Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
We studied the deformation in a partially molten mantle through detailed petrostructural analysis of plagioclase lherzolites from the northwestern part of the central Lanzo massif. These peridotites display a pervasive anastomosed network of millimetre to centimetre-scale plagioclase-rich bands, which grades locally into a planar layering marked by an intercalation of plagioclase-rich and olivine-rich layers up to tens of centimetres wide. Both the anastomosed and the planar layering are subparallel to the deformation fabric, characterized by a foliation defined by the shape-preferred orientation of olivine and by crystal preferred orientation (CPO) of olivine, orthopyroxene, and clinopyroxene. This parallelism, the coexistence of microstructural evidence for melt-rock reactions and for deformation by dislocation creep, and the predominance of axial-[010] olivine CPO patterns, characterized by [010] axes clustered normal to the layering, indicate that deformation and the reactive melt percolation that formed the layering were coeval. Strong heterogeneity in mineral composition in the planar layering domains, with Fe-enrichment in olivine and spinel and highly variable Ti content in spinel in cm-scale plagioclase-rich bands, implies that the latter interacted with higher melt volumes. High Ti contents in pyroxenes and spinel in the anastomosed layering domains point to changes in melt composition, hinting to less effective melt transport. We propose therefore that the planar layering records melt segregation in layers parallel to the shear plane, whereas the anastomosed layering results from melt alignment along grain boundaries subparallel to the shear plane, without segregation. Finite strain in the different layers cannot be quantified, but comparison of the present observations with data from experiments and from other mantle outcrops displaying shear-controlled melt organization suggests that the transition from the anastomosed to the planar layering might record an increase in finite strain, that is, strain localization associated with variations in the instantaneous melt fraction.
ISSN: 0040-1951

hal-01053509
https://hal.archives-ouvertes.fr/hal-01053509
DOI : 10.1016/j.tecto.2014.01.007

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The platinum-group element (PGE) systematics of continental mantle peridotites show large variability, reflecting petrogenetic processing of the upper mantle during partial melting and melt/fluid percolation inside the lithosphere. By removing Pd-Cu-Ni rich sulfides, partial melting events that have stabilized the sub-continental mantle lithosphere fractionated PPGEs (Palladium-group PGE; Pt, Pd) relative to IPGEs (Iridium-group PGE; Os, Ir, Ru, Rh). Residual base-metal sulfides (BMS) survive as enclosed IPGE-enriched Monosulfide Solid Solutions (Mss), which otherwise decompose into Ru-Os-Ir-rich refractory platinum-group minerals (PGMs) once the partial melts become S-undersaturated. The small-scale heterogeneous distribution of these microphases may cause extreme nugget effects, as seen in the huge variations in absolute PGE concentrations documented in cratonic peridotites. Magmas fluxing through the lithospheric mantle may change the initial PGE budgets inherited from the melting events, resulting in the great diversity of PGE systematics seen in peridotites from the sub-continental lithosphere. For instance, melt-rock reactions at increasing melt/rock ratios operate as open-system melting processes removing residual BMS/PGMs. Highly percolated peridotites are characterized by extreme PGE depletion, coupled with PGE patterns and Os-isotope compositions that gradually evolve toward that of the percolating melt. Reactions at decreasing melt-rock ratios (usually referred to as "mantle metasomatism") precipitate PPGE-enriched BMS that yield suprachondritic Pd/Ir and occasionally affect Pt/Ir and Rh/Ir ratios as well. Moreover, volatile-rich, small volume melts fractionate Os relative to Ir and S relative to Se, thereby producing rocks with supra-chondritic Os/Ir and S/Se coupled with supra-chondritic Pd/Ir and Pt/Ir. Major magmatic inputs at the lithosphere-asthenosphere boundary may rejuvenate the PGE systematics of the depleted mantle. Integrated studies of "refertilized" peridotites with worldwide provenance provide evidence for mixing between old PGM-rich harzburgitic protoliths and newly-precipitated BMS. Long-lived PGMs carry the Os-isotope compositions of ancient melt‐depletion events into seemingly undepleted fertile lherzolites. Another diagnostic feature of major refertilization processes is the increasing modal abundance of Pt-Pd-Te-Bi or Pt-As-S microphases. Due to regional-scale refertilization processes, sizeable (> 100 km) domains of the upper lithospheric mantle are now significantly enriched in Pd, Au, Cu, Se, and other incompatible chalcophile elements that are of considerable importance in PGE-ore forming events.
ISSN: 0024-4937

hal-00828626
https://hal.archives-ouvertes.fr/hal-00828626
DOI : 10.1016/j.lithos.2012.08.017

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Most peridotite massifs and mantle xenoliths show a wide range of isotopic variations, often involving significant decoupling between Hf, Nd and Sr isotopes. These variations are generally ascribed either to mingling of individual components of contrasted isotopic compositions or to time integration of parent-element enrichment by percolating melts/fluids, superimposed onto previous depletion event(s). However, strong isotopic decoupling may also arise during porous flow as a result of daughter-elements fractionation during solid-liquid interaction. Although porous flow is recognized as an important process in mantle rocks, its effects on mantle isotopic variability have been barely investigated so far. The peridotites of the Lherz massif (French Pyrenees) display a frozen melt percolation front separating highly refractory harzburgites from refertilized lherzolites. Isotopic signatures observed at the melt percolation front show a strong decoupling of Hf from Nd and Sr isotopes that cannot be accounted for by simple mixing involving the harzburgite protolith and the percolating melt. Using one dimensional percolation-diffusion and percolation-reaction modeling, we show that these signatures represent transient isotopic compositions generated by porous flow. These signatures are governed by a few critical parameters such as daughter element concentrations in melt and peridotite, element diffusivity, and efficiency of isotopic homogenization rather than by the chromatographic effect of melt transport and the refertilization reaction. Subtle variations in these parameters may generate significant inter-isotopic decoupling and wide isotopic variations in mantle rocks. (C) 2009 Elsevier B.V. All rights reserved.
ISSN: 0012-821X

hal-00413135
https://hal.archives-ouvertes.fr/hal-00413135
DOI : 10.1016/j.epsl.2008.12.033

