Bruno Messiga

An approach to the dynamics of clay firing

Abstract Results of a detailed study on microtextural and microchemical characteristics of pottery obtained during different experimental firing runs carried out at different temperature under oxidising conditions are reported. During firing, the reaction behaviour of temper grains and clay matrix is dominated by disequilibrium conditions and characterised by the presence of different reacting sub-systems. Each one of them is compositionally well established and corresponds to different mineral phases in mutual contact. Reactions occur following the two main mechanisms: (i) reactions leading to nucleation and growth of new mineral phases (modal reactions) and (ii) reactions producing only compositional variations of phases (cryptic reactions). Modal reactions are given by new phases rimming the temper grains, whereas the cryptic reactions may develop during the partial transformation of a single phase, producing compositional zonings. Kinetic aspects are also considered in order to asses the role exerted by temperature, by bulk and mineral compositions and by fluid phase on the attainment of modal or cryptic reactions during firing.

Rare earth element redistribution during high-pressure–low-temperature metamorphism in ophiolitic Fe-gabbros (Liguria, northwestern Italy): Implications for light REE mobility in subduction zones

To unravel the rare earth element (REE) redistribution during high-pressure–low-temperature metamorphism, we have analyzed by ion microprobe all the minerals from representative Fe-gabbros from Ligurian metaophiolites (northwestern Italy). Contrary to what is observed for fresh Fe-gabbros, the clinopyroxene contribution to the whole-rock REE inventory of blueschists and eclogitized Fe-gabbros is minor or negligible. In both blueschists and eclogites, REE are redistributed among newly formed minerals. In blueschists, lawsonite is the major REE carrier and concentrates almost all the light REE (LREE) in the rock, whereas titanite plays an important role for middle and heavy REE (MREE, HREE). In eclogites, LREE and HREE mainly reside in accessory allanite and garnet, respectively. The development of blueschist and eclogite facies mineral assemblages in mafic oceanic crust, due to subduction to 35–65 km depths, was not accompanied by release of significant amounts of LREE to the overlying mantle wedge. Owing to the stability of lawsonite and epidote at ultra-high-pressure conditions, subduction of mafic rocks is considered an effective mechanism to bring LREE to great depth within the mantle.

Tectonic significance of Alpine eclogites

Abstract A review of P-T peaks and paths of eo- and meso-Alpine eclogite fades rocks occurring along the axial part of the Alpine chain shows that rocks re-equilibrated under high- and low- T (group-B and -C eclogites), are, respectively, hosted within a lower and an upper tectonic level of the Penninic nappe system. If P-T estimates for eclogites are considered peak conditions the two crustal portions, otherwise undistinguishable, were sutured during the collision of the European and Adriatic continental plates, which corresponds to the latest tectonic mechanism of eclogitization. Before collision, formation and preservation of eclogitic rocks up to shallow levels was assisted by subduction of the cold oceanic crust. The two lithospheric processes of oceanic subduction and continental collision, though separated in time, contribute to continuous generation of eclogites under thermal conditions that evolve from higher to lower P-T ratios from the end of ocean consumption. Exhumation trajectories are characterized by low- or high thermal regimes in the same structural domain in different parts of the chain (Western and Eastern Austroalpine), in the same part of the chain (Penninic and ophiolites in Western, Central and Eastern Alps), or even within the same nappe (Dora-Maira, Gran Paradiso and Adula). Late orogenic collapse or slab breakoff processes may have caused late heating at very low pressure (0.3 GPa) during exhumation in some units of the Pennine nappes and ophiolites Mechanisms of nappe emplacement are demonstrably multiphase and inferences on palaeogeographic derivation of eclogitic units can be drawn from interpretation of P-T trajectories.

