Research Collection
Doctoral Thesis
Petrological and chemical investigation of a metamorphosed oceanic crust-mantle section (Chiavenna, Bergell Alps)
Author(s): Talerico, Caterina
Publication Date: 2000
Permanent Link: https://doi.org/10.3929/ethz-a-004138034
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.
ETH Library Diss. ETH No. 13934
Petrological and chemical investigation of a metamorphosed oceanic
CRUST-MANTLE SECTION (CHIAVENNA, BERGELLALPS)
A dissertation submitted to the
SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZÜRICH
for the degree of
Doctor of Natural Science
presented by
Caterina Talerico
Laurea in Scienze Geologiche - Università delta Calabria - Italia born September 11th 1967 citizen of Italy and Zurich (ZH)
accepted on the recommendation of
Prof. V. Trommsdorff, examiner
Dr. P. Ulmer, co-examiner
Prof. L. Morten, co-examiner
2000 Contents
Abstract I
Riassunto Ill
1 Introduction 1
1.1 Context Central Alps 1
1.2 Aims 1
2 Geological overview 3
2.1 Geotectonic position of the Chiavenna unit 3
2.1.1 Former studies on the tectono-metamorphic evolution of the Chiavenna unit 4
2.1.2 General discussion of the ophiolitic nature of the chiavenna unit 5
2.2 Field observations and petrography 7
2.2.1 Ultramafic body and preserved pre-alpine features 7
2.2.2 Amphibolites, calc-silicate and metamorphosed rodingites boudins 9
2.2.3 Metacarbonates: Calcite marbles and ophicarbonatic veins and pockets 14
3 Metamorphism 17
3.1 Problem Statement 17
3.2 Metamorphism in ultramafic rocks 17
3.2.1 Temperature conditions 17
3.2.2 Metamorphic evolution 20
3.2.3 Mineral chemistry 23
3.2.4 Microtextural features 26
3.2.5 Interpretation 27
3.3 Metamorphic evolution of the mafic rocks 28
3.3.1 Mineral assemblages 28
3.3.2 Metamorphic related Compositional Changes in amphibole and plagioclase 29
3.3.2.1 Amphibole breakdown reactions 30
3.3.2.2 Amphibole compositional changes 31
3.3.2.3 Plagioclase compositional changes 33
3.3.3 Summary 34
3.4 Metamorphosed calcareous rocks 35
3.4.1 Phase stability calculations 36
3.4.1.1 Implications 38
3.5 Estimated pressure conditions in the Chiavenna unit 38 3.5.1 Ultramafic and carbonate rocks 38
3.5.2 Tonalitic system 39
3.6 Summary and Interpretation 40
4 Dating of amphiboles from metabasic rocks 43
4.1 Introduction 43
4.1 WAr method 43
4.2 Sample description 44
4.3 39Ar-40Ar data presentation 46
4.3.1 Sample SM71 47
4.3.2 Sample SM44 49
4.3.3 Sample SM60 50
4.3.4 Sample SM81 52
4.3.5 Sample SM28 53
4.3.6 Sample BM4 55
4.4 Discussion and Interpretation of the results on a regional scale 57
4.4.1 Interpretation of the 39Ar/40Ar data 57
4.4.2 Overview of the existing isotope data on the surrounding areas of the Chiavenna unit
58
4.5 Conclusions 59
5 Model of Thermal cooling 61
5.1 Introduction 61
5.2 Mathematical Model 61
5.3 Results 63
5.4 Discussion 65
5.4.1 Was the Gruf Unit responsible for the thermal overprinting event in the chiavenna rocks? 65
5.4.2 Alternative heat-sources 66
5.4.2.1 Masino-Bregaglia related intrusions 66
5.4.2.2 Deep seated bodies 67
5.4.3 Geological implications of gruf unit enhanced contact metamorphism 67
5.5 Conclusions 68
6 Geochemistry 69
6.1 Introduction 69
6.1.1 Aims of geochemical analysis 69
6.2 Bulk rock chemistry of metaultramafic Rocks 69 6 2 1 Major element composition 69
6 2 2 Transition elements 72
6 2 3 Re- and trace element composition 75
6 3 Trace element distribution in ultramafic mineral phases of metamorphic parageneses a laser
ablation icp-ms study 79
6 3 1 Introduction 79
6 3 2 Samples description 79
6 3 3 Results 80
6 3 4 Significance of metamorphic amphibole for mantle processes 82
6 4 Summary and discussion on the Geochemistry of the Chiavenna metapendotites 83
6 5 Metabasic rocks 88
6 5 1 Major and Minor element-chemistry of amphibohtes 89
6 5 2 Magmatic classification 92
6 5 3 Trace and rare earth elements in amphibohtes 93
6 5 4 Rodingites 97
6 5 5 Tectono-magmatic interpretation 97
6 6 Metacarbonates and Metaophicarbonatic rocks 99
6 6 1 Compositional Features and Discussion of the metacarbonates data 99
6 7 Conclusions 102
7 Dumortiente 105
71 The Gruf complex 106
7 2 Crystal structure 107
7 3 Petrogenetic system and stability field 107
7 3 1 Chemical analysis 109
7 3 1 1 Discussion 110
8 Conclusions 113
8 1 Alpine metamorphic evolution 113
8 1 1 Recorded metamorphic events 113
8 1 2 Time constraints of the alpine evolution 115
8 1 3 Interpretation and thermal model 115
8 1 4 General considerations on the tertiary metamorphic evolution of the Chiavenna unit 116
8 2 Bulk-rock chemical characterisation 117
A Appendix 119
A 1 Analytical Methods 119 A.1.1 X-Ray Fluorescence (XRF) 119
A.1.2 Inductively Coupled Plasma Mass Spectrometry (ICP-MS) 119
A.1.3 ICP-MS Laser Ablation 120
A.1.4 Mineral separation and Ar/Ar Dating 120
A. 1.5 Electron Microprobe (EMP) 121
A.2 Sample Location 122
A.3 Analyses 123
A.3.1 XRF-Analyses 123
A.3.2 Laser ablation ICP-MS analyses 129
A.3.3 Icp-ms analyses 131
A.3.4 Argon analyses 135
A.3.5 Electron microprobe analyses 138
References 147
Ringraziamenti Curriculum vitae /
Abstract
Aim of this study is to constrain the metamorphic evolution of the Chiavenna unit (Central Alps) and to define its genetic derivation. The Chiavenna unit, exposed between the Middle Pennine Tambo nappe to the north and the North Pennine Adula-Gruf nappe to the south, consists of metamorphosed ultramafic, mafic and calcareous rocks. The entire unit underwent an intense Alpine deformation that imprinted the rocks with a typical E-W schistosity (S2). The metamorphosed ultramafics consist of deformed peridotitic rocks. Former mantle structures are rarely preserved and are limited to a weakly developed mantle layering formed by dunite and smaller pyroxenite bands crosscut by a N-S magnetite schistosity (S1) discordant to the main E-W schistosity (S2). The ultramafics contain metamorphosed boudins of rodingites surrounded by black-wall as well as various fractures and pockets filled with carbonatic material. Banded and massive amphibolites technically underlay the ultramafics. The banded amphibolites exhibit nematoblastic textures with a developed metamorphic banding. The massive amphibolites, without a preferred orientation of the mineral phases display textures typical of doleritic rocks. None of the amphibolites show typical gabbro textures. Schistose calcite marbles containing amphibolite-breccias and Cr-rich nodules directly underlay the amphibolites.
