Christian Marignac

Evidence for Li-rich brines and early magmatic fluid-rock interactionin the Larderello geothermal system☆

Abstract The geochemical features of fluids accompanying the first stages of geothermal activity linked to magmatic intrusions have been documented for the Larderello geothermal system (Italy). Deep drilling has provided samples which preserve evidence of this early geothermal activity. Four wells (San Pompeo 2, Monteverdi 7, Sasso 22, and Serrazzano, VC 11) penetrated the deeper parts of the Larderello system, located in a metamorphic basement underlying the Tertiary nappe complex which constitutes the main aquifer at Larderello. The drill holes terminated close to the inferred roof of a granitic complex thought to be responsible for geothermal activity. Fluid inclusion data were obtained from recrystallized quartz lenses and quartz veins in samples displaying high temperature assemblages (plagioclase-actinolite-biotite-tourmaline; clinopyroxene ± andradite-wollastonite) and also from magmatic quartz in a leucogranite dike. The inclusions are mainly secondary in origin, oriented in fluid inclusion planes (FIP) related to hydrothermal circulation in the Larderello system. Several generations of high temperature fluids were trapped and include: 1. (1) H2OCO2 dominated vapors displaying variable but significant contents of CH4 and N2; 2. (2) aqueous vapors containing LiCl, with variable salinity; 3. (3) aqueous LiCl brine, often oversaturated with respect to halite at room temperature; 4. (4) complex brine, always oversaturated at room temperature with respect to two (halite and sylvite) or more (n ≤ 4) salts. The presence of LiCl was confirmed by identification of the salt hydrate (LiCl5H2O) at very low temperature using Raman spectroscopy. Bulk salinities could be roughly estimated at around 30 wt% eq. LiCl for the LiCl brine. Geometric and chronologic relationships between FIP reveal close relationships and mutual contamination between the H2OCO2 vapors and LiCl brine, indicating synchronism in their trapping. These fluids were generated and trapped at pressures of 100–130 MPa, nearly 23 MPa above the estimated present-day lithostatic pressure. This implies a denudation rate between 0.2 and 0.5 mrn· a−1 since the onset of hydrothermal activity, compatible with the setting of Larderello in a young (Tortonian) collision belt. Fluid inclusion trapping temperatures (425–650°C) show a monotonous increase towards the inferred granite, and are around 100–200°C higher than the highest present-day temperatures. The results are interpreted as recording the interaction between magmatic and contact metamorphic fluids in the early Larderello system. The H2OCO2 vapors resulted from the reheating of the basement metamorphic series (often C-rich) under relatively high temperatures during contact metamorphism. Lirich fluids expelled from an underlying Li-rich leucogranite migrated through the metamorphic series and the already cooled granite dikes and experienced local boiling. The fluid inclusion data demonstrate the involvement of magmatic fluids during the initial development of this high energy geothermal field.

40AR/39AR DATING OF SHEAR ZONES IN THE VARISCAN BASEMENT OF GREATER KABYLIA (ALGERIA). EVIDENCE OF AN EO-ALPINE EVENT AT 128 MA (HAUTERIVIAN-BARREMIAN BOUNDARY) : GEODYNAMIC CONSEQUENCES

A ductile shearing event is recognized in gneisses and barite veins of the crystalline internal massifs of the Alpine Maghrebides in Algeria. Dating of shearing movement is carried out by 40Ar/39Ar laser-probe experiments performed on synchronous neoformed micas. The proposed age at 128 Ma is constrained by a two-component mixing model for the saddle-shaped age spectra on phengites and by a plateau age at 128.3±0.3 Ma on muscovite. This Hauterivian–Barremian boundary age records a strong reworking of the Alboran plate (input of detritals in the flysch basin) and surrounding areas. It can be correlated with the J magnetic anomaly in the Atlantic and with the distensive events in the Pyrenees and external Alps. Therefore, isotopic ages (80–130 Ma) obtained on the Alpine–Mediterranean basement could reflect interferences between this event at 128 Ma and late-Alpine stages.

