Sur 91, Oficina 803, Las Condes, Santiago, Chile. danielaJerez@hotmail.com ⁵ ARC National Key Centre for Geochemical Evolution and Metallogeny qfContinents (GEMOC), Department of Earth and Planetary Sciences, Macquarie University Sydney, NSW 2109, Australia. asaeed@els.mq.edu.au

ABSTRACT. Four Middle-to-Late Miocene barren plutonic complexes that occur between the giant porphyry copper deposits of the central Chilean Andes were selected for U-Pb LA-ICPMS geochronology and Hf-isotope systematics on single zircon grains. Major and trace elements and Sr-Nd-Hf isotope whole rock geochemical studies were under-taken to compare with slightly younger or coeval barren and fertile intrusive rocks between 32° and 34°S. The studied granitoids yield resolvable crystallization ages of 11.3±0.1 Ma (Cerro Mesón Alto massif), 10.3±0.2 Ma (La Gloria pluton), 14.9±0.2 Ma/14.9±0.1 Ma (Yerba Loca stock) and 11.2±0.1 Ma/14.7±0.1 Ma (San Francisco Batholith). Major and trace elements discard an adakitic signature as suggested for coeval porphyric intrusions at 32°S, slightly younger mineralized porphyries at Río Blanco-Los Bronces deposit and other Cenozoic adakites. Volcanic host rocks are less fractionated than the intrusive rock units. The same observation can be made for the unmineralized northern plutons compared to the southern ones. Initial Sr-Nd isotope data show insignificant variation (0.703761-0.704118 and 0.512758-0.512882), plotting in the mantle array. Trace element enrichment can be explained by addition of subducted-slab fluids and/or terrigenous sediments to the mantle wedge prior to and/or slight crustal input during magma ascent. Zircon grains separated from these barren intrusives share a similar initial εHf_i-data variation for the younger age group (10-12 Ma; 7.04-9.54) and show a more scattered range for the older one (14-15 Ma; 8.50-15.34); both sets plot between the DM and CLTUR evolution lines. There is evidence that magma evolution was slightly distinct through time from older to younger barren magmatism, compared to a few fertile porphyritic rocks from Río Blanco-Los Bronces porphyry copper deposit. It is suggested that chronological inconsistencies within these complexes might be related to differential shortening across the NE-SW-trending Yeso Valley transfer fault, assumed as coeval, which also explains the local lack of easterly magmatic are front migration.

Keywords: Magmatism, Miocene, Central Andes, Nd-Hf and U-Pb isotopes.

RESUMEN. Cuatro complejos plutónicos del Mioceno Temprano a Medio que se encuentran ubicados entre los depósitos de pórfidos gigantes de cobre de los Andes chileno centrales fueron seleccionados para un estudio de geocronología U-Pb por LA-ICPMS y de isótopos de Hf en granos individuales de circón. Estudios geoquímicos de elementos mayores y trazas y los isótopos de Sr-Nd-Hf en roca permiten compararlos con rocas intrusivas ligeramente más jóvenes o coetáneas estériles y fértiles entre 32° y 34°S. Los granitoides estudiados muestran edades de cristalización de 11,3±0,1 Ma (masivo Cerro Mesón Alto), 10,3±0,2 Ma (plutón La Gloria), 14,9±0,2 Ma/14,9±0,l Ma (Yerba Loca stock) y 11,2±0,1 Ma/14,7±0,l Ma (Batolito San Francisco). Elementos mayores y trazas excluyen una característica adakítica, como ha sido sugerido para los intrusivos porfídicos a los 32°S, los pórfidos mineralizados ligeramente más jóvenes de Río Blanco-Los Bronces y otras adakitas del Cenozoico. Las rocas volcánicas de caja son menos fraccionadas que las unidades de rocas intrusivas. La misma conclusión se obtiene con relación a los plutones de la zona norte comparados a los del sur. Valores isotópico iniciales Sr-Nd muestran una variación insignificante (0,703761-0,704118 y 0,512758-0,512882) en el campo del manto. El enriquecimiento de elementos traza puede ser explicado a través de una adición de fluidos de la placa subducida y/o por sedimentos terrígenos transportados hacia la cuña de manto y/o por un leve aporte cortical durante el ascenso del magma. Granos de circón separados de los intrusivos estériles del grupo más joven (10-12 Ma; 7,04-9,54) comparten valores iniciales de sHf, pero los intrusivos más antiguos (14-15 Ma; 8,50-15,34), en comparación, representan un rango más amplio; ambos rangos de valores se proyectan en la línea de evolución entre el DM y CHUR. Hay evidencia de que la evolución magmática fue ligeramente distinta a lo largo del tiempo desde un magmatismo joven y estéril hacia las rocas porfídicas mineralizadas de Río Blanco-Los Bronces. Se sugiere que la inconsistencia geocronológica entre los distintos complejos podría relacionarse con un acortamiento diferencial a través de una falla de transferencia NE-SO a lo largo del valle del río Yeso, asumida coetánea, lo cual explicaría la falta, a nivel local, de migración del frente del arco magmático hacia el este.

Palabras clave: Magmatismo, Mioceno, Andes centrales, Sistemas isotópicos de Nd-Hf y U-Pb.

1. Introduction

Miocene-Pliocene magmatism in the Andean Cordillera of central Chile between latitudes 32°S and 34°S has been of major interest since it hosts world-class porphyry Cu(-Mo) deposits. Studies of magma source based on the Sr and Nd isotopic systematics of mineralized plutonio complexes have revealed the involvement of magmas having component signatures characteristic of mantle. subduction, lower and/or upper crust. Several authors (e.g., Stern, 1991; Stern and Puig, 1991; Skewes et al., 1991; Skewes and Stern, 1994; Kay and Mpodozis, 2001) have discussed the migra-tion and petrogenetic evolution of the magmatism, suggesting that it was directly linked to the changing geometry of the subduction zone. The aim of these discussions was to develop a model for mineralization occurring in flat-slab areas of the Andean Cordillera. In contrast, the associated barren magmatism has generally not been considered, having been investigated separately by Kurtz et al. (1997), Kay et al. (1999) and Kay etal. (2005).

This paper adds geochronological, geochemical and isotopic data to the discussion of petrogenetic processes involved in the Miocene-Pliocene barren magmatism of the central Andean Cordillera. The new data set pretends to help and better understand the evolution of barren to fertile magmas related to subduction dynamics.

The geochronological and geochemical data presen-ted here focus on four selected barren intrusive units of Miocene age, from south to north: Cerro Mesón Alto massif (CMA, 33°40'S), La Gloria pluton (LGP 33°27'-35'S), Yerba Loca stock (YL,33°15'-18'S) and San Francisco Batholith (SFB, 33°10'S). They were emplaced between two supergiant Mio-Pliocene porphyry Cu (-Mo) deposits in the central Andes (Fig. 1). The first U-Pb and Hf-isotope LA-ICPMS data on single zircon grains and major and trace elements as well as Sr-Nd-Hf isotope data on whole rocks from these barren plutonic complexes presented here, allow to make progress on understanding the nature of the involved magma source(s) and their evolution in time. This new data set is compared with published analyses and interpretations for coeval, barren intrusions and slightly younger fertile porphyric rocks from the central Andean Cordillera.

Sample collection was undertaken in a limited región between 33° and 34°S, but the discussion addresses the entire Miocene-Pliocene metallogenic belt of the Main Andean Cordillera (Principal Cordillera; 32°-34°S) in central Chile.

2. Plutonism at 33°-34°S: current interpretation

Miocene-Pliocene central Andean plutonic rocks are located at the southern limit and immediately south of the flat-slab segment (ca. 29°-34°S), which separates the central volcanic zone from the southern volcanic zone. Thisca. 32^o-34^oS-70°00'-30'Wbelthoststhree major Miocene-Pliocene porphyry Cu(-Mo) deposits, from northto south, Los Pelambres, Río Blanco-Los Bronces and El Teniente (Fig. 1). Barren granitoids or smaller apophyses align with the three deposits. Mineral deposits along the Andean Cordillera may be linked to the convergence of the oceanic Nazca and South American continental plates (e.g., Sillitoe, 1974; Davidson and Mpodozis, 1991) and a causal relation between subduction and the nature (barren or fertile) of the associated magmatism has been proposedby e.g., Richards (2005).

In the study area, barren and mineralization-hosting granitoids of the San Francisco Batholith and the Teniente Plutonic Complex (cf. Kurtz et al., 1997; Kay et al., 2005), as well as smaller apophyses intrude either the Miocene Farellones Formation or the Eocene to earliest Miocene Abanico Formation. The generally assigned age of this plutonic magmatism is Middle Miocene (Table 1), but various unmineralized stocks lack precise geochronological information.