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Several recent studies have documented that reactions between melt and crystal mush in primitive gabbroic rocks (via reactive porous flow) have an important control in the formation of the lower oceanic crust and the evolution of MORBs. In this context, olivine-rich rocks can form either by fractional crystallization of primitive melts or by open system reactive percolation of pre-existing (possibly mantle-derived) olivine matrix. To address this question, we performed in-situ trace element analyses (by LA-ICP-MS) of olivine from the Erro-Tobbio ophiolite Unit (Ligurian Alps), where mantle peridotites show gradational contacts with an hectometer-scale body of troctolites and plagioclase wehrlites, and both are cut by later decameter-wide lenses and dykes of olivine gabbros. Previous studies inferred that troctolites and olivine gabbros represent variably differentiated crystallization products from primitive MORB-type melts. Olivines in the three rock types (mantle peridotites, troctolites, olivine gabbros) exhibit distinct geochemical signature and well-defined elemental correlations. As expected, compatible elements (e.g. Ni) show the highest concentrations in peridotites (2580–2730 ppm), intermediate in troctolites (2050–2230 ppm) and lowest in gabbros (1355–1420 ppm), whereas moderate incompatible elements (e.g. Mn, Zn) show the opposite behaviour. By contrast, highly incompatible elements like Zr, Hf, Ti, HREE are variably enriched in olivines of troctolites, and the enrichment in absolute concentrations is coupled to development of significant HFSE/REE fractionation (ZrN/NdN up to 80). AFC modelling shows that such large ZrN/NdN ratios in olivines are consistent with a process of olivine assimilation and plagioclase crystallization at decreasing melt mass, in agreement with textural observations. In-situ trace element geochemistry of olivine, combined with microstructural investigations, thus appears a powerful tool to investigate reactive percolation and the origin of olivine-rich rocks in the lower oceanic crust.
ISSN: 0016-7037

hal-01419938
https://hal.archives-ouvertes.fr/hal-01419938
DOI : 10.1016/j.gca.2016.06.029

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Orogenic lherzolites exhibit a specific (although reproducible) platinum-group element signature characterized by slight positive deviations of Pd/Ir, Rh/Ir and Ru/Ir ratios from the canonical chondritic model. Such a signature was alternatively considered to be a compositional feature of the primitive upper mantle or resulting from sulfide melt addition by refertilisation reactions in the continental lithosphere. To shed additional light on this conundrum, the distribution of PGE in FON B 93, an unserpentinized, fertile (3.24% Al2O3) orogenic lherzolite (French Pyrenees) used as in-house standard by our group, has been studied to different scales, from the bulk rock to trace minerals. In addition to new ICPMS analyses after separation into Nis and Te coprecipitation, its whole-rock PGE concentrations were redetermined by three different laboratories using very precise ID-ICP-MS methods after digestion of the sample in Carius tube at T=250 degrees C and 320 degrees C or in high-pressure asher (HPA-S). The four methods produce reproducible Ru, Rh and Pd contents (7.1 +/- 0.18 ppb; 1.43 +/- 0.05 ppb; 7.1 +/- 0.30 ppb) whereas the non-ID NiS-fire assay method underestimates the Os and Ir concentrations by c.a. 10 and 15% compared to ID-ICP-MS analyses (4.40 +/- 0.07 and 4.00 +/- 0.17 ppb, respectively). Platinum was the most difficult to analyse. If performed on powder aliquots smaller than 3 g, the ID-ICP-MS analyses generate strong nugget effects while the non-ID NiS-fire assay method yields statistically lower (but highly reproducible) Pt concentrations (6.92 +/- 0.26 ppb). These features reflect the strong partitioning of Pt into trace phases that SEM and laser ablation-ICP-MS analyses on thin sections identify as both high-temperature Pt-Ir-Os alloys and Pt-(Pd) tellurides of likely subsolidus origin; ten to fifteen grains ranging in maximum dimension from a few micrometres to a few hundreds of nanometers, were identified by standard polished thin section. LA-ICP-MS data on base metal sulfides (25 grains analysed), coupled with the whole-rock S concentration (277 +/- 10 ppm) and modal composition of the BMS (90% pentlandite+accessory pyrite and secondary pyrrhotite + 10% chalcopyrite) allowed the contribution of the BMS phase to the PGE budget of FON B 93 to be estimated. Except Pt that exhibits a 95% deficit in the BMS phase, the PGE concentrations measured by ID-ICP-MS can be balanced by BMS while Al-spinel is a negligible contributor, accounting for less than 0.5% of the Ru budget. The occurrence of Pt in trace phases may bias the whole-rock Pt concentrations because 1) mechanical collection of Pt-rich trace phases remains problematic with the Nis button, and 2) Pt-Ir-Os alloys may be prone to digestion problems in conventional Carius tube procedures. Since they could be stable at mantle depth, such refractory alloys that contain Os and Ir may also have enhanced the heavy PGEs/light PGEs fractionation. Our observations likely pertain to orogenic lherzolites as a whole because BMS assemblages in these mantle rocks record evolution at low sulfur fugacity, which prevents Pt-Ir-Os alloy from entering the Mss in the mantle; moreover, at subsolidus temperature, pentlandite, the major BMS, cannot accommodate 0 valence state Pt.
ISSN: 0009-2541

hal-00411486
https://hal.archives-ouvertes.fr/hal-00411486
DOI : 10.1016/j.chemgeo.2007.06.030

Éditeur(s) : HAL CCSD Oxford University Press (OUP)
Résumé : International audience
At Cima di Gagnone, garnet peridotite and chlorite harzburgite lenses within pelitic schists and gneisses correspond to eclogite-facies breakdown products of hydrated peridotites and are suitable for studying dehydration of serpentinized mantle. Thermobarometry and pseudosection modelling yield peak temperatures of 750-850°C and pressures <3 GPa. The minimum temperature recorded by the garnet peridotite corresponds to the maximum conditions experienced by the chlorite harzburgite, suggesting that these rocks recrystallized cofacially at ∼800°C. Alternatively, they might have decoupled during subduction, as achieved in tectonically active plate interface boundaries. The major and rare earth element (REE) variability of the peridotites was mostly acquired during pre-subduction mantle evolution as a result of partial melting and reactive melt flow. The ultramafic suite is also characterized by fluid-mobile element enrichments (B, Pb, As, Sb, Cs, Li, U, Be), which confirm derivation from variably serpentinized protoliths. Similarity in the U, Pb, B, Li and Sr contents of the Gagnone peridotites to present-day oceanic serpentinites suggests that these elements were partly taken up during initial serpentinization by seawater-derived fluids. Positive Be, As and Sb anomalies suggest involvement of fluids equilibrated with crustal (metasedimentary) reservoirs during subsequent subduction metamorphism and peridotite entrainment in (meta)sediments. Fluid-mobile element enrichment characterizes all peak eclogitic minerals, implying that multiple hydration events and element influx pre-dated the eclogite-facies dehydration. Peak anhydrous minerals retain B, Li, As and Sb concentrations exceeding primitive mantle values and may introduce geochemical anomalies into the Earth's mantle. The relatively low contents of large ion lithophile elements and light REE in the Gagnone peridotites with respect to much higher enrichments shown by metasomatized garnet peridotite pods hosted in migmatites (Ulten Zone, Eastern Alps) suggest that the crustal rocks at Gagnone did not experience partial melting. The Gagnone garnet peridotite, despite showing evidence for chlorite dehydration, retains significant amounts of fluid-mobile elements documenting that no partial melting occurred upon chlorite breakdown. We propose that the Gagnone ultramafic rocks represent a prime example of multi-stage peridotite hydration and subsequent dehydration in a plate interface setting.
ISSN: 0022-3530