Boiling and fluid mixing in the chlorite zone of the Larderello geothermal system

Abstract The geochemical features of the geothermal fluids produced within the boiling zone in the relatively shallow parts of the Larderello geothermal system (Italy) have been documented as a result of deep drilling which provided samples from 1480 to 2500 m depth. Four wells (Monteverdi 1, Monteverdi 2A, Sasso 22 and Capannoli 2B) have been sampled in the intermediate parts of the Larderello aquifer located in a metamorphic basement underlying the Tertiary nappe complex which constitutes the shallow aquifer at Larderello. Fluid inclusions from recrystallized quartz lenses and quartz veins in samples displaying a predominant quartz–chlorite–(epidote–adularia) paragenesis have been studied by microthermometry and Raman spectroscopy. The inclusions are primary and pseudo-secondary in origin when formed in authigenic quartz, or of secondary origin, when located in fluid inclusion planes related to microfracturing of metamorphic quartz lenses. Several generations of fluids are present and include: H 2 O–(CO 2 )-dominated vapours and liquids, and a series of aqueous liquids, for the most part of relatively low salinity. The T m -ice of both early and late inclusions are mostly between 0.0 and −4.5°C, indicating that the salinities of the hydrothermal fluid were very low to moderate. However, rare fluid inclusions with lower T m -ice (from −4.9 to −25.0°C) were also observed. These inclusions may record the occasional input of saline fluids, which may be derived from the interaction of the hydrothermal waters with the evaporites present in the shallow part of the Larderello field. At Capannoli 2B, earliest H 2 O–(CO 2 ) liquids were trapped under minimal pressures of 610–645 bars, which bracket the estimated present-day lithostatic pressure (640 bars). In all other samples, the main stage of quartz–chlorite crystallization occurs under boiling conditions attested by the presence of liquid and vapour-rich inclusions, that, in some instances, can be texturally interpreted to be coeval. Their trapping conditions (350–375°C, 160–215 bars) are higher than the present day temperatures at the same depth. Later fluid inclusions attest to a significant cooling of the fluids down to temperatures similar to the present-day temperatures. During this time, pressures were close to hydrostatic conditions. Most fluid inclusions were trapped within the liquid field, this indicating that a significant pressure drop has since affected the main aquifers or fractured zones which are, at present, under vaporstatic conditions.

Remobilisation of base metals and gold by Variscan metamorphic fluids in the south Iberian pyrite belt: evidence from the Tharsis VMS deposit

The Tharsis massive sulphide deposit, one of the major VMS-type deposits in the Iberian pyrite belt (IPB) was severely deformed by the Variscan tectono-thermal events. The question of whether or not these events affected the metal distribution in the deposit has been addressed by simultaneously studying the mineral parageneses (Tharsis stockwork) and the fluid circulation (at local and regional scales). The results are: (1) The early paragenesis in the stockwork (Q1 quartz–pyrite–chlorite–phengiteFcobaltiteFankerite) was strongly overprinted by a late post-kinematic mineral deposition, including new quartz veins (Q3 quartz) and base metal sulphides (chalcopyrite, sphalerite, Bi and Te minerals, pyrite and galena) and gold. (2) At a regional scale, fluids accompanying the peak metamorphism conditions (ca. 300 MPa, ca. 300 jC) were of C–O–H– N–NaCl type, CO2-dominated with CH4 and N2, and are considered to be ‘‘metamorphic’’ on the basis of microthermometry and geochemistry. The late- to post-kinematic evolution (‘‘retrograde’’ stage) was characterised by a pressure drop, down to 40 MPa (lithostatic to hydrostatic transition), and a heat input leading to temperatures z430 jC, then decreasing to temperature around 170 jC. Fluids of the ‘‘retrograde’’ type exhibit both dilution of the C–O–H–N–NaCl fluid by a low salinity ‘‘meteoric’’ water and progressive loss of volatile components. (3) Fluids of the retrograde type pervasively percolated through the Tharsis stockwork and were responsible for the strong mineral overprint on the early (deformed) paragenesis. All the measurable fluid inclusions (f.i.) record these late fluids. There are primary fluid inclusions in Q1, but they are systematically imploded due to the external overpressure generated by the Variscan tectonic events. Although base metal distribution in the stockwork is basically the result of the

Découverte et signification d'une paragenèse à ilménite zincifère dans les métapélites des Jebilet centrales (Maroc)

A zincian ilmenite paragenesis is found in metapelites from a contact metamorphic zone (central Jebilet, Morocco) induced by the emplacement of microgranitic intrusions. The zincian ilmenite is mainly preserved in syntectonic andalusite porphyroblasts. The growth of zincian ilmenite is related either to sphalerite breakdown during prograde metamorphism, or to the pervasive flow of a mineralizing fluid within the metapelites. The chlorine-rich fluid carried zinc and other metals leached in the microgranites, during its flow to discharge zones which were probably the Jebilet sulfide deposits.