Of the studied plutonic units, the Cerro Mesón Alto massif (CMA), La Gloria pluton (LGP) and San Francisco Batholith (SFB) have been previ-ously dated (Table 1), but not the Yerba Loca stock (YL). A single K-Ar age on biotite of 9.8 Ma (R.E. Drake, unpublished) has been presented for the La Gloria pluton by Cornejo and Mahood (1997) and interpreted as a cooling age, possibly close to its crystallization age. The Cerro Mesón Alto granitoid, which represents the intrusive lower unit of the andesitic mountain peak, was dated by Kurtz et al. (1997), who named it after its outcrop south of 'Lago Yeso' rather than after its outcrop as a part of the Cerro Mesón Alto, and assigned it to the Teniente Plutonic Complex on the basis of a hornblende ⁴⁰Ar/³⁹Ar age of 12.4±2.5 Ma. Published U-Pb ages on the San Francisco Batholith show a range of 8 to 12 Ma (in Deckart et al., 2005); the younger zircon ages of 8-9 Ma were for samples from the Río Blanco-Los Bronces mine area and were interpreted as late, but still pre-mineralization (mineralization from ca. 6.3-4.3 Ma; Deckart et al., 2005), barren magmatic pulses of the San Francisco Batholith complex.

Isotopic data indícate a decrease of ⁸⁷Sr/⁸⁶Sr and an increase of sNd isotopic signatures for Pleistocene-to-Holocene volcanic centers from the northern end of the Southern Volcanic Zone (NSVZ) (33°S) to the south (SSVZ, >36°S), coincident with crustal thinning from north to south (e.g., Kay et al., 2005). Regarding El Teniente and Río Blanco-Los Bronces Cu(-Mo) deposit, isotopic signature are increasing in ⁸⁷Sr/⁸⁶Sr but decreasing in sNd ratios with decreasing ages of magmatic rocks (Kay et al., 2005; Stern and Skewes, 1995). However, for the flat-slab segment north of the NSVZ, isotope studies demonstrated an increase of radiogenic ⁸⁷Sr/⁸⁶Sr signatures and a decrease of sNd values in west-to-east and north-to-south directions, which coincide with the postulated younging of mineralization-hosted intrusive rocks inthe same directions (e.g., Stern et al., 1984; Stern and Puig, 1991; Stern and Skewes, 1995; Skewes and Stern, 1996). This agrees with a southward younging of the mineralized centers in the central Cordillera between Los Pelambres and the coeval Río Blanco-Los Bronces and El Teniente Cu(-Mo) deposits (Deckart et al., 2005).

The geochemical patterns of the barren plutons have been compared with the contemporaneous, mainly volcanic, Farellones Formation and were interpreted to be possibly cogenetic (cf. López-Escobar et al., 1979; Kay et al., 1995). Inthe more recent study by Kay et al. (2005) it is proposed that both magmatic trends might be linked to tectonic erosión, yet not to the inception of subduction of the Juan Fernández Ridge near 33°S at 10 Ma, as reported by Yáñez et al. (2001).

3. Zircon U-Pb ages and Hf-isotopes

Three unmineralized granitoids and one intrusive, which belongs to the San Francisco Batholith com-plex, hosting the Río Blanco-Los Bronces porphyry copper deposit, were dated by the U-Pb LA-ICPMS method on single zircon grains. A summary of intrusion-related published ages of the area is presented in table 1. Analytical procedures are described in Appendix 1. U-Pb age data are documented in tables 2a-d. Hf-isotopic data for single zircon grains are summarized in tables 3a-d.

The Cerro Mesón Alto (CMA) massif was sampled south of the 'Lago Yeso' dam, about 90 km southeast of Santiago (Fig. 1). This massif intrudes the Eocene to earliest Miocene volcaniclastic Abanico Formation with sharp contacts. Twenty-five spot analyses on different zircon grains yield a mean crystallization age of 11.29±0.10 Ma with a MSWD of 0.91 and a probability of 0.59 (FA-3; Figs. 2a, 2b; Table 2a). This age is slightly younger but within error of the one (12.4±2.5 Ma; ⁴⁰Ar/³⁹Ar) from Kurtz et al. (1997). These authors ascribed the low precision of the latter to the low potassium content and/or a slight excess of ⁴⁰Ar, which is common in amphiboles and can easily lead to ages older than the 'trae' crystallization age. Consequently, it is suggested that the U-Pb zircon age of 11.29±0.10 Ma obtained here is the best estimation of crystallization age of the Cerro Mesón Alto massif.

The La Gloria pluton (LGP) is located in the Andean foothills, 50 km east of Santiago. It also intrudes the volcaniclastic Abanico Formation, dated west of the pluton at 21.1 to 30.9 Ma (Ar/Ar on plagioclase, Fock, 2005) (FA-4; Fig. 1; Table 2b). The weighted mean age calculated from nineteen ablations on different zircon grains is 10.34±0.15 Ma with a MSWD of 0.46 and a probability of 0.97 (Figs. 2d, 2e). Combined with the above-mentioned 9.8 Ma K-Ar biotite cooling age this indicates a rapid cooling history for the La Gloria pluton.

The north-south orientated San Francisco Batholith was sampled at its southern border ca. 40 km north-northeast of Santiago. Published K-Ar and Ar/Ar ages of the San Francisco Batholith range between 18.5 Ma and 11.3 Ma (e.g., Warnaars et al., 1985; Deckart et al., 2005). A recently published zircon U-Pb IDTIMS age of 11.96±0.40 Ma (Deckart et al., 2005) was interpreted as the crystallization age of an unmineralized rock unit close to the Río Blanco mine area, a few kilometres northeast of the localities sampled for this study. Two new U-Pb LA-ICPMS ages on single zircon grains from two distinct sites were obtained here. The first sample was taken from a locality adjacent to the mine site Los Bronces and gave a crystallization age of 11.16±0.07 Ma (FA-7, Table 2c, Fig. 3a). The second was collected ca. 8 km to the southeast and yielded a crystallisation age of 14.74±0.13 Ma (FA-8, Table 2c, Figs. 3b).

The Yerba Loca (YL) stock is located ca. 30 km northeast of Santiago and close to the southeastern exposedlimitofthe San Francisco Batholith. Inthis as yet undated intrusive two U-Pb LA-ICPMS ages have been obtained. A total of eleven spots were used to calculate a crystallization age of 14.88±0.16 Ma with an MSWD of 0.45 for sample FA-9 and an age of 14.96±0.15 Ma (n=ll) withaMSWD of 0.58 for sample FA-11 (Table 2d, Figs. 3c-d). Sample FA-11 was collected ca. 20 m below and 100 m laterally from sample FA-9.

Thus the Cerro Mesón Alto massif has a slightly older age than the La Gloria pluton despite its location further east. On the other hand, only the younger age obtained from the San Francisco Batholith is coeval with the Cerro Mesón Alto massif U-Pb age. The Yerba Loca stock and most of the San Francisco Batholith are older than the Cerro Mesón Alto and La Gloria plutons. This pattern only partly confirms the postulated eastward younging of the magmatic are (e.g., Stern and Puig, 1991; Skewes and Stern, 1996). However, because westward-vergent back-folding is restricted to the southern slopes of the Yeso Valley, Godoy (2005) proposed a sinistral transfer fault along that valley, which could have been active during emplacement of the Cerro Mesón Alto massif (Fig. 1). The study area therefore shows a complex structural history with Late Miocene thrust faults trending mainly north-south, northeast-southwest and northwest-southeast, which might indicate differently displaced block structures in the metallogenic belt between 32-34°S.

Hf isotopes for zircon minerals from all four plutonic complexes were also analyzed. Initial isotope ratios were calculated using the U-Pb zircon crystallization ages obtained during this study. The Cerro Mesón Alto massif and La Gloria pluton zircon samples show initial εHf_i values ranging from +7.04 to +9.60 and from +7.13 to +8.93 for FA-3 and FA-4, respec-tively. The younger San Francisco Batholith zircon sample (FA-7) yields similar εHf_i values from +7.04 to +8.84. In contrast, the older zircon sample FA-8 ranges between εHf_i +8.50 and εHf_i +12.38, whichis slightly less radiogenic than the coeval zircon samples from the Yerba Loca stock [εHf_i = +9.20 to +14.49 (FA-9) and+11.39 to +15.34 (FA-11)].

Calculated Hf model ages (Hf-T_DM), which are interpreted in the same way as Nd model ages, range from 300 to 390 Ma for sample FA-3 from the Cerro Mesón Alto massif and between 320 and 380 Ma for sample FA-4 from the La Gloria pluton. The younger San Francisco Batholith zircon sample (FA-7) shows a similar T_DM age range from 320 to 390 Ma. The older rock sample from the San Francisco Batholith (FA-8) ranges in T_DM ages from 194 Ma to 340 Ma whereas the Yerba Loca stock zircons (FA-9, FA-11) indicate the widest T_DM age range between 76 Ma and 310 Ma, pointing to an inhomogeneous source of more juvenile mantle residuals. If a direct separation of these magmas from depleted asthenospheric mantle is assumed, initial epsilon Hf- values should be all greater than +10. Since the values range generally around +10, the magmas most probably were not derived directly from a depleted MORB-like mantle source.

Since there is no significant variation in the ^17SHf/¹⁷⁷Hf ratio data set for the samples from Cerro Mesón Alto, La Gloria pluton and the youngest San Francisco Batholith, magma mixing or inheritance do not seem to have been importan! The older samples (FA-8, -9, -11) however, do show a slightly greater isotopic variation, but no analyzed zircon grain shows evidence of older nuclei (Figs. 2-3) or reveals inherited U-Pb ages (Tables 2a-d).