The Zabargad peridotite-pyroxenite association: petrological constraints on its evolution

Abstract Petrographic, textural and chemical investigations on the Zabargad Island ultramafic association of peridotites and pyroxenite layers allow the main stages of their evolution to be identified. The whole evolution was characterized by three main stages, attributed to different geodynamic situations. These three stages are: 1. (1) early upper mantle evolution and equilibration 2. (2) decompressional upwelling during continental lithosphere rifting 3. (3) late near-seafloor emplacement. The first stage, developed under relatively high P - T conditions, was probably unrelated to the Red Sea evolution and was characterized by deep-seated magmatism. Pyroxenite layers were formed during this stage, consisting of Al-Di and Cr-Di types. According to the major and trace element distributions, the Al-Di pyroxenites can be considered as liquids produced under high-pressure conditions and equilibrated under spinel peridotite fades conditions. Cr-Di pyroxenites approach cumulate compositions produced by solid/liquid fractionation, starting from liquids formed at relatively high pressure. The early lherzolite-pyroxenite association was presumably cooled during its upward movement beneath a continental crust and was completely equilibrated in the spinel facies (at a temperature which was still higher than 1000°C). The second stage was initiated when the continental lithosphere rifting and thinning took place in this region, related to the propagating tensional effects from the southernmost focal upwelling of the Afar Triple Junction. The Zabargad mantle section underwent progressive, almost adiabatic, decompression during upwelling. At that time, H 2 O-rich fluids produced equilibrium crystallization of pargasitic amphiboles under sp-lherzolite facies conditions. The solidus temperatures of the Zabargad lherzolites were lowered, causing a sporadic incipient partial melting at relatively low pressure, with the production of depleted lherzolite residua and basaltic melts, crystallized “in situ” as gabbroic material. Utilizing the available REE data and the results of the modelling of a non-modal equilibrium partial melting process on a sp-lherzolite composition, it is evident that the more depleted Zabargad lherzolites closely approach compositions of refractory residua after low-degree (about 10%) partial melting at relatively low pressure on a model mantle source very similar to the less depleted Zabargad lherzolites. This stage was accompanied by plastic deformation, the main direction of which was about north-northwest, almost parallel to the future Red Sea axis. The progressive, relatively rapid, decompression produced incomplete equilibration under low-pressure conditions (pl-bearing assemblages). The third stage of near-seafloor emplacement was accompanied by increasing interaction with H 2 O-rich metasomatic fluids, which gave rise to widespread amphibolitization of the ultramafic and gabbroic rocks and, later on, to typical metasomatic mineralization (phlogopite, apatite, scapolite and sodic plagioclase). This final stage in the evolution was facilitated by the development of almost vertical shear zones which were variably oriented, broadly parallel both to the Red Sea axis and to the main transform faults detected in the area. Important fluid/rock interaction along fractures and metasomatic mineralization (serpentine minerals, gem-quality olivine, scapolite and cancrinite) characterize the last step of this third stage, preceding the shallow basalt dyke intrusion and the final tectonic emplacement on the seafloor.

Interplay between deformation and metamorphism during eclogitization of amphibolites in the Sesia–Lanzo Zone of the Western Alps

Interactions of fabric evolution and chemical parameters driving reaction progress during the amphibolite-to-eclogite transition were investigated in eclogitized amphibolites of the Western Alps. In the Sesia–Lanzo Zone (SLZ), mafic rocks ranging from eclogitized hornblendites to true eclogites occur as layers and boudins within micaschists and are characterized by different modal amounts of amphibole, omphacite, zoisite, garnet, and phengite, constituting the Alpine HP assemblage. Across narrow zones, this array of lithologies displays gradients in planar fabrics characterized by coronitic, S-tectonitic, and mylonitic textures with different extents of eclogitization. Eclogitic parageneses are controlled not only by the bulk rock composition of the protoliths but also by the degree of fabric evolution. This is the case for omphacite occurrence, which is constrained by plagioclase composition and modal amount (NK parameter) in the protoliths, whereas the increase in modal omphacite and the concomitant decrease in modal amphibole in rocks with high NK are controlled by the strain rate. Protoliths with a low NK content develop the amphibole + garnet + epidote assemblage in eclogitized hornblendites. In protoliths with higher NK values, the co-existence of amphibole with garnet, omphacite, and epidote occurs only for low-to-medium strain textures (e.g. coronites and S-tectonites), whereas an amphibole-free assemblage defines a mylonitic foliation; in this case, amphibole relics are present exclusively as armoured inclusions in garnets and omphacite porphyroclasts. Thus, amphibole persists in the eclogitic assemblage at pressures exceeding the experimentally determined amphibole stability field. Values of confining pressure under which Sesia–Lanzo mafic rocks re-equilibrated during Alpine subduction were estimated, through equilibrium assemblage modelling, at 2.2–2.7 GPa. The amphibole-bearing eclogites of the SLZ show that large volumes of amphibole-bearing rocks can be exhumed from a depth exceeding 75 km without dehydration reactions running to completion. Petrological estimates in orogenic zones may help constrain geological and tectonic conclusions when selection of laboratory samples is assisted even by simple microstructural evaluation of planar fabric development.