Tectonomagmatic Context of Sedex Pb–Zn and Polymetallic Ore Deposits of the Nappe Zone Northern Tunisia, and Comparisons with MVT Deposits in the Region

The Nefza region (Nappe Zone, northern Tunisia) is known both as a Late Miocene magmatic province and a base-metal district. In this region, small, post-nappe, continental extensional basins (Sidi Driss and Douahria) host syndiagenetic Pb–Zn ore deposits that have been classified as Sediment-Hosted Massive Sulphide (SHMS)–Sedimentary-exhalative (Sedex). In addition to this mineralization, the Nappe zone contains other Pb–Zn and polymetallic deposits that show similarities both to the SHMS-Sedex and MVT deposits of the Dome and Graben zones, namely the “peri-diapiric” deposits of Jebel (Jb) el Hamra, Jalta, and Jebel Ghozlane. All of the Nappe Zone deposits share common characteristics, including: (1) age—the deposits formed between late collisional events (Late Tortonian) and inception of the Early Messinian extensional regime, in the context of the Alpine Maghrebide belt formation; (2) presence of pre-existing sulphates—these were likely the main source of sulphur for the sulphides; and (3) hydrothermal systems leading to their genesis—these testify to alternating influx of cold and warm fluids. The involvement of high temperature fluids are deduced from fluid inclusion studies (Th values of 140–240 °C), whereas the presence of distinctive sphalerite textures (microspherules and colloform textures) and sulphur isotopic compositions demonstrate in situ bacterial sulphate reduction, and consequently deposition at temperatures below 80 °C. These data constitute unifying characteristics for these mineralizations that are in contrast to the features of Pb–Zn(–fluorite) MVT deposits occurring in the Dome and Graben zones of Tunisia. The geodynamic context prevailing during the Late Miocene in the Nappe Zone is the underlying common factor that explains the diversity of ore deposits formed at that time in this region. Most deposits are located in the vicinity of shear zones and associated lineaments inherited from the Variscan orogeny. These deep structures were reactivated by transtensional processes during the Alpine orogeny and controlled post-collisional magmatism, circulation of hydrothermal fluids, and locations of related ore deposits and showings in the Nappe Zone.

The Hoggar Gold and Rare Metals Metallogenic Province of the Pan-African Tuareg Shield (Central Sahara, South Algeria): An Early Cambrian Echo of the Late Ediacaran Murzukian Event?

The Hoggar massif, a part of the Tuareg shield in the Trans-Saharan Pan-African orogen, is endowed in gold and rare metals (Sn, W, Ta, Be). Most of these metals are present at low levels compared to other Precambrian or collisional belts worldwide, but tantalum is concentrated in a series of evolved granitic cupolas, making the Hoggar a promising tantalum province. The Tuareg shield was built in three stages during the convergence of two cratonic masses, the West African Craton (WAC) to the west and the East Saharan craton to the east. The first stage (730–630 Ma; Cryogenian) involved the accretion of island and continental arcs to several cratons. The second stage (630–580 Ma, Ediacaran) was a collisional event, involving (1) northward escape of the Tuareg terranes between the two main cratons, along N-S mega-shear zones with up to 1000 km of lateral displacement; (2) emplacement of large high-K calc-alkaline (HKCA) linear batholiths resulting from mantle-crust interaction through linear lithospheric delamination along the mega-shear zones; and (3) concomitant high temperature-low pressure metamorphism. As a result, the small Eburnean cratons included in the Tuareg shield (In Ouzzal; Laouini-Azrou-n-Fad-Tefedest-Egere-Aleksod, or LATEA) were more or less reworked (metacratonization). The third stage (575–540 Ma, late Ediacaran) was limited to the eastern Hoggar province, and involved the intracratonic collision of terranes within the margin of the East Saharan metacraton (Murzukian event, Fezaa et al. 2010). Immediately following the Murzukian event (540–520 Ma, terminal Ediacaran-early Cambrian) was simultaneous reactivation of late (transtensional) mega-shear zones, intrusion of high-level granite plutons related to rare-metal mineralization, and inception of crustal-scale hydrothermal systems and gold mineralization. The granite plutons comprise a series of A-type granites that evolved towards F-rich (topaz-bearing) alaskites, and true peraluminous, F–Na–Li-rich rare-metal granites (RMGs), with evidence of mixing between the two lineages. However, only the RMG suites are associated with Sn–W quartz veins, whereas Ta-rich cupolas may be found either in the Taourirt lineage (Tim Mersoi and highly fractionated Rechla cupolas) or in the RMG suites (Ebelekan)—the rare-metal enrichment being, in any case, of magmatic origin. Emplacement of these late granites was controlled either by the mega-shear zones, as in the Iskel island arc terrane, or by secondary shears of various orientation (N10° E, N50° E, N140° E) that dissect the terranes, as in the LATEA metacraton. The quartz-gold deposits are of the “orogenic gold” class and display contrasting relationships with the shear-zone systems, from close spatial associations with mega-shear zones as in the shear-zone-hosted In Ouzzal deposits (Tirek-Amesmessa) and the Iskel showings, to the very distal association with the Raghane shear zone as in the Tiririne deposits, through the connexion with secondary N10° E shear zones of the In Abeggui deposits in the LATEA. In all of the deposits, the hosting quartz veins were plastically deformed prior to gold deposition, which was uniformly a very late event occurring under brittle conditions, typically during extension. Both gold lodes and RMGs were emplaced into evolving stress-strain fields, involving a rotation of the shortening direction (σ1) from ca. N100° E to N30–40° E and finally close to N-S. In addition to rare-metal and gold mineralization, minor fracturing, magma emplacement, and hydrothermal activity are associated with these orientations in the entire Tuareg shield, far to the west of the locus of the Murzukian collision. We therefore conclude that the terminal Ediacaran-early Cambrian Hoggar gold and rare-metal province was controlled, at all scales, by the latest transtensional reactivation of Pan-African mega- and second-order shear zones, following and prolonging the late Ediacaran Murzukian event. Both magmatism and hydrothermal circulation were triggered by the heat flux associated with a renewal of linear lithospheric delamination processes that accompanied this reactivation. All of the known gold and rare metal deposits are hosted by metacratonic terranes with an indisputable Eburnean basement, whereas the juvenile terranes are either very poorly endowed (Iskel) or apparently barren. Although the connection with the gold mineralization seems likely, the reasons remain obscure. It may, however, be suggested that gold endowment is ultimately linked to an influx of mantle-derived CO2 and the formation of ultra-high temperature granulites in the Eburnean lower crust.