4. Whole rock geochemical characteristics and Sr-Nd-Hf systematics

4.1. Major and trace element characteristics

Major and trace elements are presented in table 4 and figures 4-9. The Cerro Mesón Alto massif

(66-67 wt% SiO₂), La Gloria pluton (61.6-67 wt% SiO₂). Yerba Loca stock (54-57.5 wt% SiO₂) and San Fancisco Batholith (56.3-62.4 wt% SiO₂) late Miocene plutonic rocks show a calc-alkaline to high-K calc-alkaline and metaluminous character. All analyzed rocks fall in the quartz monzodiorite and quartz monzogabbro fields after Streckeisen (1976). Low MgO (<2.5 wt%), Ni and Cr concentrations from the Cerro Mesón Alto massif and La Gloria pluton are indicative of their fractionated character. The San Francisco Batholith and Yerba Loca igneous rocks show higher MgO (>2.5 wt%), Ni and Cr contents compared to the others. There is an overall increase in transition metal element (Cu, Cr, Co, Ni, V, Zn, Se) concentrations in the more mafic lithologies and at decreasing distance to the mineral district Río Blanco-Los Bronces. FA-8 from the San Francisco Batholith has the highest Cu, Mo. Se and Cr content.

A slightly negative-to-no-Eu anomaly (Fig. 4) and CaO/Al₂O₃ ratios correlated with Se, suggest clinopyroxene fractionation in all samples. Rare earth element (REE) plots show a 60- to 100-fold enrichment in light REE, around 20-fold enrichment for middle REE and a 5- to 7-fold enrichment for heavy REE relative to chondritic values (Fig. 4a). The slight enrichment of the heavy REE relative to chondrite suggests that amphibole and possibly garnet might be minor residual pilases in the magma source. (La/Yb)_n ratios range from 11.95 to 15.34 in the Cerro Mesón Alto massif, 8.96 to 9.33 in the La Gloria pluton, 6.73 to 7.76 in the Yerba Loca stock, and 8.07 to 9.16 inthe San Francisco Batholith. Cerro Mesón Alto and La Gloria have a higher potassic and SiO₂ content indicating a higher degree of fractionation. Similar major and rare earth element observations have been outlined in Kay et al. (2005) on whole rock geochemical analyses of contemporaneous rock units.

Trace element patterns are typical of continental margin subduction-related are magmas. The samples are enriched (2-38 times) in large ion lithophile elements (LILE; e.g., Rb, Ba, Th, U, Ta, K, La, Ce) relative to island are basalts (IAB; Ewart et al., 1998; Fig. 4b). High field strength elements (HFSE), Ti, Sr, and the heavy REEs show similar and/or slight enrichment or depletion relative to IAB. The strong negative anomaly for Zr and likewise for Hf (not shown) and Pb (Cerro Mesón Alto massif, La Gloria pluton) may be due to incomplete analytical acid digestión of zircon minerals. Differences between IAB and the active continental margin basalts are seen in element concentrations such as K, Sr, Rb, Ba, Zr, Th and U, which are more enriched in the latter.

The same rock concentrations normalized to average N-MORB show an even stronger enrichment in more or less the same elements (Rb, Ba, Sr, Th, U, K, Nd, La, Ce and Pb) compared to the IAB normalization and confirm a LILE transfer to the magma source area. Enrichment of these elements is usually explained through partial melts in the mantle wedge or fluids liberated from the subducted slab.

In order to compare and enlarge the major and trace element data set for the Middle-to-Late Miocene granitoids of the Central Andes, published information was included. Geochemical data of four Coya Machalí (CM) and three Farellones Formation (FF) volcanic rocks from Kay et al. (2005) were included since it has been suggested that coeval igneous rocks originated from the same magma source (e.g., López-Escobare et al., 1979; Nyström et al., 2003). A sample from the La Obra Plutonic Complex (LO; ca. 20 Ma), three (Lago Yeso, Melosas, San Gabriel) from the older Teniente Plutonic Complex (12-8 Ma; TPC-1; Kay et al., 2005) and three (Cerro Cortaderal, Carlota, Extravío) from the younger Teniente Plutonic Complex (7-3 Ma; TPC-2; Kay et al., 2005) alongwithfive samples from mineralized porphyritic intrusive units at Río Blanco-Los Bronces Cu(-Mo) deposit (RB-LB; Stern and Skewes, 1995); age range of mineralized porphyries (6.3-4.8 Ma; Warnaars et al., 1985; Deckart et al., 2005) were also included.

Harker diagrams of oxides versus SiO₂ content (Fig. 5a-h) show a negative correlation of Al₂O₃. MgO, CaO, FeO*, TiO₂ and P₂O₅ with increasing SiO₂ concentrations and an overall positive correlation of Na₂0 and K₂0 versus SiO₂ for the barren magmatism. Major element patterns are similar for the fertile rock sample group from the Río Blanco-Los Bronces deposit except for Na₂0 indicating a negative correlation with an increasing SiO₂content. Barren andesites from the Coya Machalí and Farellones formations are characterizedby the lowest SiO₂ content (50-61 wt%), followed by the barren igneous groups (58-71 wt%) without any tendency within the rock groups. The Río Blanco-Los Bronces porphyric rock group shows the highest SiO₂ content ranging from 66 to 71 wt%.

Trace elements (Ni, Se, Cr) versus SiO₂ content (Fig. 6a-e) show a negative correlation, whereas Ba indicates a positive pattern and Sr yields little correlation for the andesitic and barren plutonic rock groups. Porphyric Río Blanco-Los Bronces rocks (LOI: 0.85-2.0 wt%) show a positive correlation for Ba, a negative for Sr and rather no correlation for Se with increasing SiO₂ content. (La/Eu)_n ratios ver sus La_n concentrations (normalization after Sun and McDonough, 1989) rules out a break between volcanic and plutonic rocks, with ratios under 3 for the former and greater than 3 for the latter at a similar La_n concentration (Fig. 7a). Río Blanco-Los Bronces porphyric rocks show the highest (La/Yb)_nratios (14-28), implying either a strongly fractionated character or possible residual garnet in the source of this group (Fig. 4). These major and trace elements show that the volcanics are less differentiated than the coeval to younger barren and fertile porphyric rocks, which requires differences in magma source regions and/or differences in the entire tectonomagmatic framework at the time of their magma formation.

The trace elements ratios Ba/Th versus Ta/Yb (Pearce et al., 2005; Fig. 8a) clearly separate the herein selected volcanic from the plutonic barren groups. The higher Ba/Th ratio of the volcanic group (low Ta/Yb) may be linked to a low-temperature fluid component, whereas all barren plutonic groups have a higher Ta/Yb with lower Ba/Th ratios indicating an enriched mantle influence (N-MORB: Ta/Yb=0.04; Ba/Th=52.5; Sun and McDonough, 1989) and/or recy cled terrigeneous sediments rather than a shallow subduction (low-T, Ba-bearing, no significant Th addition) component (Pearce et al., 2005). In contrast, Kay et al. (2005) demonstrated that the Teniente Volcanic Complex group has similar characteristics withplutons of the immediate vicinity of El Teniente area. In the trace element ratio diagram (La/Yb versus Ba/La; Fig. 8b) the high Ba/La and low La/Yb ratios suggest rather dominating subduction fluids (Le., La Obra and La Gloria plutons) during genesis (e.g., Tollstrup and Gilí, 2005; Elliott, 2003), whereas a low Ba/La with increasing La/Yb ratios (Cerro Mesón Alto massif, San Francisco Batholith, and Yerba Loca stock, and 12-8 Ma Teniente Plutonic Complex [TPC-1]) might indícate a slightly increasing terrigenous sediment involvement during genesis. Río Blanco-Los Bronces porphyric intrusives show atendeney to both, slightly elevated La/Yb and Ba/La ratios pointing to a more complex magma genesis.

(Sr/P)_n ratios ranging from 3.63 to 4.46 for the Cerro Mesón Alto massif and La Gloria pluton and from 2.11 to 3.83 forthe San Francisco Batholith and Yerba Loca stock, suggest enrichment through a fluid component rather than sediment influence (Borg et al., 1997). Pelagic and ODP Leg 141 sediments (Kilian and Behrmann, 2003) yield (Sr/P)_n ratios of 1.65 and 1.43, respectively. Onthe otherhand, the higher Pb/La ratio (about 1-2) in the Yerba Loca stock could suggest a possible sediment influence during genesis or slight crustal contamination by an elevated Pb signature in the older magmatic group. Trace element ratios of La/Ta>25 (Hickey et al., 1986) indicate a typical arc-like signature for all available data from the here presented rock units. Furthermore, ratios such as Th/Yb and Ta/ Yb usually indicate the involvement of an enriched mantle source with a slight subduction component and/or slight crustal contamination during the genesis of these magmas, but do not discriminate unambiguously betweenthem (Pearce, 1983).

A slightly distinct evolution regarding the younger and slightly older plutonic rock groups may however be suggested considering the trace element ratios reported in this paper. Fluid dominance related to subduction processes are recognized in Cerro Mesón Alto and La Gloria pluton whereas sediment influence is manifested in magma evolution in the Yerba Loca stock and possibly the San Francisco Batholith.