During the Tertiary Alpine metamorphism the Chiavenna unit underwent a progressive metamorphic evolution at different temperature conditions varying from less than 500°C (diopside-out reaction) in the northern part of the unit to almost 700°C (enstatite-in reaction) in its southern part. Petrological and microtextural analyses indicate the occurrence of two different metamorphic events characterised by different deformation styles, temperature conditions and space distribution. A first dynamic metamorphic event is distinguished by: 1) A well developed antigorite or chlorite schistosity (S2) where large olivine blasts and magnetite bands are elongated parallel to this schistosity. 2) The presence of tremolite and olivine aggregates replacing rotated former diopside, preserving a relic magnetite schistosity (S1). In the amphibolites and in the calcite marbles, the synkinematic event develops foliated textures (S2) formed by the elongation of low-grade greenschist faciès metamorphic phases such as actinolite, low edenite-Mg- hornblende, epidote and low Ca-plagioclase and calcite, quartz as well as phlogopite, respectively. The successive near static thermal event overprinted the dynamic event with mineral assemblages of progressively increasing metamorphic grade. The major features of this event are: 1) The occurrence of metamorphic reactions indicating increasing temperature along a prograde path. The inferred temperature increase is of approximately 200°C, rising from less than 500°C to almost 700°C. The mapped isograds are subparallel to the contact between the Chiavenna and the Gruf units. 2) The development of coarse¬ grained granoblastic textures partially replacing the schistose microfabrics. 3) The occurrence of exchange reactions which involve solid solution phases such as hornblende and plagioclase. During the temperature increase the amphiboles of the mafic system modified their compositions towards progressively higher edenite, AIIV, AIVI and Ti contents, while plagioclase increased its anorthite component until clinopyroxene appeared. The compositional variations of these phases occurred without inducing significant changes in the rock-microfabric. The changes are linked to increasing temperature conditions and correlate with the distance from the Gruf unit. Independent pressure estimates obtained from the high-grade metamorphic calcite marbles as well as from tonalitic dikes crosscutting the amphibolites yield values of approximately 4 kbar. The calcite marbles contain a five-mineral phase assemblage whose composition permits to determine pressure and temperature conditions (4.3 kbar and approximately 670°C) for a fluid composition of 0.17 Xco2. The amphibole composition of the tonalité dikes has an average total aluminium content (AIT) of 1.47 (a.p.f.u.). The substitution of this AIT value in a barometric equation calibrated for amphiboles of calc-alcaline intrusive rocks yields pressures of approximately 4 ± 0.6 kbar. This value refers to conditions achieved during the high temperature conditions and represents the pressure acting during the near static thermal event. 39Ar/40Ar isotope-data of different hornblende compositions in amphibolites from different temperature zones and distances from the Gruf contact confirm the occurrence of multiple metamorphic events. In all analysed amphibole separates, a metamorphic event characterised by the degassing of low Ca/K ratio // "ism
Mg-hornblende can be dated between 30 and 35 Ma. Samples, which underwent temperatures higher than the amphibole blocking temperatures, refine the age-data between 31 and 33 Ma. Amphibole samples that may have reached the amphibole blocking temperature preserve information of an older event characterised by the occurrence of high Ca/K ratio and low edenite hornblende and is dated at approximately 45-47 Ma. In this study it is proposed to consider the metamorphic dynamic event indicated above as a synkinematic greenschist facies metamorphism associated to the E-W syn-collisional extension phase. It occurred after the nappe staking and was dated through the occurrence of edenite poor amphiboles at approximately 45-47 Ma. For the near static thermal event, the narrow isograd zones and the temperature increase of almost 200°C across an average distance of 2.25 kilometres argue for a local metamorphic event possibly induced by the emplacement of a neighbouring hot body enhancing contact metamorphism. For this near static thermal event the argon data indicate an age of approximately 30 and 33 Ma, in agreement with the emplacement age of the migmatitic Gruf complex. Computer modeling of the heat distribution during the emplacement of the Gruf unit within the adjacent Chiavenna ultramafics shows that the computed temperature conditions (approximately 680°C at the contact) are sufficient to stabilise the different mineral-assemblages at similar distances from the Gruf contact as they are observed in the field. These features provide additional evidence that the Gruf unit may have furnished sufficient enthalpy to enhance contact metamorphism in the rocks of the Chiavenna Unit.