Geochemical Signature of Magmatic-Hydrothermal Fluids Exsolved from the Beauvoir Rare-Metal Granite (Massif Central, France): Insights from LA-ICPMS Analysis of Primary Fluid Inclusions

The Beauvoir granite (Massif Central, France) represents an exceptional case in the European Variscan belt of a peraluminous rare-metal granite crosscutting an early W stockwork. The latter was strongly overprinted by rare-metal magmatic-hydrothermal fluids derived from the Beauvoir granite, resulting in a massive topazification of the quartz-ferberite vein system. This work presents a complete study of primary fluid inclusions hosted in quartz and topaz from the Beauvoir granite and the metasomatized stockwork, in order to characterize the geochemical composition of the magmatic fluids exsolved during the crystallization of this evolved rare-metal peraluminous granite. Microthermometric and Raman spectrometry data show that the earliest fluid (L1) is of high temperature (500 to >600°C), high salinity (17–28 wt.% NaCl eq), and Li-rich ( 100 m) and interaction with external fluids.

Edough-Cap de Fer Polymetallic District, Northeast Algeria: II. Metallogenic Evolution of a Late Miocene Metamorphic Core Complex in the Alpine Maghrebide Belt

During the late Oligocene-early Miocene, three main hydrothermal events formed polymetallic deposits of the Edoug-Cap de Fer in the Edough massif of the Alpine Maghrebide belt. At ca. 17 Ma, the Karezas As (lollingite)-F (fluorite)-W (scheelite) deposit formed at a depth of ca. 2 km and temperatures of ca. 450–500 °C, from mixing between magmatic-hydrothermal hypersaline fluids issued from a concealed rare-metal granite and several metamorphic fluids derived from the metamorphic core complex. Slightly later, at ca. 16 Ma, the intrusion of microgranites produced high-enthalpy, liquid-dominated geothermal fields at the basement-Kabylian flysch boundary, with Numidian flysch acting as an impermeable lid and host for “mesothermal” polymetallic vein fields (Ain Barbar, Mellaha, Saf-Saf). Temperatures as high as ca. 350–375 °C were attained in the deep parts of the Ain Barbar field, at depths of ca. 1.3–1.5 km, accompanied by massive input of sodium that formed metasomatic plagioclase-rich hornfels (Chaiba domain); higher in the Cretaceous flysch aquifer, influx of hydrothermal fluids (300–270 °C) produced hydrothermal metamorphic assemblages of quartz-chlorite, calcite-chlorite, wairakite-chlorite, and epidote. The source of these hot fluids was a basement of the Edough type, in which heat advection was likely related to emplacement of a granite batholith at depth. Concomitant with the paleogeothermal circulations, fault activity created N170° E-trending fracture zones that progressively channeled fluid flow, with the development of propylitically altered linear zones and ore precipitation (Zn–Pb–Cu) at temperatures between 330 and 285 °C. At ca. 15 Ma, renewed magmatic activity (subvolcanic rhyolite dikes) was associated with the generation of new and shallow (ca. 800 m depth) geothermal fields, wherein convected surficial fluids (meteoric and possibly seawater) formed “epithermal” deposits including polymetallic quartz veins, quartz-stibnite metasomatic deposits in marble, and quartz-arsenopyrite-gold showings, at mostly lower temperatures of 300–250 °C. Excepting the Karezas skarn, for which a magmatic origin of the tungsten is likely, the metals deposited by the different hydrothermal systems were mainly sourced in rocks of the metamorphic core complex and its tectonically emplaced cover of Cretaceous flysch. Only a minor contribution of metals came from the magmatic rocks, as shown by lead isotope data for the Ain Barbar