4.2. Isotopic characteristics

Whole rock Sr, Nd and Hf isotope ratios were measured in two samples from the Cerro Mesón Alto massif, three from La Gloria pluton, two from the Yerba Loca stock and two from the San Francisco Batholith (Tables 3-5). Initial ratios were calculated back to the individual intrusive ages ranging between 15-10 Ma. Since the numberof samples involved is rather small, isotope modelling was not attempted. Instead, isotopic ratios are plotted in conventional diagrams or combined with trace element ratios (Figs. 9-11).

Initial Sr- and Nd-isotope ratios (0.703761-0.704118 and 0.512758-0.512882) show a restricted range for all four intrusions of the study area. The similarity in Hf-isotope ratios for whole rock and zircons, probably especially in the case of the younger intrusive rocks, indicates that zircon is most possibly controlling the Hf-isotope values.

Isotope diagrams show that initial Nd-Sr isotopic ratios of the new data plot in the mantle array between Depleted Mantle (DM) and Bulk Silicate Earth (BSE) (Fig. 9). The southern intrusives (Cerro Mesón Alto and La Gloria pluton) are slightly more radiogenic in their ⁸⁷Sr/⁸⁶Sr ratio than the northern group (San Francisco Batholith and Yerba Loca stock) at equivalent ¹⁴³Nd/¹⁴⁴Nd values. As a result of this characteristic the southern group plots inside many fields: the Chilean Coastal Batholith of central Chile (31°-34°S) (Parada et al., 1999), where its depleted portion overlaps with the SSVZ volcanic field (Kilian and Behrmann, 2003), Teniente Volcanic Complex (Kay et al., 2005) and in parts with the Farellones and Abanico formations (Nyström et al., 2003). Río Blanco-Los Bronces porphyry (Stern and Skewes, 1995) ⁸⁷Sr/⁸⁶Sr ratios are similar with the SSVZ volcanic field but ¹⁴³Nd/¹⁴⁴Nd_i values are less depleted compared to the rest of the rock groups of the depleted quadrant.

Ba/Nb versus Sr and Nd initial isotopic compositions confirmthe relative stability of the Sm-Nd(¹⁴³Nd/¹⁴⁴Nd)_i: 0.512758-0.512882) systemin a subduction environment and suggests a fluid dominated process (increased Ba/Nb) for Cerro Mesón Alto, La Gloria pluton and Río Blanco-Los Bronces porphyric intrusives compared to the San Francisco Batholith and Yerba Loca stock (Fig. 10a-b).

The slightly wider range of the Sr isotope ratios compared to Nd might be related to higher Rb mobility during subduction processes and the robustness of the Sm-Nd based isotopic system. In the initial ¹⁴³Nd/¹⁴⁴Nd versus Th/La diagram, the data (Fig. 10c) plot mainly at an average SSVZ volcanics isotopic ratio but at higher Th/La values. Exeptions are two samples, Cerro Mesón Alto and La Gloria pluton, which are characterized by significantly higher Th/La ratios. Whereas higher Th/ La ratios might account for recycling subarc mantle lithosphere or continental crust into the mantle source, lower Th/La ratios of the new data and characteristic for Río Blanco-Los Bronces porphyries indicate recycling of oceanic crust through the subduction zone into a mantle source (Kerrich et al., 1999).

Initial εHf_i-isotope data plotted against eNd ratios (Fig. 11a) separate one sample (FA-1) of the Cerro Mesón Alto massif from the rest of all plu-tons. The more radiogenic eNd value of the Cerro Mesón Alto massif (FA-1) is not accompanied by a higher εHf_i value. This result can be considerad to represent an outlier since major and trace elements do not indicate any unusual behaviour compared to the other Cerro Mesón Alto sample. Plotting ¹⁷⁶Hf/¹⁷⁷Hf_i versus ⁸⁷Sr/⁸⁶Sr_i isotopic data a split between the Cerro Mesón Alto and La Gloria pluton group (about 10-11 Ma) and the San Francisco Batholith and Yerba Loca stock rocks (about 14.8 Ma) becomes evident, with the first group being slightly more radiogenic in Sr isotopes (Fig. 11b). Variations in initial isotopic composition are still less in Nd and smallest in the case of Hf. This observation underlines the fact that Hf isotope systematics represents the most robust system in subduction weakly or not affected by dehydration. It possibly represents original or residual mantle source values. Figure 11c shows whole rock and zircon initial Hf-isotope ratios versus age indicating that all samples plot above the CHUR and below the DM evolution lines. This observation indicates a residual depleted mantle involved at an early stage of magma evolution. 'Mantle '-T_DM ages calculated for the Hf whole rock data from the Cerro Mesón Alto and La Gloria pluton granitoids, however, lie inthe future (Table 3a-b). Since ¹⁷⁶Lu/¹⁷⁷Hf ratios are high and samples are young, as confirmed by geochemical data, they most probably did not derive directly from a depleted mantle source. An estimated 'crustal' T_DM (two-stage model) with a ¹⁷⁶Lu/¹⁷⁷Hf ratio=0.015 (average continental crust; bulk silicate earth, Goodge and Vervoort, 2006) should give a more realistic age of separation of the host magma source from depleted mantle (DM) or crustal forming events. The widest whole rock crustal-T_DM age range is given by the Yerba Loca stock (510-590 Ma), whereas the San Francisco Batholith lies between 520 and 530 Ma. Both, the Cerro Mesón Alto massif and the La Gloria pluton values range between 550 and 570 Ma. Figure 11c shows that whole rock and zircon single grain initial ¹⁷⁶Hf/¹⁷⁷Hf isotopic data are principally overlapping except for the ca. 14.8 Ma zircon samples from the Yerba Loca stock and San Francisco Batholith (zircon: 110-540 Ma; WR: 510-590 Ma) with initial Hf-ratios slightly more radiogenic than the entire whole rock and zircon mineral data set.

5. Middle-to-Late Miocene source rock

In a subduction environment, element enrichment is often explained through either varying amounts of essentially lower crastal components, metasomatic fluid circulation from the subducting ocean píate. dehydration of subducted sediments, orpartial melting of sediments. Adakitic geochemical rock signatures (high Sr/Y (>40), La/Yb(n) subscript (>20) ratios; low high field strength element concentrations; depleted (<0.704) ⁸⁷Sr/⁸⁶Sr ratios (Defant and Drummond, 1990)) have been described as relatively common in the Neogene of the central Andes, but can be excluded for the granitoids herein presented.

The ages of all samples versus Sr/Y ratios (Fig. 12) indicate a slight-to-no tendency to increasing Sr/Y ratios with decreasing ages in the temporal interval from 20-5.5 Ma. It is notable that there is a higher Sr/Y ratio (88 and 113) detected for the two dacite porphyry samples (Stern and Skewes. 1995) from the Río Blanco-Los Bronces deposit as observed by several authors (e.g., Thieblemont et al., 1997) who suggested that adakitic magmas (Sr/Y >40) are involved in giant porphyry copper deposits. Furthermore, La Gloria pluton and San Gabriel granitoids are barren, yet plot on the limit or inside the adakite field.

An arc-like signature generally characterizes the Middle to Late Miocene rocks. Ba/Th, Ta/Yb and Th/Yb ratios indicate for all plutonic rocks an enriched mantle influence, which might be explained through a slight subduction component in the source area and /or crastal contamination during magma ascent. Higher Ba/La ratios show a tendency to subduction fluids as enrichment factor (La Gloria pluton and the older La Obra pluton), whereas higher La/Yb might suggest terrigeneous sediment influence (Cerro Mesón Alto). The re-maining samples (e.g., Río Blanco-Los Bronces porphyries) plot intermedíate to these trends and might indicate both processes for the enrichment of especially large ion lithophile elements (LILE).

The available isotopic compositions are compatible with two distinct mantle sources for two distinct rock and age groups: the 10-11 Ma group and the older, 14.8 Ma group (Fig. 9). All samples plot in the mantle array between N-MORB and Bulk Silicate Earth compositions. Whereas the first group shows slightly more enriched Sr-isotope characteristics, Hf isotopic signatures are relatively uniform in both data sets. A major difference is shown by ¹⁸⁷0s/¹⁸⁶0s_(i)isotopic data from magmatic magnetite minerals of each of the four plutons. Samples from the first group (CMA, LGP) are clearly characterized by continental crust input (average ¹⁸⁷Os/¹⁸⁸Os_(i)=0.33; εOs_(i)=+151 and +168 %; Table 5) whereas the second group (YL, SFB) indicates with its slightly positive εOs_(i) values (average ¹⁸⁷Os/¹⁸⁸Os_(i)=0.17; εOs_(i)=+28 and +37%; Table 5) that the magma source was enriched in Re, presumably being the result of ocean floor sediments recycled into the source area orby the way of crustal contamination during magma ascent. These isotope characteristics agree with the trace element ratios (e.g., Ta/Hf, Ta/ Yb, Th/Yb) characteristic of the intrusive bodies. as presented above. Increased Ba/Nb ratios might suggest preferential trace element enrichment by fluid processes orpelagic sediments or, whencombined with increasing initial ⁸⁷Sr/⁸⁶Sr ratios, to crustal input. Variation in Th/La ratios with relatively constant radiogenic initial Nd isotopic values imply slight crustal or ocean-floor sediment input for nearly all samples except one La Gloria pluton (FA-5) and one Cerro Mesón Alto (FA-1) sample. Their increased Th/La might reflect high temperature mobilization enrichment of Th compared to La.