Amphibole evolution in Variscan eclogite-amphibolites from the Savona crystalline massif (Western Ligurian Alps, Italy): Controls on the decompressional P-T-t path

Abstract The Variscan metamorphic evolution of eclogite-amphibolites from the Savona Crystalline Massif is marked by the occurence of calcic amphiboles. Microtextural relations together with coupled electron microprobe and X-ray investigations on amphiboles pointed out a complex decompressional evolution. A prograde stage of the evolution in the eclogite facies (temperature conditions estimated as ca. 620°C) is recorded by compositional zoning and inclusions in garnet. The eclogitic event (pressure conditions 12kbar) produced assemblages of garnet, omphacite, Ca-amphibole, zoisite, quartz and rutile. Composition and zoning of matrix eclogitic amphiboles reveal that the subsequent evolution in the eclogite facies contemplates a late re-equilibration under lower conditions of temperature (ca. 540°C) and pressure. The eclogitic amphiboles are pargasitic to edenitic hornblendes, and are marked by rather high Na contents at the M4 site (up to 0.5 apfu). The forming of coarse-grained diopside-plagioclase symplectites and of amphibole-plagioclase coronas represent the early results of the destabilization of omphacites and garnets in the amphibolite facies. Pressure conditions of this stage have been estimated of ca. 10 kbar, while temperature conditions are poorly constrained. However, the increase in edenite (NaAlSi −1 ) at the outermost rims of the matrix eclogitic amphiboles and the very high values of edenite and tschermack (Al 2 Mg −1 Si −1 ) components in the coronitic amphiboles (subsilicic pargasites) suggest that temperature increase accompanied the early decompression to the amphibolite facies. Subsequent retrogression in the low-grade amphibolite facies resulted in the complete breakdown of omphacite, producing finer-grained symplectites. This stage is also recorded by the amphibole growth (magnesio-hornblendes to edenitic hornblendes with negligible Na amounts at the M4 site) at the expense of the clinopyroxene symplectites and as outward rims around the coronitic amphiboles. Frequently, amphibolites are completely recrystallized and do not retain any relic of high pressure assemblages. Temperature conditions for this late amphibolitic event have been estimated of ca. 510°C, while pressure conditions are inferred in the range 4–7 kbar. Such a complicated decompressional evolution is indicative of complex uplift tectonics after a subduction event.

Mafic and ultramafic pods with eclogitic relics from the Proterozoic Nagssugtoqidian mobile belt of East Greenland

Abstract The cores of amphibolitic and ultramafic pods within the basement complex of the Nagssugtoqidian mobile belt of East Greenland preserve metastable relics of older high-pressure assemblages. The bulkrock geochemistry of amphibolites indicates that their protoliths derive from low-pressure crystal fractionation (mainly controlled by plagioclase, pyroxene and olivine) of tholeiitic melts of probable MORB affinity. Amphibolites from the cores of the pods are corona-textured and provide microtextural evidence of a polyphase retrogression subsequent to a high-T eclogitic event, producing omphacite-garnet assemblages. Symplectitic intergrowths of Ca-clinopyroxene and plagioclase are the result of the unmixing of the older omphacite, whilst garnet is partly replaced by a fine-grained pseudomorph of orthopyroxene, anorthitic plagioclase and magnetite. This suggests subsequent partial re-equilibration in the intermediate-pressure granulite facies, producing clinopyroxene-orthopyroxene-plagioclase-magnetite assemblages. The final event in the amphibolite facies was controlled by the influx of dominantly hydrous fluids. According to kinetic constraints, the amphibolites developed coronitic or granoblastic textures. The coronas consist of plagioclase and hornblende between remaining garnet and unmixed clinopyroxene. Re-crystallized granoblastic amphibolites do not retain relics; according to the bulk rock chemistry, plagioclase and hornblende coexist with either garnet or clinopyroxene. Adjacent ultramafic rocks conform with this metamorphic evolution, being retrogressed from the spinel ±amphibole lherzolite facies to the tremolite-chlorite peridotite facies. Estimated equilibrium conditions for the final amphibolite facies event are T = 600 + 70°C and P = 5.2 ± 1 kbar. Mafic and ultramafic pods from the basement complex of the Nagssugtoqidian mobile belt of East Greenland probably represent fragments of an oceanic crust which was subjected to a subduction event of Proterozoic age.