The geochemical characterisation of this metamorphic ultramafic, mafic and carbonate rock-sequence leads to significant results. The metapendotites are characterised by high AI2O3 contents (2.7-3.9 wt.%) and high middle to heavy Rare Earth Element contents. The SrWYbN ratio is between 0.21 and 0.69 and can be considered as former Iherzolites documenting a composite mantle history. Major element oxides and REE distribution in bulk-rock as well as in single mineral phases evidence early partial melting processes probably commenced in the garnet stability field (low Tb/Yb ratios between 0.1-0.22) which depleted the Iherzolites in light REE. The negative Zr anomalies point to partial melting processes successively occurring in the spinel stability field. Part of these depleted Iherzolite residua (type B enriched metalherzolites) underwent cryptic mantle metasomatism, which selectively enriched the mantle rocks in light REE as shown by the REE budget of tremolite in the enriched metalherzolites. The Ca-depletion and the negative europium anomalies were caused by the subsequent serpentinisation processes on the seafloor. The banded and massive amphibolites as well as the metamorphosed rodingites display identical chemical features. The analysed mafic rocks plot within the compositional fields of Alpine basaltic rocks and within the field of fresh mid-ocean ridge basalts. The igneous fractionation degree (mafic index) of these rocks varies between 0.4 and 0.7 (Ml) and their major and minor element distribution is in agreement with a tholeiitic differentiation trend. The AFM content of the Chiavenna amphibolites and rodingites is typical for subalkaline basalts of tholeiitic composition. Their chondrite enriched REE patterns show a strong affinity to N-type mid ocean ridge basalts (MORB). The banded amphibolites were probably formed from crystallised liquids, while the massive amphibolites represent shallow depth dolerite bodies. None of the amphibolites show textural or compositional evidence that indicates that they were former gabbros. The calcite marbles are characterised by low MgO contents (<10 wt.%) and plot in the same major element compositional field of other calcite-marbles from the Central Alps. They are interpreted as former carbonates of probably sedimentary origin deposited in an oceanic environment. Evidence for an oceanic stage of the entire sequence are the presence of serpentinites and metamorphosed rodingite dikes as well as ophicarbonatic material filling fractures and pockets within the metapendotites. Furthermore, the occurrence of Cr-rich boudinaged nodules and mafic breccias within the calcite-marbles argues for a direct contact between ultramafic and carbonate rocks and indicate tectonic or sedimentary processes involving the two lithologies before the Alpine collision. These aspects lead to consider the Chiavenna unit as an "incomplete" ophiolitic sequence where subcontinental mantle rocks instead of "typical" oceanic lithosphère are directly exposed on the ocean floor and covered with tholeiitic N-type basalts and carbonates. ///
RlASSUNTO
Lo scopo di questo studio consisite nel definire l'evoluzione metamorfica dell'unità di Chiavenna (Alpi Centrali) e di caratterizzarne la derivazione genetica. L'unità di Chiavenna affiora tra la falda Medio Pennidica Tambo a Nord, e la falda Nord Pennidica Adula-Gruf a Sud. L'unità di Chiavenna è costitutita da rocce ultrafemiche, mafiche e carbonatiche e fu interessata da una intensa deformazione Alpina che sviluppô nelle rocce una tipica scistosità (S2) con direzione E-O. Le rocce metaultrafemiche sono costituite da peridotiti deformate. Strutture mantelliche di età pre-Alpine sono raramente preservate e consistono in una stratificazione debolmente sviluppata costituita da sottili bande dunitiche e pirossenitiche attraversate da una scistosità N-S (S1) di magnetite allungata. Tale scistosità è discordante rispetto alla più diffusa scistosità E-0 (S2). Le ultrafemiti contengono dicchi budinati di rodingiti metamorfiche, circondati da pareti di black-wall. Inoltre, le ultrafemiti presentano varie fratture e cavità contenenti materiale carbonatico. Tettonicamente al di sotto délie ultrafemiti affiorano anfiboliti bandate e massive. Le anfiboliti bandate presentano una struttura nematoblastica e un marcato banding composizionale. Le anfiboliti massive, prive di anisotropia planare presentano strutture tipiche di rocce doleritiche. Nessuna délie anfiboliti bandate e massive mostra tipiche strutture gabbroidi. Direttamente al di sotto délie anfiboliti affiorano marmi a calcite contenenti brecce di anfiboliti e noduli ricchi in Cr.
Durante il metamorfismo Tertiario Alpino, l'unità di Chiavenna subi una evoluzione metamorfica progressiva, le cui condizioni di temperatura variarono a partire da meno di 500°C (reazione diopside-out) nella parte dell'unità più a Nord fino a quasi 700°C (reazione enstatite-in) nella parte più a Sud. Analisi petrologiche e microstrutturali indicano la presenza di due eventi metamorfici distinti, caratterizzati da distinte deformazioni, condizioni di temperatura e da una diversa distribuzione spaziale. Si distingue un primo evento metamorfico dinamico caratterizzato da: 1) Una sviluppata schistosità (S2) data da antigorite e chlorite, in cui grossi blasti di olivina e bande di magnetite sono allungati parallelamente alla scistosità. 2) La presenza di aggregati di tremolite ed olivina che sostituiscono blasti di diopside ruotati aH'intemo délia scistosità. Queste pseudomorfosi preservano una scistosità relitta definita da magnetite (S1). Il primo evento metamorfico sviluppô nelle anfiboliti di basso grado una foliazione data da anfiboli a basso contenuto di edenite, epidoto e plagioclasio povero in Ca, mentre nei marmi a calcite lo stesso evento détermina la formazione di una schistosità definita daH'allungamento di calcite, quarzo e flogopite. L'evento termico successivo di natura statica si sovrappose all'evento dinamico con lo sviluppô di associazioni mineralogiche di grado metamorfico progressivamente crescente. Le caratteristiche principali di questo evento sono: 1) La presenza di reazioni metamorfiche indicanti un aumento délia temperatura lungo un cammino progrado. L'aumento di temperature fu di circa 200°C, durante il quale la temperatura passô da meno di 500°C fino a quasi 700°C. Le corrispondenti isograde hanno un andamento subparallelo al contatto tra le unità del Gruf e di Chiavenna. 2) Lo sviluppô di strutture granoblastiche a grana grossa che localmente sostituiscono le strutture anisotrope. 3) La presenza di reazioni di scambio chimico che interessarono fasi soluzioni solide corne orneblenda e plagioclasio. Durante l'aumento di temperatura, nelle rocce mafiche, l'orneblanda modified la sua composizione verso composizioni progressivamente più rieche in edenite, AIIV, AIVI e TL, mentre il plagioclasio si arricchi in anortite fino alla formazione del clinopirosseno. Le variazioni composizionali di queste fasi avvennero senza modificare significatamente la struttura délia roccia. Le variazioni sono legate all'aumento di temperatura e sono in relazione con la distanza dall'unità del Gruf. Valori di pressione stimati indipendetemente per i marmi a calcite e per dicchi tonalitici che intrudono le anfiboliti indicano pressioni di circa 4 kbar. I marmi a calcite sono costituiti da una associazione mineralogica a cinque fasi, la cui composizione, per un fluido con contenuti in Xco2 di 0.17, permette di determinare le condizioni di pressione e temperatura (rispettivamente, 4.3 kbar e circa 670°C). Le composizioni degli anfiboli presenti nei dicchi tonalitici raggiungono un valore medio di Al totale (AIT) pari a 1.47 (c.p.f.u). la sostituzione di taie valore in una equazione barometrica, calibrata per determinate composizioni di anfiboli tipiche di rocce calcalcaline, indica una pressione di circa 4 kbar. Valore che si riferisce aile condizioni di pressione esistenti durante le alte temperature e di consequenza rappresenta il valore di pressione raggiunto durante l'evento termico statico. IV
« - "-«»HI
La datazione 39Ar/40Ar di anfiboli di diversa composizione di rocce anfibolitiche, provenienti da zone a diversa temperatura e da diverse distanze dal Gruf, conferma la presenza di eventi metamorfici differenti. Per tutti i separati di anfibolo analizzati, è possibile individuare un evento metamorfico, tra 30-35 Ma, caratterizzato dal degassamento di Mg-omeblenda a basso rapporta Ca/K. I campioni che si trovarono al di sopra della temperatura di chiusura per il sistema anfibolo permettono di definire meglio l'intervallo d'età di questo evento tra 31-33 Ma. I campioni di anfiboli che raggiunsero la temperatura di chiusura dell' anfibolo, invece contengono informazioni su un evento metamorfico précédente, caratterizzato dalla presenza di anfiboli ad alto rapporta Ca/K e basso contenuto in edenite. Tale evento è datato a circa 45- 47 Ma. Questo studio propone di considerare l'evento metamorfico dinamico (vedi sopra) come un evento sincinematico in facies metamorfica degli scisti verdi associato alia fase deformativa estensionale E-0 avvenuta durante la collisionale Alpina e successivamente alia fase di impilamento delle falde. Questo evento pud essere datato, dalla presenza di anfiboli a bassi contenuti edenitici a circa 45-47 Ma. Per il metamorfismo termico di natura statica, invece, la presenza di isograde ravvicinate e l'aumento di temperatura di quasi 200°C, lungo una distanza media di 2.25 chilometri, indicano piuttosto un evento metamorfico locale, forse indotto dalla vicinanza di un corpo caldo. Per questo evento i dati argon indicano una età compresa tra 30-33 Ma, età che è in accordo con la possibile età di messa in posto del complesso migmatitico del Gruf. La modellizzazione della distribuzione del calore generata dalla messa in posto dell'unità del Gruf nell'adiacente unità di Chiavenna, dimostra che le condizioni calculate di temperatura (circa 680°C al contatto tra le due unità) sono sufficient a stabilizzare le diverse associazioni mineralogiche a distanze dal Gruf simili a quelle osservate sul terreno. I risultati di questo modello termico costituiscono una indicazione addizionale a sostegno dell'ipotesi che considéra l'unità del Gruf teoricamente capace di fornire un'entalpia sufficients a generare un metamorfismo di contatto, del tipo osservato nelle rocce dell'unità di Chiavenna.