Magmatic evolution of the Middle-to-Late Miocene intrusions may thus be explained either by a slightly enriched mantle source or by a two-stage model. The two-stage model projects the emplacement composition back to a hypothetical crustal formation/mantle depletion event, assuming a crustal ¹⁷⁶Lu/¹⁷⁷Hf of 0.015; and yields a minimum age for the hypothetical crustal reservoir.

Such model ages range between 470 and 630 Ma (Early Ordovician to Neoproterozoic) for sample FA-3 from the Cerro Mesón Alto massif and a shor-ter age range of 510 to 620 Ma (Early Cambrian to Neoproterozoic) for both FA-4 from the La Gloria pluton and FA-7 from the younger San Francisco Batholith sample. A more significant variation in ¹⁷⁶Hf/¹⁷⁷Hf ratios is seen in zircon samples from Yerba Loca and the older sample (14.7 Ma) from the San Francisco Batholith. Variable model crustal residence T_DM ages are recognised: 300-540 Ma (FA-8, San Francisco Batholith), 160 to 490 Ma (FA-9, Yerba Loca stock) and 110 to 360 Ma (FA-11, YL). This distinct isotopic data set for the two age groups (about 11 Ma and 14.8 Ma) suggests either a temporal change in a single magma source composition by different portions of contaminants, or the involvement of at least two magma sources or distinct differentiation processes.

Whole rock crustal Hf-T_DM ages from Neoproterozoic to Early Cambrian show the same crustal Hf -T_DM age range as zircon single grain ages for the Cerro Mesón Alto massif and La Gloria pluton, being slightly less pronounced for the San Francisco Batholith. The oldest San Francisco Batholith rock and Yerba Loca stock are exceptional in that they yield the same crustal Hf-T_DM ages for whole rock analyses but a range from 540 Ma to 110 Ma (Cambrian to Early Cretaceous) for zircon single grains. This rather identical Neoproterozoic to Early Cambrian crustal Hf-T_DM age for zircon and whole rock from the y ounger plutons suggests that zircon grains originated from the same magma as the zircon-bearing intrusives and. therefore, the modelled crustal residence age may be used as an estímate for a crustal-type magma separation from a residual depleted mantle source. Hf isotopes from the older zircon samples suggest contribution from a more juvenile mantle residual. These results do not agree with the 2 Ga Hf-model ages and their interpretation presented in Monteemos et al. (2008) on coeval (15-10 Ma) intrusions at 33°S. They described their obtained model ages as consistent with Sm-Nd model and upper intercept U-Pb ages from zircons of distinct Proterozoic and Phanerozoic lithologies (1.9-2.1 Ga) in the central Andes between 21°S and 32°S and 36° and 41°S (in Franz et al., 2006). Consequently, this inconsisteney outlines a distinctive petrogenetic evolutionforthe herein studied 33°S to 34°S plutons compared to the one proposed in Monteemos et al. (2008) using the much older lithologies reported by Franz et al. (2006).

Inheritance in zircon single grains in all intrusives bodies is not characteristic, as has been recognized in mineralized porphyric intrusions (6.3-5.0 Ma) of the Río Blanco-Los Bronces Cu (-Mo) deposit (Deckart et al., 2005), where zircon core ages range mainly between the Carboniferous and Mesoproterozoic. Furthermore, zircon grains from andesites of the volcanic Farellones Formation (ca. 17 Ma) cropping out in the Río Blanco mine sector yield one Neoproterozoic age for its upper discordia intercept (Deckart et al., 2005), whereas U-Pb SHRIMP dating of a Miocene ignimbrite in the El Teniente area indicated abundant Oligocene ages together with a few 114 Ma and 268 Ma ages (Godoy et al., in press). Zentilli et al. (2001) described as a common phenomenon xenocrystic zircons in older Cretaceous rocks of the Punta del Cobre Cu-Fe (Au) District and in the El Soldado Cu District of northern and central Chile, respectively. They suggested that during ascent felsic magmas associated with the mineralization picked up zircons from the Coastal Cordilleran granitic basement. However, zircon results in this study do not confirm this type of contamination process for our plutonic setting, which points to a different magma generation and emplacement process for the 15-10 Ma barren intrusions compared to the fertile plutonio younger and older pulses in the study area and further northeast in the Coastal Cordillera area. respectively. Instead, assimilation seems to nave taken place on amacroscopic scale. Mafic enclaves (mainly composed of hornblende, plagioclase and quartz minerals) are characteristic in both the Cerro Mesón Alto massif and La Gloria pluton. It is important to point out that these mafic enclaves are completely absent from mineralization-associated rocks in the Río Blanco mine area (e.g., San Francisco Batholith, Yerba Loca stock). The homogeneity and occurrence of the entire isotope range in the mantle array of the depleted quadrant supports, on the other hand. the suggestion that crustal components must have been of rather minor importance in the generation of these rocks.

Geochemical and isotope characteristics of sub-duction zone-related plutonic bodies in the central Andes between 33°and 34°S indicate an enriched subarc mantle source which was preferentially enriched in LIL and some HFS elements. It is suggested that element enrichment occurred either through dehydration fluids from subducted sediments which replenished the subarc mantle source (YL, SFB), or/and through partial melting or incorporation of a small amount of fragments of crustal material wi-thout homogenisation within the ascending magma (CMA, LPG). Finally, source enrichment processes do not appear to have been homogenous, and can be discarded as a process responsible for the separation of the younger age group (10-11 Ma), Cerro Mesón Alto massif, La Gloria pluton and the younger San Francisco Batholith from the older (14.8 Ma) San Francisco Batholith and Yerba Loca stock rocks.

6. Conclusions

Barren plutons occur between two supergiant porphyry Cu(-Mo) deposits in the Central Andes. Chile, between 33° and 34°S. Four of them (Cerro Mesón Alto massif, La Gloria pluton (3 3 °27' -3 5' S), Yerba Loca stock and San Francisco Batholith) have U-Pb LA-ICPMS emplacement ages ranging from 10.3 to 15 Ma. In the study area magmatism does not reflect the general eastward are front migration postulated by e.g., Kurtz et al. (1997) and Kay et al. (1999,2005). The Cerro Mesón Alto intrusive is slightly older than the La Gloria pluton but coeval with one age obtained from the more westerly located San Francisco Batholith. The magmatic are younging hypothesis is applicable on a regional scale (e.g., from the coast to the Principal Cordillera W to E transect) but not on a local scale (70°30'-69°45'). It, however, does not take into account either displacement by covered sinistral transfer faults, a complex tectonic control of plutonic emplacement or even younger intrusive events at about 1.1 Ma, as described for the Volcán valley by Godoy (1998) and Baeza (1999), mixed inthe general Miocene-Pliocene magmatism of the study area. The latter observation points to a stationary (12-1 Ma) magmatic are at these latitudes.

A quartz monzodioritic to quartz monzogabbroic composition and a metaluminous character charac-terize the plutonic complexes.

Rare earth element patterns suggest that amphi-bole and garnet were possibly minor components on the Late Miocene magma source in contrast with the Neogene adakites and the Río Blanco-Los Bronces porphyries. The (La/Yb)_n ratios indicate an only slightly higher fractionation degree for the southern plutons compared to the northern ones. Río Blanco-Los Bronces porphyries indicate with their elevated ratio most probably residual garnet in the source. The adakite -like geochemical features outlined for Late Miocene porphyritic intrusions at Los Pelambres located ca. 100 km north of the study area (Reich et al., 2003) could not be supported by the coeval barren intrusions analyzed herein. When the Middle-to-Late Miocene plutonic rocks studied here are compared with their volcanic Miocene Farellones Formation host rocks, a direct genetic relation, as mentioned for the Early-to-Middle Miocene intrusión and the Eocene-Early Miocene Abanico Formation (cf López-Escobare et al., 1979; Kurtz et al., 1997), could not be confirmed. Yet, anarc-like signature is suggested for all rocks, including volcanics and mineralized rocks. Trace element ratios for all plutonic rocks indicate an enriched mantle influence with possibly a slight subduction component and/or crustal contamination during magma generation.

Isotopic ratios for the plutons also indicate a slightly enriched mantle source with only small isotopic variations, which might be accounted for by a slightly heterogeneous source. Nd-Sr isotopic ratios show small variations, whereas Hf-Sr isotopic data separate the southern from the northern intrusions, with higher radiogenic Sr for the southern ones. Crustal Hf-T_DM modelling indicates a magma source that separated from the depleted mantle residual around 550-500 Ma and evolved until 11-10 Ma ago when magma re-melted and was emplaced at crustal levels as indicated by both whole rock (entire data set) and single zircon Hf-isotope (CMA, LGP, SFB) data. The older group (14.8 Ma; SFB, YL) is characterizedby similar Hf-isotopic whole rock data but shows a wide range of zircon Hf-data suggesting contributions from a more juvenile mantle residual. Magma ascent has been direct for partly coeval volcanics but shows slight LIL element enrichment for the entire middle to Late Miocene plutonic magmatism.