area. In particular, amphibolite of the Marble Complex in the Edough sequence may have been a major source of copper and the epithermal antimony (and gold?). The Edough-Cap de Fer district is directly linked to the evolution of the Edough metamorphic core complex. Although metallogenic activity began after the end of ductile deformation, metamorphic fluids derived from the core complex seem to have played a key role in the first stages of the hydrothermal circulation and related mineralization (Karezas W skarn, mesothermal polymetallic veins). However, the role of the late Miocene magmatism, induced by collisional processes through slab break-off and/or lithospheric delamination, was of equal importance in the genesis of the Edough-Cap de Fer metallic deposits, being the source of the heat advection responsible for hydrothermal convection during the meso- and epithermal mineralization. Finally, it appears that the transition from extension (related to opening of the Algerian-Provencal oceanic basin) to transpression (when the collision resumed), at the end of the Miocene, was the ultimate control on the mineralizing events in the Edough metamorphic core complex.

Edough-Cap de Fer Polymetallic District, Northeast Algeria: I. The Late Miocene Paleogeothermal System of Aïn Barbar and Its Cu–Zn–Pb Vein Mineralization

In northeast Algeria, the internal Edough massif of the Alpine Maghrebide belt, is an inlier of basement rocks under a cover of Cretaceous (Kabylian) and Cenozoic (Numidian) flysch nappes. During the late Oligocene-early Miocene, the Edough massif was an Oligo-Miocene metamorphic core complex involving the basement rocks (Pan-African gneiss, marble, amphibolite) and its Paleozoic cover. In a short time interval from latest Burdigalian to early Langhian (ca. 17–15 Ma), felsic intrusive rocks were emplaced in the basement and its tectonic cover under progressively shallower conditions (granite to rhyolite) that define the Edough-Cap de Fer magmatic district. At Ain Barbar, during intrusion of microgranites at ca. 16 Ma, a high-enthalpy, liquid-dominated geothermal system was active in the Cretaceous flysch reservoir, with Oligo-Miocene Numidian flysch serving as an impermeable cap. Temperatures as high as ca. 350–375 °C were attained in the deep parts of the Ain Barbar paleogeothermal field, at a depth of ca. 1.3–1.5 km. Input of massive amounts of sodium resulted in the formation of metasomatic plagioclase-rich hornfels (Chaiba domain), whereas higher in the Cretaceous flysch aquifer, invasion of hot fluids (300–270 °C) was associated with hydrothermal metamorphism (quartz-chlorite, calcite-chlorite, wairakite-chlorite, and epidote domains). The source of these hot fluids was a basement of the Edough type, in which advection of heat was likely related to emplacement of a granite batholith at depth. Concomitant with the paleogeothermal circulations, fault activity created N170° E fracture zones that progressively channelled fluid flow, with related development of linear propylitically altered zones and precipitation of Zn–Pb–Cu sulphides at temperatures between 330 and 285 °C. At ca. 15 Ma, renewed magmatic activity (subvolcanic rhyolite dikes) was associated with a new and shallower (ca. 800 m depth) geothermal system, involving the convective circulation of surficial fluids (meteoric and possibly seawater) at temperatures between 300 and 250 °C. Epithermal quartz and sulphides were deposited in the same vein systems as in the previous mineralization stage, but remained uneconomic. However, concomitant formation of massive adularia during alteration of the Chaiba rhyolite produced an economic K-feldspar body mined for ceramics.