La caratterizzazione geochimica della sequenza di rocce ultramafiche, mafiche e carbonatiche porta ad importanti risultati. Le metaperidotiti caratterizzate da alti contenuti in AI2O3 (2.7-3.9 % in peso) e da alti contenuti in Terre Rare medie e pesanti, possono essere considerate originarie Iherzoliti che documentano una composita storia mantellica. II contenuto in elementi maggiori e in Terre Rare sia della roccia totale che delle singole fasi-mineralogiche, evidenziano la presenza di processi di fusione parziale, probabilmente iniziati nei campo di stabilité del granato (bassi rapporti Tb/Yb compresi tra 0.1-0.22), i quali impoverirono le Iherzoliti in Terre Rare leggere. Le anomalie negative di zirconio, invece, sono spiegate con processi di fusione parziale awenuti successivamente nei campo di stability dello spinello. Parte di queste Iherzoliti residuali impoverite (metalherzoliti arricchite di tipo B) subirono un metasomatismo criptico mantellico, che selettivamente arricchi le rocce mantelliche in Terre Rare leggere come dimostrano i contenuti in Terre Rare presenti nelle tremoliti delle metalherzoliti arricchite. L'impoverimento in calcio e la presenza di anomalie negative di europio, indicano una successiva serpentinizzazione sul fondo marino. Le anfiboliti bandate e massive nonché le rodingiti metamorfiche presentano caratteri chimici identici. Le composizioni delle rocce mafiche analizzate ricadono nei campi composizionali di rocce basaltiche Alpine e nei campo composizionale di basalti freschi di fondo oceanico. I dati si dispongono lungo trend di correlazione per i quali il frazionamento igneo (indice mafico) varia tra 0.4 e 0.7 Ml e la concentrazione in elementi maggiori e minori è in accordo con trend di differenzzazione tholeiitica. II contenuto in AFM delle anfiboliti e rodingiti di Chiavenna associa tali rocce a basalti subalcalini di composizione tholeiitica. Inoltre l'andamento delle Terre Rare normalizzate rispetto alia composizione condritica, tipico di rocce basatiche, revela una forte affinité con basalti di dorsale oceanica di tipo normale (MORB tipo-N). Le anfiboliti bandate probabilmente rappresentano liquidi cristallizzati, mentre le anfiboliti massive sono corpi doleritici di bassa profondità. Nessuna delle anfiboliti présenta evidenze tessiturali 0 composizionali tipiche di gabbri. I marmi a calcite sono caraterizzati da bassi contenuti in MgO (< 10% in peso) e da contenuti in elementi maggiori compresi nell'intervallo composizionale di altri marmi a calcite delle Alpi Centrali. I marmi a calcite sono interpretati come originari sedimenti carbonatici, probabilmente depositati in V .-mta/m
ambiente oceanico. La presenza di serpentiniti, di dicchi rodingitici e di tasche e fratture riempite con materiale oficarbonatico indicano una- fase di oceanizzazione per Tintera sequenza di rocce. Inoltre, all'intemo dei marmi a calcite, la presenza di noduli boudinati ricchi in Cr e di brecce anfibolitiche, suggerisce un diretto contatto tra rocce ultrafemiche e carbonatiche. Un tale contatto spiegherebbe attraverso processi tettonici o sedimentari di età pre-Alpina la presenza di tali litologie nei marmi. Questi aspetti inducono a considerare l'unità di Chiavenna una sequenza ofiolitica "incompleta", in cui rocce di mantello subcontinentale piuttosto che di litosfera oceanica vennero esposte sul fondo oceanico e ricoperte con basalti di tipo N-MORB e carbonati. 1 Introduction
1 Introduction
1.1 Context Central Alps
The Central Alps represent a significant area of the entire Alpine orogen, where several geotectonic nappe systems of different paleogeographic derivation are intimately associated. This area offers complete sections through the Pennine zone, whose position is considered between the Australpine above, and the Helvetic zone below. These three zones constitute the remnants of the Alpine orogen where basement, fragments of oceanic crust and covering rocks are intimately associated. Here the tectono-metamorphic evolution during the Alpine orogeny was particularly intense leading to the occurrence of high metamorphic conditions which locally obliterated former pre-Alpine structural and metamorphic features. Metamorphism is often polyphase and several parageneses may be identified in a single rock sample. In addition, several significant tectonic and magmatic events accompanied the formation of the belt complicating significantly its regional tectono metamorphic configuration.
The Chiavenna unit belongs to the Central Alps and represents a remnant of the Alpine orogen where the two continental margins of Adria and Europe collided. It records several of the Tertiary processes, which accompanied the evolution of this portion of the orogenic belt.
This study largely benefits from the work carried out by Schmutz (1976), who created a detailed map of the area enclosing the Chiavenna unit and furnished an accurate structural and pétrographie description which documents several aspects of the tectono-metamorphic evolution of the unit. The author emphasised the necessity of performing geochemical bulk-rock and mineral-phase analyses of the different lithologies in order to solve questions such as the ophiolitic nature of the Chiavenna unit, which could not be established in his work. Since the seventies, the scientific knowledge about the evolution of the Central Alps substantially improved and especially the large amount of published geochronological data contributed to establishing the various genetic and metamorphic processes. These processes signed the evolutionary history of the Alpine nappes within a time span extending from the Permian and earlier to the present. Together with the development of accurate thermobarometric measurements it was possible to unravel the complex structural and metamorphic evolution of several nappes and units. New geological concepts, were introduced modifying the existing vision on ancient continent-ocean transition zones.