The data presented here supports a slightly dis-tinctive magma evolution through time from older to barren younger magmatism, different from the one recognised in the later, fertile one. Further studies on barren intrusives as well as a larger geochemical data set on the fertile magma are necessary. Some features that need more investigation and may relate to differences in the magmatic evolution of barren to mineralized intrusives are: the barren plutons contain mafic enclaves and show no zircon inheri-tance, mineralization-associated or -hosting intrusión show generally no mafic enclaves and no zircon inheritance, whereas the mineralized intrusives lack mafic enclaves and contain inherited zircon grains.

Acknowledgements

This study was financed through the Fundación Andes Project No. C-13960/32. Geologists M. Farías and A. Fock are thanked for assistance during field trips and J. Vargas for sample preparations in the facility of the Geology Department, University of Chile. We would like to thank N. Pearson and W. Griffin for supplying analytical work at GEMOC, Australia and F. Barra for the Os data on magnetite at the University of Arizona, USA. We appreciate the constructive comments of S.M. Kay and J.-R Liégeois on a prior versión of this manuscript. The authors acknowledge the critical review by S.M. Kay, R. Pankhurst and D. Morata. J. Le Roux is friendly thanked for having checked the English of this paper. This research used instrumentation funded by ARC LIEF and DEST Systemic Infrastructure Grants, Macquarie University and Industry. This is the contribution No. 618 from the ARC National Key Centre for Geochemical Evolution and Metallogeny of Continents. We acknowledge willingness from FONDECYT project 1095099 to include our results into its modelling.

References

Baeza, O. 1999. Análisis de litofacies, evolución depositacional y análisis estructural de la Formación Abanico en el area comprendida entre los ríos Yeso y Volcán, Región Metropolitana. Memoria de Título (Inédito), Universidad de Chile, Departamento de Geología: 119 p. [ Links ]

Black,L.P; Gulson,B.L. 1978. The age of the Mud Tank carbonatite, Strangways Range, Northern Territory. BMR Journal of Australian Geology and Geophysics 3: 227-232. [ Links ]

Blichert-Toft, J.; Albarède, F. 1997. The Lu-Hf geochemistry of chondrites and the evolution of the mantle-crust system. Earth and Planetary Science Letters 148: 243-258. Erratum. Earth and Planetary Science Letters 154(1998): 349 p. [ Links ]

Borg, LE.; Clynne, M.A.; Bullen, T. 1997. The variable role of slab-derived fluids in the generation of a suite of primitive calc-alkaline lavas from the southernmost Cascades, California. The Canadian Mineralogist 35: 425-452. [ Links ]

Creaser, R.A.; Papanastassiou, D.A.; Wasserburg, GJ. 1991. Negative thermal ion mass spectrometer of Os, Re and Ir. Geochimica et Cosmochimica, Acta 55: 397-401. [ Links ]

Cornejo, P; Mahood, G. 1997. Seeing past the effects of re-equilibration to reconstruct magmatic gradients in plutons: La Gloria Pluton, central Chilean Andes. Contribution to Mineralogy and Petrology 127: 159-175. [ Links ]

Davidson, L; Mpodozis, C. 1991. Regional geologic setting of epithermal gold deposits, Chile. Economic Geology 86: 1174-1186. [ Links ]

De Bievre, P; Taylor, PD.P 1993. Table of the iso-topic composition of the elements. International Journal of Mass Spectrometry and Ion Processes 123: 149 p. [ Links ]

Deckart, K.; Clark, A.H.; Aguilar, C; Vargas, R. 2005. Magmatic and hydrothermal chronology of the supergiant Río Blanco Porphyry Copper Deposit, Central Chile: Implications of an integrated U/Pb and ⁴⁰Ar/³⁹Ar database. Economic Geology 100 (5): 905-934. [ Links ]

Defant, M. J.; Drummond, M.S. 1990. Derivation of some modern are magmas by melting of young subducted lithosphere. Nature 347: 662-665. [ Links ]

Drummond, M.S.; Defant, M.J.; Kepezhinskas, PK. 1996. Petrogenesis of slab-derived trondhjemite-tonalite-dacite/adakite magmas. Transactions of the Royal Society of Edinburgh: Earth Sciences 87: 205-215. [ Links ]

Elliot, T. 2003. Tracers of the slab. In Inside the subduction factory (Eiler,J.; editor). American Geophysical Union Geophysical Monograph 138: 23-45. [ Links ]

Ewart, A.; Collerson, K.D.; Regelous, M.; Wendt, J.I.; Niu, Y. 1998. Geochemical evolution within the Tonga-Kermadec-Lau arc-backarc system: the role of varying mantle wedge composition in space and time. Journal of Petrology 39: 331-368. [ Links ]

Fock, A. 2005. Cronología y tectónica de la exhumación en el Neógeno de los Andes de Chile Central entre los 33° y los 34°S. Memoria de Título (Inédito), Universidad de Chile, Departamento de Geología: 179 p. [ Links ]

Franz, G.; Lucassen, E; Kramer, W.; Trumbull, R.B.; Romer, R.L.; Wilke, H.-G.; Viramonte, J.G.; Becchio, R.; Siebel, W. 2006. Crustal Evolution at the Central Andean Continental Margin: a Geochemical Record of Crustal Growth, Recycling and Destruction (Oncken, O.; Chong, G.; Franz, G.; Giese, R; Götze, H.-J.; Ramos, V; Strecker,M.R.; Wigger, R; editors). The Andes-Active Subduction Orogeny. Springer, Chapter 3: 45-64. [ Links ]

Godoy, E. 1998 Intrusivos sintectónicos entre los ríos Aconcagua y Cachapoal, Andes de Chile Central. In Congreso Latinoamericano de Geología, No .10, Actas 2: 149-154. Buenos Aires. [ Links ]

Godoy, E. 2005. Comentario sobre 'High magma oxidation state and bulk crustal shortening: key factors in the genesis of Andean porphyry copper deposits, Chile (31-34°S). Revista Geológica de Chile 32 (1): 155-158. [ Links ]

Godoy, E.; Schilling, M.; Solari, M.; Fock, A. En prensa. Geología del Área Rancagua-San Vicente de Tagua Tagua. Servicio Nacional de Geología y Minería, Carta Geológica de Chile, Serie Geología Básica, No. 118, 1 mapa escala 1:100.000. [ Links ]

Goodge, J.W.; Vervoort, J.D. 2006. Origin of Mesoproterozoic A-type granites in Laurentia: Hf isotope evidence. Earth and Planetary Science Letters 243: 711-731. [ Links ]

Griffin, W.L.; Pearson, N.J.; Belousova, E.; Jackson, S.E.; O'Reilly, S.Y.; van Achterberg, E.; Shee, S.R. 2000. The Hf isotope composition of cratonic mantle: LAM-MC-ICPMS analysis of zircon megacrysts in kimberlites. Geochimica and Cosmochimica, Acta 64: 133-147. [ Links ]

Hickey, R.L.; Frey, F.A.; Gerlach, D.C.; López-Escobar, L. 1986. Multiple source forbasaltic are rocks from the southern volcanic zone of the Andes (34-41°S): Trace element and isotopic evidence for the contributions from subducted oceanic crust, mantle and continental crust. Journal of Geophysical Research 91: 5963-5983. [ Links ]

Jackson, S.E.; Pearson, N.J.; Griffin, W.L.; Belousova, EA. 2004. The application of laser ablation-inductively coupled plasma-mass spectrometry to in situ U/Pb zircon geochronology. Chemical Geology 211:47-69. [ Links ]

Kay, S.M.; Mpodozis, C. 2001. Central Andean ore deposits linked to evolving shallow subduction systems and thickening crust. Geological Society of America (GSA) Today 11 (3): 4-9. [ Links ]

Kay, S.M.; Godoy, E.; Kurtz, A. 2005. Episodic are migration, crustal thickening, subduction erosión, and magmatism in the south-central Andes. The Geological Society of America Bulletin 117 (1-2): 67-88. [ Links ]

Kay, S.M.; Mpodozis, C; Coira, B. 1999. Neogene magmatism, Tectonism, and Mneral Deposits of the Central Andes (22° to 33°S Latitude). In Geology and ore deposits of the central Andes (Skinner, B.; editor). Society of Economic Geology, Special Publication 7: 27-59. [ Links ]

Kay, S.M.; Kurtz, A.; Godoy, E. 1995. Tertiary magmatic and tectonic framework of the El Teniente copper deposit, southern Chile (34°S-35°S). Geological Society of America, Abstracts with Programs 27 (6): p. 409. [ Links ]

Kerrich, R.; Wyman, D.; Hollings, P; Polat, A. 1999. Variability of Nb/U and Th/La in 3.0 to 2.7 Ga Superior Province ocean plateau basalts: implications for the timing of continental growth and lithosphere reeyeling. Earth and Planetary Science Letters 168 (1): 101-115. [ Links ]