Origin of prismatine from the Sondalo granulites (Central Alps, northern Italy)

A granulite from the Sondalo femic complex, Italian Central Alps, contains prismatine, the boron-rich member of the kornerupine group. This is the first report of prismatine in the Alps. The granulite consists of albite-rich plagioclase + cordierite + sillimanite + rutile + hercynite + corundum + quartz and is interpreted as a restite formed after partial melting of amphibolite-facies tourmaline-bearing metasediments at about 900°C and 0.8 GPa. Prismatine grains, which form a coarse aggregate with tourmaline, biotite and albitic plagioclase, have inclusions of sapphirine, hercynite and corundum. Secondary-ion mass spectrometry (SIMS) analyses on prismatine gives B 2 O 3 = 2.30–2.89 wt.%, Li 2 O = 0.067–0.125 wt.%, BeO = 0.005–0.007 wt. %, F = 0.32–0.49 wt. % and H 2 O = 0.90–1.02 wt.%. The cell parameters a and c and V of the Sondalo prismatine fit with the B 2 O 3 co-variation reported in literature. We propose a mechanism of prismatine formation involving the breakdown of tourmaline during the anatexis of the amphibolite-facies metasediments and the development of prismatine as a refractory phase. No intervention of a metaso-matic boron-rich fluid is required. The possible tourmaline breakdown process is the reaction 7.82 tourmaline + 1.03 biotite + 2.41 sapphirine + 3.22 quartz = 8.00 prismatine + 5.93 melt + 1 B 2 O 3 .

The Neolithic pottery of Abri Pendimoun (Castellar, France): a petro-archaeometric study

Abstract Middle and Late Neolithic ceramics from Abri Pendimoun (Castellar, France) and their geological raw materials have been investigated to characterize the ceramic bodies and to determine the possible provenance of raw materials. Petrographic, mineralogical and energy-dispersive spectrometry analyses were undertaken to define the compositional parameters of sherds and to clarify the relationship between Square Mouthed Pottery-phase I (VBQ I) and Chassey Culture. The ceramic bodies were generally made from glauconite-rich layers and terra rossa, unprocessed or mixed in variable proportions. Different kinds of temper, such as carbonates and/or aplite fragments, were added to the mixtures. Although most of the analysed ceramics were produced locally, a few mixtures show the addition of exogenous rocks. Although these ceramics could be interpreted as imported, we demonstrate that local clayey materials were used at Abri Pendimoun. The hypothesis that pottery was imported can therefore be ruled out. A small amount of crushed calcite (5%) was added to some glauconitic pellet mixtures. Pots made with this mixture are normally referred to the VBQ I. This combination of mixture and shape indicates that there was an important link between the VBQ I and Chassey Cultures.

Technological features of the ‘Cotto Variegato’: a petrological approach

Abstract The ‘Cotto Variegato’ are tiles used in Lombardy, between the XVII and XIX centuries as flooring for several historical buildings. Tiles are produced by the processing of two compositionally distinct clays. The main stylistic character of these tiles is a banded texture producing a veined aspect, in which white and red bands are also folded. The artefacts were hand crafted using two clayey raw materials of different composition, that are only partially mixed before the firing. The colour differences are produced during the firing. In all samples white and red portions are always composed of Ca-rich and Ca-poor clay, respectively. The multi-layered texture was obtained by a multiple folding and pressing process of the mixture. The interference of fold limbs with the tile surface gives the ‘variegato’ style to tiles. The paper explains how basic petrological knowledge can be applied to the study of ceramic artefacts in order to define provenance of the raw material, firing technology and how ancient craftsmen transformed the natural clayey materials into floor tiles. The methodological approach is that commonly applied to the study of the rocks and consists of textural analyses, at a different observational scale, combined with X-ray powder diffraction, X-ray fluorescence and microprobe analyses.