1.2 Aims
The aim of this study is to redefine the significance of the Chiavenna unit gathering new analytical data to support and extend older studies (Schmutz, 1976) and attempt to: 2 Introduction
Distinguish among distinct metamorphic phases overprinting the unit at different physical conditions.
Establish time constraints for the polyphase-metamorphism in the Chiavenna unit.
Identify the regional events that contributed to the metamorphic evolution of the unit.
Set the specific evolution of the Chiavenna unit in a larger regional context.
Propose a possible genetic derivation of the Chiavenna unit defining its significance prior to the Alpine
collision.
The first part (Chapter 3,4 and 5) of this study focuses on the Tertiary evolution of the Chiavenna unit, investigating its metamorphic assemblages and textures. In addition, 39Ar/40Ar dating of amphiboles help constrain the timing of the Tertiary metamorphic events overprinting the unit. Chapter 5 presents the results of a mathematical model that reproduces the petrologic situation of the Chiavenna unit. The second part (Chapter 6) of this study discusses detailed bulk-rock features as well as data carried out on single mineral phases in the attempt to identify the pre-Alpine nature of the unit. Chapter 7 focusses on the composition of the dumortierite samples found in the study area. This mineral occurs as needles within aplitic veins hosted by erratic gneiss blocks of supposed Gruf affiliation. This chapter is not related to main arguments discussed in this study, but presents the chemical results carried out on dumortierite samples. 3 Geological overview
2 Geological overview
This chapter illustrates the geotectonic position of the Chiavenna unit at the regional scale. A brief description of the surrounding nappes and units, with a review of the existing studies on the tectono- metamorphic evolution and paleogeographic derivation of the Chiavenna unit is presented. It also contains a detailed description of the field relationships among the different rock types, as well as their pétrographie features.
2.1 Geotectonic position of the Chiavenna unit
The Chiavenna unit, has been defined as a mafic-ultramafic complex (Schmutz, 1976), and is geographically situated along the southeastern Swiss-Italian border and belongs to the Central Alps. It borders the Middle Pennine Tambo nappe to the north and the North Pennine Gruf unit (considered part of the Adula nappe) to the south (Figure 2-1).
Figure 2-1: Geotectonic map of the Central Alps modified after Schmid et al. (1996a), C = Chiavenna Unit.
The Gruf unit, tectonically underlying the Chiavenna unit, is essentially a gneissic unit consisting of migmatite-bearing granulite faciès rocks (Bucher-Nurminen and Droop, 1983) of unresolved origin. It has been correlated with the Bellinzona-Dascio zone (Milnes and Pfiffner, 1980; Wenk, 1973) and with the
Adula nappe (Huber, 1999). Several studies (Wenk, 1970 and 1973; Gulson, 1973; Trommsdorff and
Nievergelt, 1983; Davidson et al., 1996) showed that the Gruf unit is crosscut by the Tertiary Masino-
Bregaglia intrusions. The age of the granulites facies metamorphism in the Gruf unit is uncertain. Whether it is pre-Alpine (Gulson, 1973) or Alpine in age (Bucher-Nurminen and Droop, 1983; Davidson et al, 1996; 4 Geological overview ^ ^ is «* * ilölli
Wenk et al., 1974) is still an open question. The emplacement of the Gruf unit in its present position is thought be related to late Alpine events (Droop and Bûcher, 1984).
The Middle Penninic Tambo nappe is composed of basement rocks covered by Permo-Mesozoic sediments. The basement rocks underwent a polymetamorphic evolution (Baudin and Marquer, 1993
Marquer, et al, 1994; Marquer, 1991) and are crosscut by a Permian monometamorphic granitic intrusion
(Truzzo granite) (Marquer, 1991; Marquer et al, 1996). Together with the Novate granite, the Masino-
Bregaglia tonalite-granodioritic intrusions are considered the major Alpine magmatic events of the Central
Alps. Although both magmatic bodies are not directly abutting the Chiavenna unit and are distinct in age, they influenced the tectono-metamorphic evolution of this unit, as well as the entire Central Alps.
2.1.1 Former studies on the tectono-metamorphic evolution of the Chiavenna unit
A basic study on the metamorphic evolution of the Chiavenna unit was presented by Schmutz, (1976) who compiled a detailed geological map (Figure 2-2) as well as the occurrence of metamorphic isograds in the area. Schmutz (1976) interpreted the isograds as the results of a progressive regional Alpine metamorphism up to amphibolite fades condition at a pressure of 3.5-4 kbar.
Figure 2-2: Geological map of the Chiavenna unit, modified after Schmutz (1976).
Recent works (Huber and Marquer, 1996; Huber and Marquer, 1998 and references therein) redefined the tectono-metamorphic evolution of the Chiavenna unit during the Tertiary collision. These studies compared the structural phases and the metamorphic evolution observed in peridotites of the Chiavenna unit with those in the surrounding nappes (Tambo and Suretta). In the Tambo and Surretta nappes four main deformation phases overprinting the pre-Alpine structures were recognised. During the Eocene, NW- directed thrusts brought the Pennine units into a subduction zone environment (D1 deformation) (Marquer et al. 1994, Baudin and Marquer, 1993). During subduction of the Pennine oceanic crust, the upper
Pennine units (Tambo and Suretta nappes) built an accretionary prism. The Eocene subduction was 5 Geological overview
followed by the Oligo-Miocene collision generating penetrative D2 structures, related to ductile, syn-
collisional E-W extension, which crosscut the tectonic contacts between the nappes. D3- and D4-
structures are less penetrative than D2 structures. D3 structures are linked to the vertical extrusion of the
crustal block situated to the north of the Insubric lineament and formed during late folding. This
deformation overprinted and steepened the previous structures. At this time, the Bregaglia tonalite-
granodioritic intrusions (D3) also were emplaced. The latest structures (D4) correspond to a NE-SW
extension and consist of brittle normal faults (i.e. Forcola fault) crosscutting all previous structures. The
younger Novate granitic intrusion (25 Ma) underwent only D4-deformation (normal faulting).
In the Chiavenna unit the most pervasive structures consist of a well developed E-W schistosity which
correlates with the syn-collisional E-W extensional phase (D2) (see above). Older structures were found in
ultramafic rocks and are restricted to a weak mantle layering formed by pyroxenitic and dunitic layers
crosscut by a approximately N-S oriented magnetite-schistosity discordant to the S2 schistosity. Boudins
of metamorphosed rodingites within ultramafic rocks indicate that the deformation phase related to the
boudinage of the rodingites (competence contrast between rodingite and serpentinite (Scambelluri et al,
1995)) occurred after serpentinisation. Whether it is Alpine or pre-Alpine is not clear. The youngest
structures are folds (D3), which deformed the S2 schistosity.