Kilian, R.; Behrmann, J.H. 1997. Southern Chile trench sediments: Chemical and isotopic constraints on their source. In Congreso Geológico Chileno, No. 8, Actas 3: 1661-1665. Antofagasta. [ Links ]

Kilian, R.; Behrmann, J.H. 2003. Geochemical constraints on the source of Southern Chile trench sediments and their reeyeling in are magmas of the Southern Andes. Journal of the Geological Society 160: 57-70. [ Links ]

Kurtz, A.C; Kay, S.M.; Charrier, R.; Farrar, E. 1997. Geochronology of Miocene plutons and exhumation history of the El Teniente region, Central Chile (34-35°S). Revista Geológica de Chile 24 (1): 75-90. [ Links ]

López-Escobar, L.; Kilian, R.; Kempton, P.D.; Tagiri, M. 1993. Petrology and geochemistry of Quarternary rocks from the Southern Volcanic Zone of the Andes between 41 °30'and 46 °00'S, Chile. Revista Geológica de Chile 20(1): 33-56. [ Links ]

López-Escobar, L.; Frey, F.A.; Oyarzún, J. 1979. Geochemical characteristics of Central Chile (33°-34°S) granitoids. Contributions to Mineralogy and Petrology 113(3): 381-393. [ Links ]

Lucassen, F.; Trumbull, R.; Franz, G.; Creixell, C; Vásquez, P; Romer, R.L.; Figueroa, O. 2004. Distinguishing crustal recycling and juvenile additions at active continental margins: the Paleozoic to recent compositional evolution of the Chilean Pacific margin (36-41°S). Journal of South American Earth Sciences 17: 103-119. [ Links ]

Michard, A.;Gurriet, P; Soudant, M.; Albarede, F. 1985. Nd isotopes in French phanerozoic shales: external versus internal aspects of crustal evolution. Geochimica et Cosmochimca Acta 49: 601-610. [ Links ]

McCulloch, M.T.; Black, L.P 1984. Sm-Nd isotopic characteristics of Enderby Land granulites and evidence for the redistribution of Sm and Nd during metamorphism. Earth and Planetary Science Letters 71:46-58. [ Links ]

Monteemos, P; Scharer, U.; Vergara, M.; Aguirre, L. 2008. Lithospheric origin of Oligocene-Miocene magmatism in Central Andes: U-Pb ages and Sr-Pb-Hf isotope composition of minerals. Journal of Petrology 49 (3): 555-580. [ Links ]

Nagler, T.F.; Frei, R. 1997. Plug inplug osmium distillation. Schweizer Mineralogische und Petrographische Mitteilungen 77: 123-127. [ Links ]

Niu, Y; O'Hara, MJ. 2003. Origin of ocean island basalts: A new perspective from petrology, geochemistry and mineral physics considerations. Journal of Geophysical Research: Solid Earth 108 (B4): 2209. [ Links ]

Nowell, G.M.; Kempton, P.D.; Noble, S.R.; Fitton, J.G.; Saunders, A.D.; Mahoney, J.J.; Taylor, R.N. 1998. High precision Hf isotope measurements of MORB and OIB by thermal ionization mass spectrometry: Insights into the depleted mantle. Chemical Geology 149:211-233. [ Links ]

Nyström, J.O.; Vergara, M.; Morata, D.; Levi, B. 2003. Tertiary volcanism during extension in the Andean foothills of central Chile (33°15'-33°45'S). Geological Society of America Bulletin 115 (12): 1523-1537. [ Links ]

Othmann, B.D.; White, W.M.; Patchett, J. 1989. The geochemistry of marine sediments, island are magma genesis, and crust-mantle recycling. Earth and Planetary Science Letters 94: 1-21. [ Links ]

Pankhurst, R.J.; Weaver, S.D.; Hervé, F.; Larrondo, R 1999. Mesozoic-Cenozoic evolution of the North Patagonian Batholith in Aysen, southern Chile. Journal of the Geological Society of London 156: 673-694. [ Links ]

Parada, M.A.; Nyström, J.O.; Levi, B. 1999. Multiple source for the Coastal Batholith of central Chile (31-34°S): geochemical and Sr-Nd isotopic evidence and tectonic implications. Lithos 46: 505-521. [ Links ]

Pearce, J.A. 1983. Role of sub-continental lithosphere in magma genesis at active continental margins. In Continental Basalts and Mantle Xenoliths (Hawkesworth, C. J.;Norry,M. J.; editors). Shiva Publishing, Nantwich: 203-249. UK. [ Links ]

Pearce, JA.; Stern, R.J.; Bloomer, S.H.; Fryer, P 2005. Geochemical mapping of the Mariana arc-basin system: Implications forthe nature and distribution of subduction components. Geochemistry, Geophysics, Geosystems 6, Q07006, doi:10.1029/2004GC000895. [ Links ]

Reich, M.; Parada, M.A.; Palacios, C; Dietrich, A.; Schultz, F.; Lehmann, B. 2003. Adakite-like signature of Late Miocene intrusions at the Los Pelambres giant porphyry copper deposit in the Andes of central Chile: metallogenic implications. Mineralium Deposita 38: 876-885. [ Links ]

Richards, J.P 2005. Cumulative factors in the generation of giant calc-alkaline porphyry Cu deposits. In Super porphyry copper and gold deposits: A global perspective (Porter, T.M.; editor). Porter GeoConsultancy (PGC) Publishing, Linden Park 1: 7-25. South Australia. [ Links ]

Rivano, S.; Godoy, E.; Vergara, M.; Villarroel, R. 1990. Redefinición de la Formación Farellones en la Cordillera de los Andes de Chile Central (32-34°S). Revista Geológica de Chile 17 (2): 205-214. [ Links ]

Shirey, S.; Walker,R. 1995. Cariustube digestion for low-blank rhenium-osmiumanalysis. Analytical Chemistry 67 (13): 2136-2141. [ Links ]

Skewes, M.A.; Stern, C.R. 1994. Tectonic trigger forthe formation of late Miocene Cu rich megabreccias in the Andes of central Chile. Economic Geology 2: 551-554. [ Links ]

Skewes, M.A.; Stern, C.R. 1996. Late Miocene mineralized breccias in the Andes of central Chile: Sr- and Nd-isotopic evidences formultiple magmatic sources. In Andean copper deposits: new discoveries, mineralization, sty les andmetallogeny (Camus, F.; Sillitoe, R.; Petersen,R.; editors). Society of Economic Geologists, Special Publication 5: 33-41. [ Links ]

Skewes, M.A.; Stern, C.R.; Holmgren, C; Contreras, A.; Godoy, S.; Vela, I.; Rivano, S. 1991. Evolución magmática cerca del borde sur del segmento de bajo ángulo de subducción en Chile Central (32-34°S). In Congreso Geológico Chileno, No. 6, Extended Abstracts: 146-148. Viña del Mar. [ Links ]

Sillitoe, R.H. 1974. Tectonic segmentation of the Andes: Implications for magmatism and metallogeny. Nature 250: 542-545. [ Links ]

Stern, C.R. 1991. Role of subduction erosion in the generation of Andean magmas. Geology 19: 78-81. [ Links ]

Stern, C.R.; Futa, K.; Muehlenbachs, K. 1984. Isotope and trace element data for orogenic andesites from the austral Andes. In Andean Magmatism; chemical and isotopic constraints (Harmon, R. S.; Barreiro, B. A; editors). Shiva Publication: 31-46. U.K. [ Links ]

Stern, C.; Puig, A. 1991. Geochemical Evolution of Magmatic Rocks in the Vicinity of El Teniente Copper Deposit (34°S), Central Chilean Andes. In Congreso Geológico Chileno, No. 6, Extended Abstracts: 265-267. Viña del Mar. [ Links ]

Stern, C; Skewes, A. 1995. Miocene to the present magmatic evolution at the northern end of the Andean Southern Volcanic Zone, central Chile. Revista Geológica de Chile 22 (2): 261-272. [ Links ]

Streckeisen, A. 1976. To each plutonic rock its proper name. Earth Science Reviews 12: 1-33. [ Links ]

Sun, S.S.; McDonough, W.F. 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and process. In Magmatism in ocean basins (Saunders, A.D.; Norry, M.J.; editors). Geological Society of London, Special Publication 42: 313-345. [ Links ]

Thieblemont, D.; Stein, G.; Lescuyer, J.-L. 1997. Gisements epithermeaux et porphyriques: la connexion adakite. C.R. Academie des Sciences Comptes Rendus, serie 2: Sciences de la terre et des planetes, Earth and Planetary Sciences 325: 103-109. [ Links ]

Tollstrup, D.L.; Gill, J.B. 2005. Hafnium systematics of the Mariana are: Evidence for sediment melt and residual phases. Geological Society of America 33: 737-740. [ Links ]

Vergara, M.; Drake, R. 1978. Edades potasio-argón y su implicancia en la geología regional de Chile Central. Universidad de Chile, Departamento de Geología, Comunicaciones 23: 1-11. [ Links ]

Warnaars,F.W.; Holmgren, C; Barassi, S. 1985. Porphyry copper and tourmaline breccias at Los Bronces-Río Blanco, Chile. Economic Geology 80: 1544-1565. [ Links ]

Wiedenbeck, M.; Alié, P; Corfú, F.; Griffin, W.L.; Meier, M.; Oberli, E; von Quadt, A.; Roddick, J.C.; Spiegel, W. 1995. Three natural zircon standards for U-Th-Pb, Lu-Hf , trace element and REE analyses. Geo standard Newsletters 19: 1-23. [ Links ]

Wilson, S.A. 1997. The collection, preparation and testing of USGS reference material BCR-2, Columbia River, Basalt. U.S. Geological Survey Open-File Report 98-00x. [ Links ]

Yáñez, G.A.; Ranero, C.R.; von Huene, R.; Díaz, J. 2001. Magnetic anomaly interpretation across the southern central Andes (32°-34°S): The role of the Juan Fernández Ridge in the late Tertiary evolution of the margin. Journal of Geophysical Research 106 (B4): 6325-6345. [ Links ]

Zentilli, M.; Pop, N.; Heaman, L.; Boric, R. 2001. Evidence of Proterozoic crust under the Coastal Cordillera of Central Chile: Grenville age xenocrystic zircons in Cretaceous volcanic rocks. In South American Isotope Conference, CD-ROM 642. Chile. [ Links ]

Manuscript received: January 09, 2008; revised/accepted: June 01, 2009.