hal-01054319
https://hal.archives-ouvertes.fr/hal-01054319
DOI : 10.1093/petrology/egt068

Éditeur(s) : HAL CCSD Mineralogical Society of America
Résumé : International audience
To constrain the crystallization of alkaline and volatile-rich lavas present in intraplate settings, we studied the petrological features and the geochemical composition of major, trace, and volatile elements of mineral and bulk-rock of two sodalite-bearing phonolites: (1) haüyne-plagioclase-bearing Si-K-rich phonolite from Laacher See (Germany) and (2) nosean-nepheline-bearing Si-poor phonolite from Saghro (Morocco). In haüyne-bearing phonolites (55–59 wt% SiO2, K > Na, Na+K/Al = 0.96–1.08), we found that the low silica and low sodium activity promoted the early crystallization of S-rich haüyne (13.7–13.9 wt% SO3, 0.4 wt% Cl) + S-rich apatite (0.7–0.9 wt% SO3), titanite, and rare pyrrhotite followed by clinopyroxene-plagioclase-sanidine at relatively low pressure and temperature (P = 250 MPa and T = 850 °C) and oxidized condition (ΔNNO-NNO+1, where NNO is nickel-nickel oxide buffer). The crystallization of haüyne occurred at fluid-undersaturated conditions from a silicate melt with 6 wt% H2O, 0.17–0.23 wt% Cl, 0.11–0.4 wt% S, and 0.07–0.14 wt% F. Nosean-bearing phonolites from Saghro are silica-poor and peralkaline (52–54 wt% SiO2, Na > K, Na+K/Al = 1.2) and crystallized at higher P and T (300 MPa and 950 °C) and more reduced conditions (NNO) compared to haüyne-bearing phonolites. The incongruent reaction to form nosean requires high silica and Na2O activity. The mineral assemblage and composition suggest early crystallization of nepheline followed by nosean (7.8–8.8 wt% SO3; 1–1.1 wt% Cl). The sequence of crystallization is: clinopyroxene + nepheline + S-poor apatite (<0.04 wt% SO3) + pyrrhotite followed by nosean and titanite. Nosean-bearing magmas are fluid-undersaturated with relatively low volatile content (4 wt% H2O, <0.25 wt% Cl, <0.056 wt% S, 0.08–0.1 wt% F), although Cl may have exsolved during ascent and formed a fluid phase (NaCl-bearing).Both haüyne- and nosean-bearing phonolites are last equilibrated at relatively low pressure and high temperature. Haüyne and nosean crystallized at oxidized and volatile-rich pre-eruptive conditions. They record the volatile concentrations at depth and may be used as oxybarometer. The incongruent reactions involved to form haüyne and nosean suggest that phonolitic magmas became more oxidized during crystallization. The initial volatile concentrations in basanite/nephelinite magmas, from partial melting of volatile-bearing K2O-rich mantle rock, should have been one important factor influencing the crystallization of haüyne-bearing Si-K-rich phonolite and nosean-bearing Si-poor phonolite in intracontinental setting.
ISSN: 0003-004X

hal-01278067
https://hal.archives-ouvertes.fr/hal-01278067
DOI : 10.2138/am-2015-5318

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Major and trace element signatures and Sr-Nd-Pb isotope data for muscovite (Ms)-bearing amphibolite blocks and associated muscovite-bearing trondhjemite and quartz-muscovite rocks from the Sierra del Convento melange (eastern Cuba) indicate that Proto-Caribbean oceanic crust underwent wet partial melting processes during Mesozoic subduction and after accretion to the upper plate. Trace element normalized patterns of Ms-bearing amphibolites are enriched in light rare earth elements (LREE) and large-ion lithophile elements (LILE) and evidence variable trace element transfer from the Proto-Caribbean subducting slab to the mantle wedge. Ms-bearing trondhjemites show LREE enrichment and HREE depletion and have geochemical features similar to adakites, including SiO2>56 wt.%, high Na2O contents (5.5-9.0 wt.%) and high Sr/Y (16-644). We consider that the trondhjemites represent primary and natural melts formed by deep partial melting of the subducting slab which did not significantly react with the mantle wedge before intrusion in the subduction channel. The Ms-bearing trondhjemites show different geochemical and petrological signatures compared with the Ms-free tonalites-trondhjemites from Sierra del Convent, that are interpreted as primary slab melts. These differences support the idea that partial melting processes in the Sierra del Convento subduction channel were triggered by the infiltration of fluids derived from three distinct subducted sources: sediments, altered mid-ocean ridge basalts (MORB), and serpentinites. In this scenario, the pegmatitic quartz-muscovite rocks, which are highly enriched in LILE, probably represent the crystallization products of fluids derived by differentiation of the trondhjemitic melts.
ISSN: 0024-4937

hal-00669881
https://hal.archives-ouvertes.fr/hal-00669881
DOI : 10.1016/j.lithos.2011.07.011

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
The Sapat peridotite in Kohistan (NW Pakistan) is a 12x1 km sized body of meta-harzburgites, dunites, pyroxenites and meta-gabbros in the hanging wall of the Indus Suture Zone. Field relations, micro-textures, bulk and mineral chemistry document melt-rock reactions producing dunites at the expense of meta-harzburgites, and a consecutive olivine-clinopyroxenite formation from the same melt type. The association of mesa-gabbros, dunites, and pyroxenites is interpreted as a crust-mantle transition zone produced by the percolation of highly depleted primitive arc melt. The meta-harzburgites represent large portions of refractoty mantle that have suffered melt extraction and metasomatism prior to their interaction with the dunite-forming primitive arc melt and the establishment of the crust-mantle transition zone. Uncommon vein assemblages in the dunites that include Cr-clinochlore, calcite and olivine document a late percolation of slab-related, volatile-rich fluids. Taking into account the highly refractory composition of the Sapat mantle, its structural position, the evidence for interaction with highly depleted arc melts and the sub-solidus interaction with volatile-rich arc fluids, the Sapat ultramafic rocks are interpreted as being derived from the frontal Kohistan arc, representing a crust-mantle transition formed by arc magmatism.
ISSN: 0024-4937

hal-00424515
https://hal.archives-ouvertes.fr/hal-00424515
DOI : 10.1016/j.lithos.2008.07.004