2.1.2 General discussion of the ophiolitic nature of the chiavenna unit
The mafic-ultramafic lithological sequence of the Chiavenna unit was regarded by Schmutz (1976) as
an overturned ophiolite body genetically related to the Tambo and Surretta nappes. According to this
author, the ophiolite body comprises basal Iherzolites overlain by massive metagabbros. The metagabbros
are progressively replaced by banded fine-grained amphibolites, interpreted as relics of basaltic lavas and
diabase dikes. In addition to the ultramafic-mafic lithologies, metacarbonates are present in the form of
thin deformed layers (2.2.3). Schmutz suggested that the carbonates were emplaced on the above
mentioned mafic-ultramafic sequence before it was overturned, and the emplacement of the carbonates
occurred either by a tectonic mechanism or through sedimentary processes. Despite the absence of
features characteristic of "normal" ophiolitic sequences such as cumulus layers or silica-rich sediments,
Schmutz considered the layered sequence exposed in the Chiavenna area as a possible "ophiolite".
The work of Huber and Marquer (1998) suggested an alternative paleogeographic scenario for the
Chiavenna unit. These authors, based on structural and field observation, consider the Chiavenna unit as
a thinned continental margin within a lithospheric extensional regime, where subcontinental mantle rocks
have been exposed along normal faults and subsequently covered by. Coarse-grained amphibolites were
interpreted as former gabbros, which intruded the mantle-crust boundary before the Mesozoic extension.
Fine-grained amphibolites were interpreted as volcanic basalts, which were extruded during the Mesozoic opening of the Valais oceanic-basin. This is a possible scenario if the overlying carbonates are assumed 6 Geological overview
to be Mesozoic in age, whereas the oceanic stage of the Chiavenna unit and therefore its ophiolitic nature
remain dubious.
A scenario, whereby subcontinental mantle rocks are exposed by lithospheric faulting, has also been
proposed for the present day passive continental margin of the Galicia bank (Boillot et al, 1987) and for
other localities of the Alpine orogen such as the Malenco unit (Trommsdorff et al, 1993) and the Tasna
nappe (Florineth and Froitzheim, 1994). These studies showed that the Alpine oceanic rocks were part of
small basins, often floored by exhumed subcontinental mantle rocks instead of typical oceanic lithosphère
(Froitzheim et al, 1996; Lemoine et al, 1987).
This new concept of ophiolitic sequences, contrasting with the classical definition, has become more
and more accepted in the research community. The common ophiolite term indicating distinctive
assemblages of mafic and ultramafic rocks forming a sequence consisting of: variable propotions of
Iherzolite, harzburgite and dunite, gabbroic complex, mafic sheeted dike complex and volcanic pillow rocks
associated with cherts, shales and limestones, as well as chromite and dunite bodies, sodic felsic intrusive
and extrusive rocks (Anonymous, 1972), has been substituted by a new vision. In fact, modern studies
(Hébert et al, 1990; Hekinian et al, 1993, and others) have shown that ophiolite bodies are rarely
completely developed as previously assumed and are often lacking in some of the above-defined layers or
may be derived from different oceanic environments. Comparison between drilled oceanic samples from
the Galicia margin and rock-samples from the Alpine ophiolites demonstrated strong analogies between
the present day oceanic crust along passive continental margins and the ancient one. In this respect, the
idea of Schmutz (1976) that the Chiavenna mafic-ultramafic complex is a former ophiolitic body, is
consistent with current ideas on oceanic crust and its origin.
The Valais basin represents one of the small oceanic basins flooring the Tethys, within the context of a
"multi-oceanic" basin system (Platt et al, 1989). Paleogeographically, the Valais basin is located between
the Briançonnais terrain nappes and the southern border of the European margin. In the Central Alps, the
Briançonnais sequences are built up by the cover nappes Schams, Falknis and Sulzfluh in Graubünden
and by the Tambo and Surretta basement-cover nappes in the Valtellina area (Froitzheim et al, 1996).
The Briançonnais is considered to represent a continental fragment of Iberia, separating the southern
Ligure-Piemontese oceanic basin from the northern Valais oceanic basin (Stampfli, 1993; Triimpy, 1980).
Towards the northeast, this latter basin borders the European margin formed by the Adula nappe (Schmid etal., 1997; Triimpy, 1980).
In the eastern part of the Valais basin, continental break-up occurred in the Early Cretaceous
(Froitzheim et al, 1996). The closure of this basin started with the rotation of Iberia in the Late
Cretaceous. The age of subduction of the Valais basin is still a matter of debate, especially since no subduction-related sediments have been dated. Comparison of the Valais basin with the Provence basin indicates that the final closure along a suture zone occurred in the Eocene (40 Ma). This suture zone can be followed from the northern part of Graubünden (Switzerland) through the Misoxer zone (Switzerland- 7 Geological overview
Z - a-v^1 .UBS«—
Italy) between the Tambo and the Adula nappes and finally between the Penninic gneiss nappes of the
Lepontine area to join the Valais zone in the West (Froitzheim et al, 1996; Schmid et al, 1996b).
On the basis of these constraints, the Chiavenna unit, exposed between the middle Penninic Tambo nappe and the north Penninic Gruf unit, may be interpreted as oceanic remnants of the Valais basin and part of the suture zone of the southern part of the Misoxer area (Schmid et al, 1996b).
2.2 Field observations and petrography
The Chiavenna unit is mainly formed by metaultramafic rocks, amphibolites and metacarbonates, whereby more than 90% of the investigated area is covered with metaultramafic rocks and amphibolites
(Figure 2-2). Metacarbonate outcrops are limited to a few localities. A complete outcrop sequence comprising all the major lithologies is exposed along the Schiesone River near Prata Camportaccio.
2.2.1 Ultramafic body and preserved pre-alpine features
The metamorphosed ultramafic body of the Chiavenna unit consists of variously deformed peridotitic rocks. Important outcrops exist along the Schiesone River, where the continuous water flow of the river denuded and smoothed the underlying rocks. Other outcrops can be found near Uschione, South of the town of Chiavenna, within the town and on the southern side of Val Bregaglia next to the locality of Prosta, where boudins of metamorphosed rodingites are embedded within metapendotites. The main street connecting Villa di Chiavenna with Chiavenna crosscuts the northeastern outcrops. Foliated metapendotites are also present in the upper part of the steep valley Val Aurosina and on the Swiss part of Val Bregaglia near Alpe Foppate.