*APPENDK 1. ANALYTICAL METHODS

For single-zircon grain LA-ICP-MS U-Pb and Hf-isotope analysis, magmatic zircons were separated from 2 kg samples using routine crashing, grinding. Gemini Table, heavy liquid and Frantz magnetic separation methods in the Geology Department, University of Chile. Preliminary grinding was undertaken in the same facility for samples selected for geochemistry. All geochemical and isotope analyses were carried out in the geochemical facilities at Macquarie University, Sydney, Australia.

For U-Pb geochronology and Hf-isotope analyses, zircon mineral separates were mounted in epoxy in a 2.5 cm diameter sample-holder and polished. Backscatter electron (BSE)/cathodoluminescence (CL) images were taken of all analysed zircons to study the internal struc-ture of the grains prior to LA-ICP-MS analysis. Mineral ablation was carried out using a LUV213 laser ablation system. LA-ICP-MS analyses were done in an Agilent 4,500 high-sensitivity quadrupole ICPMS providing a sensitivity of more then 200 million counts per second (cps)/ppm for mono-isotopic heavy elements (atomic mass >85) in standard solution mode. The transporter gas of the ablated material was He. Spot diameters ranged between 50 and 80 um, ablation time was about 60 s and the ablated masses of zircon were approximately 500-1,500 ng. Detection limits for heavy elements applying laser ablation sampling are typically <10 ppb for a 60 s ablation at 40-50 µm resolution. Gas blank ²⁰⁸Pb signals of 30-40 cps were achieved routinely. The in-house GJ-1 zircon standard (608.5±0.4 Ma) was used for optimizing ICP-MS operating conditions. Reproducibility and accuracy were controlled by zircon standards 91500 (1,065.4±0.3 Ma; Wiedenbeck et al. ,1995) and Mud Tank (732±5 Ma; Black and Gulson, 1978). The GLITTER software program for LAM-ICPMS analyses undertakes data reduction and calculation of age estimates from isotopic ratios. Ages are quoted at 2σ error intervals.

Hf-isotope analyses were carried out in situ using a Merchantek EO LUV laser-ablation microprobe, attached to a Nu-Plasma multi-collector ICP-MS. Beam diameter of approximately 80 µm resulted in total Hf signals of 1-6x1^-11 Adepending onHfconcentrations, which ranged from 0.9 to 1.8% Hf. Ablation time was 80-120 s and pit depth 40-60 um; typically a total amount of 50 ng of Hf was consumed. For correction of the interference of ¹⁷⁶Yb on ¹⁷⁶Hf, a 1 ppm solution of JMC475 Hf spike with 80 ppb Yb was used with the aim to obtain the appropriate value of ¹⁷⁶Yb/¹⁷²Yb to calculate ¹⁷⁶Yb/¹⁷⁷Hf ratios. The recommended ¹⁷⁶Lu/¹⁷⁵Lu ratio of 0.02669 (De Bievre and Taylor, 1993) was used to calculate ¹⁷⁶Lu/¹⁷⁷Hf ratios. The 2 SE uncertainty on single analyses of ¹⁷⁶Lu/¹⁷⁷Hf ratios was <l-2%. Typical within run precision (2SE) on the ¹⁷⁶Hf/¹⁷⁷Hf ratio analyses were ±0.000020. Hf isotope analyses indicated mean values for ¹⁷⁶Hf/¹⁷⁷Hf ratios of JMC475 Hf standard identical to the recommended value of 0.28216 by Nowell et al. (1998). Chondritic values of Blichert-Toft and Albaréde (1997) were used for calculation ofepsilonHf values. Depleted-mantle source model ages are based on ¹⁷⁶Hf/¹⁷⁷Hf=0.279718 and ¹⁷⁶Lu/¹⁷⁷Hf=0.0384-values which produce ¹⁷⁶Hf/¹⁷⁷Hf=0.28325 similar to average MORB over 4.56 Ga.

Further detailed information on methodology, sample preparation and instrumentation are described by Jack-son et al. (2004) for U-Pb analyses and by Griffin et al. (2000) for Hf-isotope measurements.

Major and some trace elements (V, Cr, Co, Ni) were analyzed by X-ray fluorescence spectrometry using a Spectro XLAB2000 energy dispersive spectrometer in the Geochemical Analysis Unit, GEMOC, Macquarie University. Samples for major element analysis were prepared as glass discs using a lithium metaborate flux. Trace elements were determined on pressed powder pellets using a PVA binder. Replicate analyses of USGS reference materials (BCR-2, Columbia River Basalt; Wilson, 1997) give precision values of <1 wt% relative standard deviation (rsd) for all elements with concentrations >0.2 wt%, except for Na which has a precision of 2-3 wt% rsd. For trace elements typical precision is as follows: >500 ppm, 0.5-2.0 wt% rsd; 100-500 ppm, 1-3 wt% rsd; 10-100 ppm, 3-10 wt% rsd; 1-10 ppm, 12-25 wt% rsd; <1 ppm, 25-40 wt% rsd.

All other trace elements were analyzed by an Agilent 7500cs ICP-MS Octopole Reaction System using standard mode with no collision cell reaction gas. Argón was used as carrier gas. Other instrument parameters include a Pt shield on torch, quartz torch and spray chamber and a self aspirating PFA nebuliser. Standard deviation for replicated measurements on sample FA-3 lies generally between zero and 3% with one exception for Ba yielding a 7% standard deviation on its replica-te. The standard deviation between the measured and recommended standard values of the analysed (BCR-2) generally ranges from 0 to 4% with the exception of Sr and Ba yielding a 12 and 17%, respectively. Instrument drift was corrected by spiking samples and standards with internal standards as ⁶Li, As, Rh, In, Tm and Bi. The used digestión procedure employed 100 mg of rock powder in 15 mi Teflon Savillex beakers mixed with HF/HNO₃ and HC1O₄ acids over night. This procedure was repeated several times before finally mixed with 2%-HNO₃ and analysed on the ICP-MS.

For magmatic magnetite analysed 0.2-0.5 g of hand-picked minerals were loaded in Carius tubes (Shirley and Walker, 1995). Re and Os spikes and 16 mi of Aqua Regia were added and kept under cold conditions to avoid evaporation and early chemical reaction, then sealed. Afterwards, the tube was placed into an oven at 240°C overnight. Os was separated from the solution in a two-stage distillation process (Nagler and Frei, 1997), then purified and loaded with Ba(OH)₂ on a Pt-filament for analysis by a negative thermal ion mass spectrometer (NTIMS-VG54 mass spectrometer; Creaser et al., 1997). Re was separated after Os by using the remaining acid solution, After drying the solution, it was re-dissolved in 0.1 HNO₃. Re was extracted and purified by using AG1-X8 (100-200 mesh) resins. After-wards Re was loaded with Ba(SO)₄ on Pt-filament and analysed by NTIMS.

Whole rock isotopic analyses were obtained by using the multi-collector Nu Plasma Instrument ICP-MS. The chemical separation for Sr-Nd isotopes from the whole rock samples was obtained by using catión and anión exchange resin columns. An exponential correction for mass fractionation was applied using ¹⁴⁶Nd/¹⁴⁵Nd=2.07204 to avoid an iterative correction for the interference of ¹⁴⁴Sm on ¹⁴⁴Nd. The value for ¹⁴⁶Nd/¹⁴⁵Nd is an average of results obtained on a 1 ppm JMC321 Nd solution and is normalised to ¹⁴⁶Nd/¹⁴⁴Nd=0.7219. The interference of ¹⁴⁴Sm on¹⁴⁴Nd was correetedbymeasuring¹⁴⁷Sm and using ¹⁴⁴Sm/¹⁴⁷Sm=0.2070. Corrections for mass fractionation of Sr and Rb isotope ratios are made using an exponential law using a normalizing value for ⁸⁶Sr/⁸⁸Sr=0.1194. The precision (2 sd) and aecuracy of standard analyses generally average for Sr isotpes 0.000066 and for Nd 0.000040 a.