Éditeur(s) : HAL CCSD Springer Verlag (Germany) Springer Verlag
Résumé : International audience
Peridotite xenoliths erupted by alkali basaltic volcanoes in the western Pannonian Basin can be divided into two fundamentally contrasting groups. Geochemical characteristics of the abundant protogranular, porphyroclastic and equigranular nodules suggest that these samples originate from an old consolidated and moderately depleted lithospheric mantle domain. In contrast, the geochemical features of the worldwide rare, but in the Pannonian Basin relatively abundant, poikilitic xenoliths attest to a more complex evolution. It has been argued that the origin of the peculiar texture and chemistry may be intimately linked to melt/rock reactions at successively decreasing liquid volumes in a porous melt flow system. The most likely site where such reactions can take place is the asthenosphere-lithosphere boundary. In this context, poikilitic xenoliths may provide petrological and geochemical evidence for reactions between magmatic liquids issued from the uprising asthenosphere and the solid mantle rocks of the lithosphere. These reactions are important agents of the thermal erosion of the lithosphere; thus, they could have considerably contributed to the thinning of the lithosphere in the Pannonian region. We suggest that in the Pannonian Basin, there could be a strong relation between the unusual abundance of poikilitic mantle xenoliths and the strongly eroded lithosphere.
ISSN: 0010-7999

hal-01107421
https://hal.archives-ouvertes.fr/hal-01107421
DOI : 10.1007/s00410-014-1054-y

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Geochemical studies of primitive basalts have documented the presence of crustal-derived garnet pyroxenite in their mantle sources. The processes whereby melts with the signature of garnet pyroxenite are produced in the mantle are, however, poorly understood and somewhat controversial. Here we investigate a natural example of the interaction between melts of garnet pyroxenite derived from recycled plagioclase-rich crust and surrounding mantle in the Ronda peridotite massif. Melting of garnet pyroxenite at ∼1.5 GPa generated spinel websterite residues with MREE/HREE fractionation and preserved the positive Eu anomaly of their garnet pyroxenite precursor in whole-rock and clinopyroxene. Reaction of melts from garnet pyroxenite with depleted surrounding peridotite generated secondary fertile spinel lherzolite. These secondary lherzolites differ from common spinel lherzolite from Ronda and elsewhere by their lower-Mg# in clinopyroxene, orthopyroxene and olivine, lower-Cr# in spinel and higher whole-rock Al2O3, CaO, Sm/Yb and FeO⁎ at a given SiO2. Remarkably, secondary spinel lherzolite shows the geochemical signature of ghost plagioclase in the form of positive Eu and Sr anomalies in whole-rock and clinopyroxene, reflecting the transfer of a low-pressure crustal imprint from recycled pyroxenite to hybridized peridotite. Garnet pyroxenite melting and melt-peridotite interaction, as shown in the Ronda massif, may explain how the signature of subducted or delaminated crust is transferred to the mantle and how a garnet pyroxenite component is introduced into the source region of basalts. The efficiency of these processes in conveying the geochemical imprint of crustal-derived garnet pyroxenite to extruded lavas depends on the reactivity of pyroxenite melt with peridotite and the mantle permeability, which may be controlled by prior refertilization reactions similar to those documented in the Ronda massif. Highly fertile heterogeneities produced by pyroxenite-peridotite interaction, such as secondary spinel lherzolite in Ronda, may nucleate magmatic channels that remain chemically isolated from the ambient mantle and act as preferential pathways for melts with the signature of recycled crust.
ISSN: 0012-821X

hal-00814010
https://hal.archives-ouvertes.fr/hal-00814010
DOI : 10.1016/j.epsl.2012.11.047

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : International audience
The composition of S-rich apatite, of volatile-rich glass inclusions in apatite, and of interstitial glasses in alkaline xenoliths from the 1949 basanite eruption in La Palma has been investigated to constrain the partitioning of volatiles between apatite and alkali-rich melts. The xenoliths are interpreted as cumulates from alkaline La Palma magmas. Apatite contains up to 0.89 wt% SO3 (3560 ppm S), 0.31 wt% Cl, and 0.66 wt% Ce2O3. Sulfur is incorporated in apatite via several independent exchange reactions involving (P5+, Ca2+) vs. (S6+, Si4+, Na+, and Ce3+). The concentration of halogens in phonolitic to trachytic glasses ranges from 0.15 to 0.44 wt% for Cl and from < 0.07 to 0.65 wt% for F. The sulfur concentration in the glasses ranges from 0.06 to 0.23 wt% SO3 (sulfate-saturated systems). The chlorine partition coefficients (D (Cl) (apatite/glass) ) range from 0.4 to 1.3 (average D (Cl) (apatite/glass) = 0.8), in good agreement with the results of experimental data in mafic and rhyolitic system with low Cl concentrations. With increasing F in glass inclusions D (F) (apatite/glass) decreases from 35 to 3. However, most of our data display a high partition coefficient (similar to 30) close to D (F) (apatite/glass) determined experimentally in felsic rock. D (S) (apatite/glass) decreases from 9.1 to 2.9 with increasing SO3 in glass inclusions. The combination of natural and experimental data reveals that the S partition coefficient tends toward a value of 2 for high S content in the glass (> 0.2 wt% SO3). D (S) (apatite/glass) is only slightly dependent on the melt composition and can be expressed as: SO3 apatite (wt%) = 0.157 * ln SO3 glass (wt%) + 0.9834. The phonolitic compositions of glass inclusions in amphibole and hauyne are very similar to evolved melts erupted on La Palma. The lower sulfur content and the higher Cl content in the phonolitic melt compared to basaltic magmas erupted in La Palma suggest that during magma evolution the crystallization of hauyne and pyrrhotite probably buffered the sulfur content of the melt, whereas the evolution of Cl concentration reflects an incompatible behavior. Trachytic compositions similar to those of the (water-rich) glass inclusions analyzed in apatite and clinopyroxene are not found as erupted products. These compositions are interpreted to be formed by the reaction between water-rich phonolitic melt and peridotite wall-rock.
ISSN: 0010-7999

hal-00644803
https://hal.archives-ouvertes.fr/hal-00644803
DOI : 10.1007/s00410-011-0606-7