Figure 2-3: Ultramafic rock-outcrops of the Chiavenna Unit: a) weathered (brown coloured domains) massive peridotite outcrop on the southern slope of the Bregaglia valley, b) outcrop of ultramafic rocks preserving evidence of former mantle layering: detail showing a dunitic layer associated with a small pyroxenite layer, both layers are discordant to the main magnetite-foliation. 8 Geological overview
The entire unit underwent a complex Alpine deformation that imprints the ultramafitites with a typical
schistosity (Figure 2-3) defined by the orientation of minerals such as magnetite, serpentine minerals,
chlorite and olivine. The grain size of the rock-minerals varies from coarse grained to medium-fine grained.
Frequently, a brown coloured alteration pattern covers the entire rock (Figure 2-3). Fresh dark-green
coloured metapendotites are present along the Mera River in the most northern part of the study area,
where serpentine (i.e. antigorite) and chlorite give the ultramafic rock a dark green colour and a smooth
surface. Veins with centimetre-scale olivine blasts and chlorite and in some case talc are often present
within the metapendotites. Dunites and rare pyroxenites form thin boudinaged layers parallel to the main
foliation of the host metaperidotite. The intense Alpine deformation obliterated any former mantle
structures or primary relationship between the dunites and metapendotites. One single ultramafic outcrop
(Figure 2-3b) (locality of Lotteno) preserves structural evidence of a former mantle layering. In this
outcrop, the main Alpine foliation defined by slightly elongated magnetite is discordant to layering formed
by dunitic and smaller pyroxenitic bands. These features are considered the only structural indication of
pre-Alpine relics in the study area.
Typical minerals characterising the Chiavenna metamorphosed ultramafic rocks are olivine, tremolite,
chlorite, and magnetite. In addition, other minerals such as antigorite, talc, anthophyllite, cummingtonite,
diopside, enstatite and spinel are present (Figure 2-4), depending on the particular temperature, pressure
conditions experienced, as well as compositional.
Antigorite characterises the northern ultramafitites, it occurs together with olivine blasts, chlorite,
magnetite and talc. Antigorite, chlorite and porphyroblastic or coarse-grained olivine define the main
foliation observed in the ultramafites. In rare outcrops it is also possible to recognise relict rotated
clinopyroxene within a schistose antigorite-chlorite matrix, or else tremolite and olivine aggregates
replacing pyroxene, crosscut by oriented magnetite bands. Anthophyllite needles can be found in the
southern outcrops. Anthophyllite is normally associated with olivine, chlorite, talc and magnetite, whereas
it also may occur with enstatite in the south. Cummingtonite needles are invariably associated with tremolites and often form a thin border around the tremolite needles. Enstatite together with green Al-
spinel, anthophyllite, olivine and chlorite, as well as magnetite may be found close to the Gruf contact. Al-
spinel rarely preserves a brownish Cr-rich core.
The microfabric of these mineral assemblages varies from north to south, especially towards the
contact with the Gruf unit. The presence of high-temperature phases such as anthophyllite, cummingtonite and enstatite support the development of a coarse grained texture, hence the typical foliation of the
northern outcrops tends to disappear or to become weaker with increasing degrees of recrystallisation.
Elongated porphyroblastic olivines are typical of the antigorite- and chlorite-bearing foliated
metapendotites of the north, while in the south granoblastic olivines of smaller grain size occur in a
polygonal texture associated with anthophyllite- and enstatite-bearing assemblages. 9 Geological overview
Figure 2-4: Photomicrographs of reppresentative mineral assemblages observed in the ultramafics of the Chiavenna Unit (crossed polars): a) coarse-grained enstatite-olivine-tremolite-chlorite metaperidotite (1069), b) anthophyllite needles within an olivine-chlorite-magnetite matrix (SU76), c) detail green spinel within enstatite, olivine, tremolite and chlorite metaperidotite (plane polars, 1069).
2.2.2 Amphibolites, Calc-silicate and metamorphosed rodingites boudins
More than 50% of the entire Chiavenna unit consists of amphibolite rocks tectonically underlying the
ultramafic rocks. Based on field observations, it is possible to distinguish between dark-green coloured
banded amphibolites and massive amphibolites (Figure 2-5). Banded amphibolites display alternating thin white feldspar bands and dark-green amphibole bands. This mineral banding is parallel to the main foliation formed by slightly elongated amphibole and plagioclase blasts and often it is difficult to distinguish
between metamorphic banding and foliation. Massive amphibolites are less abundant than banded amphibolites and are made up of millimetre-size white needles of feldspar with minor amounts of quartz
randomly dispersed in an amphibole matrix. The amphibolites of the Chiavenna unit are crosscut by various generations of dikes and veins (Figure 2-6). The intrusions show variable compositions and textures, ranging from granitic to tonalitic and from aplitic to pegmatitic. The structural relationship between the intrusions and the main foliation (S2) of the host rock is mainly syn-deformation for the tonalitic dikes and post deformation for the aplitic veins. In thin-section the tonalité dikes are prevalently composed of plagioclase, green-hornblende, biotite with lesser amounts of quartz and rare potassic 10 Geological overview
feldspar. The amphibolites are also marked by the presence of thin epidote or feldspar veins post-dating the compositional banding and the foliation of the host rock.
Figure 2-5: Samples of dark-green amphibolites of the Chiavenna unit: a) banded amphibolites: metamorphic banding consists of thin feldspar bands and amphibole-bearing layers, b) detail of massive amphibolites: needles of plagioclase randomly dispersed in an amphibole matrix.
The banded and massive amphibolites display a mineral assemblage (Figure 2-7) composed of amphibole, plagioclase, titanite, quartz, ilmenite, chlorite and to a lesser extent, biotite, apatite, rutile and zircon (Table 3-2, section 3.3.1). In the northern outcrops, clinozoisite-epidote also is observed, while towards the south, close to the Gruf unit, diopside appears. The banded amphibolites show relatively simple textural features. In the northern outcrops, the amphiboles develop nematoblastic textures associated with sub-grain recrystallisation of feldspar and quartz. The grain size of the mineral blasts varies from fine- to coarse-grained. The foliation, mainly given by the elongation of amphibole and plagioclase, is easily recognisable in the clinozoisite-epidote bearing amphibolites and becomes weaker or parallel to the metamorphic banding towards the south. The hornblende- and diopside-bearing amphibolites of the southern outcrops often show a granoblastic polygonal-texture, indicating that significant recrystallisation took place during high-grade metamorphic conditions. The textural analysis of these amphibolites shows that typical gabbro textures such as flaser texture or inhomogeneous phase distributions are not preserved. In contrast, the banded amphibolites show a homogenous distribution of plagioclase and amphibole blasts that are characterised by a relatively small grain size. These features may be regarded as relic features of a basaltic rather than of a gabbroic texture.