Éditeur(s) : HAL CCSD Oxford University Press (OUP)
Résumé : International audience
Despite a relatively 'uniform' fertile composition (Al2O3 = 2 center dot 19-4 center dot 47 wt %; Fo% = 89 center dot 2 +/- 0 center dot 3%; Cr#(Spl) = 8 center dot 9 +/- 1 center dot 5%), the Montferrier peridotite xenoliths show a wide range of S contents (22-590 ppm). Most sulphides are interstitial and show peculiar pyrrhotite-pentlandite intergrowths and low abundances of Cu-rich phases. Sulphide-rich samples are characterized by strong enrichment in the light rare earth elements and large ion lithophile elements without concomitant enrichment of the high field strength elements. Such trace-element fractionation is commonly ascribed to metasomatism by volatile-rich melts and/or carbonatitic melts. S and Se (11-67 ppb), as well as S/Se (up to approximate to 12 000), are correlated with La/Sm. Cu, however, remains broadly constant (30 +/- 5 ppm). These features strongly suggest that the percolation-reaction of such volatile-rich fluids has led to sulphide enrichment with an atypical signature marked by strong fractionation of the chalcophile elements (i.e. S vs Se and Cu). S-rich xenoliths are also characterized by high (Pd/Ir)(N) (1 center dot 2-1 center dot 9; where subscript N indicates normalized to chondrite), (Pd/Pt)(N) between 1 center dot 5 and 2 center dot 2, and (Os/Ir)(N) up to 1 center dot 85. Despite the relative uniform fertile composition of the xenoliths, Re/Os ranges between 0 center dot 02 and 0 center dot 18. Os-187/Os-188 is extremely variable even within a single sample and can be as high as 0 center dot 1756 for the most S-rich samples. Sulphides show highly fractionated and variable abundances of the highly siderophile elements (HSE) [0 center dot 03 (Pd/Ir)(N) < 1283] and Re-Os isotopic composition (0 center dot 115 < Os-187/Os-188 < 0 center dot 172). Such variation can be observed at the thin-section scale. Whole-rock and in situ sulphide data demonstrate that chalcophile and HSE systematics and the Os isotopic composition of the upper mantle could be significantly modified through metasomatism, even with volatile-rich fluids. These features highlight the complex behaviour of the HSE in fluid-rock percolation-reaction models and suggest a complex interplay between sulphide addition (crystallization or sulphidation) and partial equilibration of pre-existing sulphide. The specific fractionations observed in chemical proxies such as S and Se, Os and Ir, and Pd and Pt, as well as the low abundance of Cu-rich sulphides, suggest that sulphide addition may not have occurred via sulphide melts. Rather, we infer that S was present as a dissolved species in a (supercritical) oxidizing, volatile-rich fluid (C-O-H-S +/- Cl) along with other chalcophile and siderophile elements such as Os, Pd, Re and Au. The highly radiogenic Os composition of this fluid ( Os-187/ Os-188 > 0 center dot 17) would imply that such fluids are derived from an uncommon type of mantle source possibly related to carbonatite melts.
ISSN: 0022-3530

hal-00644524
https://hal.archives-ouvertes.fr/hal-00644524
DOI : 10.1093/petrology/egr038

Éditeur(s) : HAL CCSD
Résumé : This study provides additional constraints on the relations between deformation, hydration and percolation of fluids and melts in the subcontinental lithospheric mantle beneath a craton and a rift, as well as their implication on its geodynamical behaviour. I have analysed the microstructures, the CPOs, and the hydrogen content of mantle xenoliths from the Kaapvaal craton, and two sets of xenoliths from different localities along the East African Rift (North Tanzanian Divergence and SE Ethiopia).The coarse-granular microstructures and the well-defined CPOs in Kaapvaal peridotites suggest a deformation followed by a long quiescence time. Orthorhombic olivine CPOs predominates, but axial-[100] and axial-[010] are also measured. Cratonic peridotites record multiple metasomatic episodes, leading to a significant compositional heterogeneity, which cannot be imaged by seismic studies. Olivine hydrogen contents are variable, but tend to increase until 150 km depth, reaching up to 50 ppm wt. H2O. The deeper samples are almost dry. Piston-cylinder experiments on hydrogen diffusion between a volatile-rich kimberlitic melt and forsterite suggest that the presence of CO2 in the system could significantly decrease water fugacity and thus forsterite hydration. These experimental results indicate that the hydrogen contents measured in olivine were acquired during a metasomatic event rather than during xenolith extraction by kimberlites. However, this metasomatism was not followed by remobilization of the cratonic root.In the North Tanzanian Divergence, localities within the rift axis and the volcanic transverse belt (Lashaine and Olmani) show significant differences in microstructures and olivine CPO patterns. In Lashaine, coarse-granular microstructures and orthorhombic to axial-[100] CPO patterns in olivine can be explained by transpressional deformation during the formation of the Mozambique belt, or by the occurrence of a remnant of a cratonic domain embedded within the Mozambique belt. Within the rift axis, porphyroclastic to mylonitic microstructures suggest a recent rift-related deformation accompanied by syn-kinematic melt-rock reactions, and followed by variable annealing. The strong heterogeneity in microstructures and olivine CPO suggests that this deformation was acquired during multiple tectonic events probably linked to episodic magma percolation, separated by quiescence episodes. The axial-[100] patterns in olivine and the oblique fast directions reported by SKS studies are coherent with transtensional deformation within the lithospheric mantle beneath the rift.The peridotites from SE Ethiopia are less recrystallized than the rift-axis Tanzanian peridotites, displaying coarse-porphyroclastic microstructures. Microstructures and orthorhombic CPOs in olivine suggest syn- to post-metasomatic deformation. S-waves polarization anisotropies calculated for these samples cannot explain alone the delay times reported by SKS studies in this part of the East-African Rift.
Les travaux réalisés durant cette thèse apportent de nouvelles contraintes sur les relations entre déformation, hydratation et percolation de fluides et/ou de magmas dans le manteau subcontinental sous un craton et sous un rift, et leurs implications sur son comportement rhéologique. Il repose sur l'analyse des microstructures, des OPRs et des teneurs en hydrogène de xénolites mantelliques du craton du Kaapvaal, et sur deux séries de xénolites provenant de différentes localités le long du rift Est-Africain (Divergence Nord Tanzanienne et SE de l'Ethiopie). Les microstructures granulaires à gros grains et les OPRs bien définies des péridotites du craton du Kaapvaal sont cohérentes avec un épisode de déformation suivi d'une longue période de quiescence. Les OPRs de l'olivine sont majoritairement à symétrie orthorhombique, mais des symétries axiale-[100] et axiale-[010] sont aussi mesurées. Les péridotites cratoniques enregistrent de multiples épisodes métasomatiques, ayant entraîné une hétérogénéité de compositions à petite échelle ne pouvant être détectée par les études sismiques. Les teneurs en hydrogène mesurées dans l'olivine sont variables, mais ont tendance à augmenter jusqu'à 150 km de profondeur, atteignant alors jusqu'à 50 ppm wt. H2O. En dessous de cette profondeur, les échantillons montrent des teneurs en hydrogène très faibles. Les expériences réalisées en piston-cylindre sur la diffusion de l'hydrogène issue d'un liquide kimberlitique vers de la forstérite suggèrent que la fugacité en eau pourrait fortement être diminuée par la présence de CO2, empêchant l'hydratation de l'olivine durant extraction des xénolites par les kimberlites. Ces résultats expérimentaux suggèrent que les teneurs en hydrogène dans l'olivine des péridotites du craton du Kaapvaal ont été acquises durant un épisode métasomatique en profondeur et non pendant leur extraction par les kimberlites. Ces teneurs n'ont toutefois pas à ce jour entraîné de remobilisation de la racine cratonique. Enfin, le calcul des propriétés sismiques des péridotites cratoniques révèle que les anisotropies générées par les OPRs de ces échantillons sont suffisantes pour expliquer les anisotropies mesurées par les ondes SKS et les ondes de surface.Les xénolites de la Divergence Nord-Tanzanienne, montrent des variations significatives de microstructures et d'OPR de l'olivine entre les péridotites des localités dans l'axe du rift et celles de la chaîne volcanique transverse (Lashaine et Olmani). A Lashaine, les microstructures granulaires à gros grains et les OPRs de type orthorhombique et axial-[010] peuvent être expliquée par une déformation en transpression liée à la formation de la chaîne Mozambique ou par la présence d'une relique d'un domaine cratonique à l'intérieur de la chaîne Mozambique. Dans l'axe du rift, les microstructures porphyroclastiques à mylonitiques suggèrent une déformation plus récente, accompagnée de réactions magma-roche sous des conditions proches du solidus, suivie d'un recuit variable. L'hétérogénéité des microstructures enregistrées par les échantillons du rift suggère de multiples épisodes de déformation localisée, probablement liés à l'injection percolation épisodique de magmas, espacés de périodes d'accalmie. Les OPRs de l'olivine de type axial-[100] et l'orientation des directions de polarisation des ondes SKS suggère que le rift s'est formé en régime de transtension Les péridotites du Sud-Est de l’Éthiopie présentent des microstructures porphyroclastiques à gros grains moins recristallisées qu'en Tanzanie. Les microstructures et les OPRs principalement de type orthorhombique suggèrent une déformation syn- à post-métasomatisme. Les anisotropies de polarisation des ondes S calculées pour ces échantillons sont insuffisantes pour expliquer à elles seules les déphasages des ondes SKS dans cette partie du rift.
https://tel.archives-ouvertes.fr/tel-01684759

Droits : info:eu-repo/semantics/OpenAccess
NNT : 2014MON20139
tel-01684759
https://tel.archives-ouvertes.fr/tel-01684759
https://tel.archives-ouvertes.fr/tel-01684759/document
https://tel.archives-ouvertes.fr/tel-01684759/file/43443_BAPTISTE_2014_archivage_cor.pdf

Éditeur(s) : HAL CCSD Elsevier
Résumé : Siberian cratonic mantle
International audience
The processes that control water distribution in nominally anhydrous minerals from peridotites are twofold. Melt depletion will remove water while metasomatism can potentially add water to these minerals. These processes can lead to a wide range of outcomes in water contents, which in turn could play a role in mantle rheology and long-term cratonic root stability. To examine these complexities, water concentrations in minerals from well-characterized peridotites from the Udachnaya kimberlite in the central Siberian craton were analyzed by FTIR. The peridotites span a complete top to bottom cross-section of typical cratonic lithospheric mantle (2-7 GPa and 700-1400 °C). Diffusion modeling of water content profiles across olivine grains shows that water loss during decompression is limited to the 100 μm rims of olivines; the cores preserved their mantle water contents. Water contents range from 6 to 323 ppm wt H2O in olivine, 28-301 ppm H2O in orthopyroxene (opx), 100-272 ppm H2O in clinopyroxene (cpx) and 0-23 ppm H2O in garnet. Melting modeling cannot reproduce the high water contents of cratonic mantle peridotites and any potential partial melting trend must have been erased by later events. The water contents of minerals, however, are correlated with modal abundances of clinopyroxene and garnet, bulk rock FeO, TiO2 and SiO2 as well as with light and middle rare earth elements in clinopyroxene and garnet. These relationships are best interpreted as interaction of residual, melt-depleted peridotites with silicate melt, which produced modal and cryptic metasomatism. Importantly, the water enrichment in the Siberian cratonic mantle took place prior to kimberlite magmatism and eruption. Water addition by metasomatism occurred from pressures >4 GPa all the way to the base of the cratonic root below central Siberia, but was limited to shallower levels (<5 GPa) in the Kaapvaal cratonic lithosphere. The difference in olivine water contents at the deepest levels of the Kaapvaal (<5 ppm H2O) and Siberian (6-323 ppm H2O) cratonic roots may be linked to oxygen fugacity and resulting fluid speciation or, alternatively, to reaction with different metasomatic agents. Calculated viscosities for the deepest Udachnaya samples are similar to those inferred for the asthenosphere. If these xenoliths are representative of the deep cratonic lithosphere, water is not as important a parameter as previously thought in the strength of cratonic lithosphere, otherwise the cratonic root beneath Udachnaya would have been delaminated. Alternatively, the metasomatic xenoliths may not be representative of the Siberian cratonic root and kimberlites preferentially sample cratonic mantle lithosphere material located near, and metasomatized by, melt conduits, which served as channels for upward migration of water-rich melts and fluids including kimberlites. In that case, the cratonic root overall still may have relatively low water contents, which in addition to its less metasomatized (more refractory) and thereby buoyant nature, still play a role in making it strong enough to resist delamination by the surrounding asthenosphere.
ISSN: 0016-7037

hal-01053920
https://hal.archives-ouvertes.fr/hal-01053920
DOI : 10.1016/j.gca.2014.04.011

Éditeur(s) : HAL CCSD Oxford University Press (OUP)
Résumé : International audience
The Ronda peridotite massif in Southern Spain shows a well-defined 'recrystallization' front that separates a large-scale partial melting domain formed at the expense of the continental lithospheric mantle from a 'preserved' lithospheric domain. To investigate the processes allowing a transient lithosphere-asthenosphere boundary to propagate in the lithospheric mantle, we performed a joint petrostructural and geochemical study of an similar to 2.5 km(2) zone extending from the melting front to the mylonites that mark the western limit of the massif. This study emphasizes a feedback between heat transfer, melt percolation, and deformation in the lithospheric mantle. Petrographical observations and geochemical data show that heterogeneous reactive percolation of melts produced in the underlying partial melting domain led to refertilization of lithospheric peridotites up to 1.5 km ahead from the melting front, producing metre-scale layering of fertile and refractory mantle rocks. Within 800 m from the front, pre-existing garnet pyroxenite layers were partially molten and the resulting melts probably contributed to the refertilization process. Detailed structural mapping and analysis of the microstructures and crystal preferred orientations highlight the relations between reactive melt transport and deformation, and the control of the temperature gradient on both processes. Parallelism between the recrystallization front, compositional boundaries, and deformation structures, as well as variations in the deformation intensity of pyroxenes and spinels, suggest syn- to late-tectonic melt transport controlled by both the deformation and the thermal gradient. Variations in the strength of olivine crystal preferred orientations as a function of the modal and chemical composition of the spinel tectonites point to a higher contribution of diffusion to deformation in the most fertile rocks, corroborating the hypothesis that deformation occurred in presence of melt. Finally, the systematic dispersion of olivine [100] and orthopyroxene [001] axes in the foliation plane suggests a dominantly transpressive deformation regime.
ISSN: 0022-3530

hal-00420904
https://hal.archives-ouvertes.fr/hal-00420904
DOI : 10.1093/petrology/egp032

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
In the Tahalgha district (southwestern Hoggar, Algeria), the Cenozoic volcanism has sampled subcontinental mantle beneath two crustal terranes that collided during the Pan-African orogeny: the “Polycyclic Central Hoggar” to the east and the “Western Hoggar” to the west. Two major lithospheric shear zones separate these terranes: the “4°35” and the “4°50” faults. Mantle xenoliths were collected between the two faults and across the 4°35 fault. In addition to a range in equilibrium temperatures and chemical compositions reported elsewhere, the samples show variations in their microstructures and crystallographic preferred orientations. Equilibrium temperatures and geochemical characteristics allow dividing them into low — (LT; 700–900 °C), intermediate — (IT; 900–1000 °C), and high-temperature (HT; 1000–1100 °C) xenoliths. The LT and IT peridotites occur on both sides of the 4°35 fault; they are usually coarse-grained. HT xenoliths are present only east of the 4°35 fault, in the narrow domain stuck between the two faults; they are fine-grained and extensively affected by annealing and melt–rock reactions. Microstructures and crystallographic textures indicate that deformation in the LT- and IT-xenoliths occurred through dislocation creep under relatively high-temperature, low-pressure conditions, followed by post-kinematic cooling. The fine-grained HT-xenoliths were deformed under relatively high-stress conditions before being annealed.Combining microstructural and CPO data with petrological and geochemical informations suggests that: (1) the LT xenoliths are remnants of the Neoproterozoic lithospheric mantle that preserved microstructural and chemical characteristics inherited from the Pan-African orogeny, and (2) the HT xenoliths record localized Cenozoic deformation associated with melt channeling through feed-back processes that culminated in the formation of high-permeability porous-flow conduits. Limited grain-growth in HT xenoliths suggests that advective heating of melt conduits was transient and rapidly followed by thermal relaxation due to conductive heat loss into wall-rock peridotites represented by the IT xenoliths, then by exhumation due to volcanic activity.
ISSN: 0040-1951

hal-01171761
https://hal.archives-ouvertes.fr/hal-01171761
DOI : 10.1016/j.tecto.2014.11.012

Éditeur(s) : HAL CCSD AGU and the Geochemical Society
Résumé : International audience
Analysis of the microstructures in the km-scale mantle shear zone that separates the northern and the central parts of the Lanzo peridotite massif provides evidence of an evolution in time and space of deformation processes accommodating shearing in the shallow mantle within an extensional setting. This shear zone displays an asymmetric distribution of deformation facies. From south to north, gradual reorientation of the foliation of coarse porphyroclastic plagioclase-bearing peridotites is followed by development of protomylonites, mylonites, and mm-scale ultramylonite bands. A sharp grain size gradient marks the northern boundary. Early deformation under near-solidus conditions in the south is recorded by preservation of weakly deformed interstitial plagioclase and almost random clinopyroxene and plagioclase crystal orientations. Feedback between deformation and melt transport probably led to melt focusing and strain weakening in the shear zone. Overprint of melt-rock reaction microstructures by solid-state deformation and decrease in recrystallized grain size in the protomylonites and mylonites indicate continued deformation under decreasing temperature. Less enriched peridotite compositions and absence of ultramafic dykes or widespread melt-impregnation microstructures north of the shear zone and clinopyroxene and amphibole enrichment in the mylonites and ultramylonites suggest that the shear zone acted as both a thermal barrier and a high-permeability channel for late crystallizing fluids. These observations, together with chemical data indicating faster cooling of central Lanzo relative to the northern body, corroborate that this shear zone is a mantle detachment fault. All deformation facies have crystal preferred orientations consistent with deformation by dislocation creep with dominant activation of the (010)[100] and (100)[001] systems in olivine and orthopyroxene, respectively. Dynamic recrystallization produces dispersion of olivine CPO but not a change of dominant deformation mechanism. Evidence for activation of grain boundary sliding is limited to mm-scale ultramylonite bands, where solid-state reactions produced very fine grained polymineralic aggregates. Except for these latest stages of deformation, strain localization does not result from the microstructural evolution; the grain size decrease is a consequence of the need to deform a rock volume whose strength continuously increases because of decreasing temperature conditions. Strain localization in the intermediate levels thus essentially results from the more localizing behavior of both the deep, partially molten, and shallow parts of this extensional shear zone distribution.
ISSN: 1525-2027

hal-00644844
https://hal.archives-ouvertes.fr/hal-00644844
DOI : 10.1029/2011GC003627

Éditeur(s) : HAL CCSD
Résumé : Subduction is one of the major geodynamical process active during the Earth evolution. The characterization of arc magmas, their evolution through space and time, and the relationship between arc magmatism and continental crust forming processes, constitute the main research topics in petrology and geochemistry since the last three decades. This study focuses on an exhumed portion of the Kohistan Arc (N Pakistan), which is an exceptional ‘in situ' laboratory to study and follow through time the processes active on subduction zones. Based on a pluri-geochemical approach, this study allows unraveling two major issues : 1/ The lack of a cogenetic relationship between the arc root (i.e. ultramafic rocks) and the overlying gabbroic crustal section. The Jijal ultramafic section does not bear a crustal origin and was emplaced at ~117 Ma via a magma-rock reaction between REE-depleted ‘boninite-like' melts and the pre-existing lithospheric mantle with Indian-MORB affinities; 2/ A three-stage geodynamical model covering the arc evolution through ~30 Ma. This model starts with the onset of the subduction followed by the building of the volcanic arc (1st stage). The 2nd stage corresponds to a major thermal event, characterized by abundant magma underplating and by granulitisation of the arc base. Recycling of the residual/cumulative lower crust into the convective asthenospheric mantle was efficient during this stage, and probably related to thermo-mechanical erosion processes. The last stage, between ~95 and 85 Ma, seals the end of the intra-oceanic subduction and corresponds to an “amagmatic” period. This step is followed by a voluminous, but short, magmatic pulse at ~85 Ma, occurring just before the arc-continent collision. At last, this study emphasizes the more mafic composition of the investigated crustal arc section when compared to the bulk continental crust. Nevertheless the arc section composition nicely fits with the lower continental crust average.
Les zones de subduction représentent des zones d'évolution clés dans la compréhension des processus géologiques majeurs actifs sur notre planète. En particulier, la caractérisation et l'évolution des magmas générés dans les zones d'arcs océaniques et la relation entre leur genèse et les processus de croissance crustale sont des axes primordiaux. Cette étude est focalisée sur la portion d'arc océanique exhumée du Kohistan (Nord Pakistan) qui constitue un laboratoire exceptionnel pour appréhender cette problématique. Basée sur une approche géochimique multi-méthodes (éléments majeurs et en trace, isotopes), cette étude a permis d'établir : 1/ l'absence de relation génétique directe entre les roches ultrabasiques de la racine de l'arc et la section crustale sus-jacente. La formation de la séquence ultrabasique de Jijal se produit à ~117 Ma via une réaction de type magma-roche entre des liquides de type boninitique (appauvris en terres rares), et le manteau lithosphérique de type MORB-Indien ; 2/ un modèle géodynamique en trois stades majeurs résumant l'évolution de l'arc océanique et de la subduction sur une période d'environ 30 Ma. Ce modèle débute par l'initiation de la subduction et la formation de l'arc volcanique s.s. (1er stade) ; le 2ème stade correspond à un évènement thermique majeur. Il est représenté par un sous-placage important de magmas et une granulitisation intense de la base de l'arc. Le recyclage de la croûte inférieure cumulative et résiduelle, dans le manteau sous-jacent, se produit durant cette étape vraisemblablement suite à des processus d'érosion thermo-mécanique. Le dernier stade, entre 95 Ma et 85 Ma, scelle la fin du fonctionnement de la zone de subduction et correspond à une période amagmatique suivie par une brève reprise du magmatisme avant la collision de l'arc après 85 Ma. Enfin, une modélisation numérique comparative avec la croûte continentale globale met en évidence la composition significativement plus basique de la croûte de l'arc insulaire du Kohistan. De fortes similitudes sont en revanche observées entre la section d'arc étudiée et la croûte continentale inférieure.
https://tel.archives-ouvertes.fr/tel-00437770

tel-00437770
https://tel.archives-ouvertes.fr/tel-00437770
https://tel.archives-ouvertes.fr/tel-00437770/document
https://tel.archives-ouvertes.fr/tel-00437770/file/thesedhuime.pdf

Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Spinel (± plagioclase) peridotite xenoliths from the Tallante and Los Perez volcanic centres in the eastern Betics (SE Spain) range from depleted (clinopyroxene-poor) harzburgites to fertile (clinopyroxene-rich) lherzolites and orthopyroxene-free wehrlites. Significantly, only one harzburgite, which is depleted in heavy rare earth elements (HREE), retains the imprint of ca. 20% ancient melting of an original garnet lherzolite source. In contrast, REE abundances of other harzburgites and lherzolites from the eastern Betics have been increased by melt-rock reaction. The whole-rock and mineral compositions of these mantle rocks are largely controlled by three types of modal metasomatism: 1) common clinopyroxene-orthopyroxene addition and olivine consumption which increased FeOt, SiO2 and Al2O3, and decreased MgO compared to the refractory melting products; 2) subordinate orthopyroxene dissolution and precipitation of clinopyroxene and olivine, which led to higher FeOt and MgO and lower SiO2 than in common (orthopyroxene-rich) lherzolites; and 3) rare orthopyroxene consumption and olivine addition that caused higher FeOt and lower SiO2 compared to the original melting residues. These mineral modal and major element variations have been produced mostly by interactions with relatively FeOt-rich/SiO2-poor melts, likely derived from a peridotite-pyroxenite lithospheric mantle with a highly heterogeneous isotopic composition. Melting of the lithospheric mantle in the western Mediterranean was triggered by upwelling of the asthenosphere induced by back-arc extension in the Late Oligocene-Early Miocene. Trapping of small fractions of exotic melts in whole-rocks — likely the parental magmas of Miocene back-arc dykes that intruded the Betic crust — caused local disequilibrium between the trace element signatures and Pb isotopic compositions of clinopyroxene and whole-rock. Subsequent interaction with SiO2-undersaturated magmas, similar to the parental melts of the Pliocene alkali basalts that host the xenoliths, promoted orthopyroxene consumption and clinopyroxene-olivine enrichment at locations close to magma conduits, and finally generated orthopyroxene-free wehrlites. This event constitutes the last episode of the Cenozoic magmatic evolution of the westernmost Mediterranean which is recorded in the mantle xenoliths from the eastern Betics.
ISSN: 0024-4937

hal-01553092
https://hal.archives-ouvertes.fr/hal-01553092
DOI : 10.1016/j.lithos.2016.12.011