Compared with the banded amphibolites, the massive amphibolites display the same mineral- assemblages, but with distinct textural features. In thin-section, characteristic white needles are formed by aggregates of recrystallised plagioclase and small amphibole blasts, whereas the remaining part of the rock is composed of amphibole, quartz and titanite. A first generation of plagioclase (G1) in the central part of the feldspar-bearing domains is completely substituted by younger polygonal plagioclase blasts (G2) along the borders. The absence of a preferred orientation of the mineral phases suggests that deformation was lacking or extremely weak and that temperature was the only important factor driving the recrystallisation of the rocks. The textural features characterising the massive amphibolites are very 11 Geological overview
similar to textures typical of doleritic rocks, indicating that the massive amphibolites may represent
metamorphosed shallow-depth mafic intrusive bodies.
Figure 2-6: Several generations of small intrusions and veins crosscutting the amphibolites from Chiavenna: a) tonalitic dike within amphibolites, the dike is subparallel to the metamorphic layering and the foliation, b) folded aplitic vein in foliated amphibolites, c) evidence of a discordant relationship between foliation of the small amphibolite pebble enclosed in an aplitic vein and the foliation of the intruded amphibolite.
The banded amphibolites often contain green nodules and boudins of calc-silicate rocks (Figure 2-8a)
embedded within the metamorphic layering, parallel to the foliation of the host rock.
The calc-silicate layers or nodules display an heterogranular texture characterised by coarse-grained
pale-green diopside and epidote layers, embedded within a nematoblastic amphibole matrix. Diopside and
clinozoisite-epidote are elongated parallel to the foliation and to the compositional banding of the rock.
Commonly, diopside consists of large elongated blasts, whereas clinozoisite and epidote form smaller subrounded granules. The border between amphibolitic host rock and calc-silicate layers is not generally sharp, and various amounts of Ca-rich minerals may be included in the amphibole-plagioclase matrix of the host amphibolite.
Rare metamorphosed rodingite rocks (Figure 2-9) form decimetre-scale boudins of metasomatised
mafic material enclosed within the metapendotites. On the rock surface, elongated red grossular blasts are disseminated in a dark-green, fine-grained matrix. The boudins are surrounded by a centimetre-thick black-wall separating them from the host metapendotites. The occurrence of boudins parallel to the metaperidotite-foliation suggests a competence contrast between the rigid metamorphosed rodingite and the ductile antigorite-bearing metaperidotite (Scambelluri et al, 1995), which supports the idea of a pre- deformation intrusion and rodingitisation of the mafic dikes within the metapendotites.
Rodingitisation is defined as a metasomatism of mafic rocks enclosed in metamorphic ultramafitites undergoing serpentinisation whereby Ca-exchange between the host peridotites and the mafic rock causes Ca enrichment in the mafitite (Coleman, 1966). This metasomatic bulk-rock exchange drives the 12 Geological overview
crystallisation of calc-silicate phases such as grossular, minerals of the epidote-group, vesuvianite and other Ca rich minerals in the mafic rocks.
Figure 2-7: Photomicrographs of banded and massive amphibolites from Chiavenna (plane-polars): a) banded fine-grained epidote bearing amphibolites (BM4), b) massive amphibolites (SM82), c) small rutile inclusions in amphibole (SM82), d) typical titanite blasts with ilmenite inclusions (SM54).
Figure 2-8: a) boudinaged calc-silicate layer within amphibolites, b) amphibolite-breccia within deformed calcite marbles.
The metamorphic assemblage of the metarodingites consists of several generations of mineral-phases such as grossular, titanite, clinozoisite, epidote, diopside, vesuvianite and green hornblende (Figure 2-10). 13 Geological overview
Elongated grossular porphyroblasts enclose epidote, titanite, diopside and microgranular aggregates of epidote-clinozoisite. Coarse-grained clinozoisite and epidote also form aggregates with vesuvianite. In addition, green-hornblende bands are frequently associated with chlorite, separating the elongated garnets from domains of coarse-grained epidote-clinozoisite, hornblende, chlorite, garnet and titanite.
Figure 2-9: Boudinaged metarodingite mafic dikes of the Chiavenna unit: a) decimetre-long deformed boudin of metarodingites within foliated peridotites. Reddish garnet blasts immersed in a green matrix, with black-wall borders (dark- green) at the contact to the foliated metaperidotite, the axis of the metarodingitic boudin is parallel to the metaperidotite foliation, b) polished drilled sample of metarodingite.
ib).
Chlorite Amphibole matrix red grossular blasts
Noteworthy is the presence of amphibolite breccias within the metacarbonates directly underlying the amphibolites. The breccias are deformed subparallel to the schistosity of the metacarbonates suggesting a pre-deformation relationship between the metacarbonates and the amphibolite breccias (Figure 2-8b).
Figure 2-10: Photomicrograph of metarodingite samples from the Chiavenna unit (crossed polars): a) microgranular inclusions of epidote-clinozoisite aggregates in elongated garnet (grossular) blasts (BM34), b) intergrowth of clinozoisite and epidote with vesuvianite blasts (BM34). 14 Geological overview
2.2.3 Metacarbonates: Calcite marbles and ophicarbonatic veins and pockets
The metacarbonates of the Chiavenna unit can be divided into two main groups: calcite-marbles and ophicarbonates (Figure 2-11). The metacarbonates are subordinate to the metaultramafitites and amphibolites. Calcite-marbles are limited to a few deformed outcrops of metre-scale, while decimetre-size veins and pockets within the ultramafics are filled with ophicarbonatic material.
Figure 2-11: Metacarbontes of the Chiavenna unit: a) banded and deformed calcite-marbles, the dashed line indicates the schistosity, b) ophicarbonatic pockets and veins within metapendotites, c) contact between the calcite-marbles and the overlying amphibolites, d) Cr-rich nodules within the banded calcite- marbles, the dashed line indicates the schistosity within the marbles.
The calcite-marbles consist of folded and deformed banded metacarbonate material of different colours. They appear to be in primary contact with the overlaying foliated amphibolites, since the compositional banding and foliation of the marbles parallels the main foliation and banding of the amphibolites. Often they are intensely weathered and exhibit a "sugary" consistence. Similar to the amphibolites, the calcite-marbles are crosscut by magmatic dikes, which are partially brecciated and incorporated in the marbles. Rarely, decimetre-size intense green-coloured nodules of calcsilicate-rich minerals are embedded within the calcite marbles. The calcite-marbles consist of calcite, diopside, plagioclase, titanite and quartz. Small compositional differences and variation in metamorphic grade, may locally develop phlogopite, hornblende, epidote-clinozoisite, garnet, scapolite, potassium feldspar and 15 Geological overview
wollastonite (Figure 2-12). Generally a weakly schistose granoblastic texture is present. Relict phases can often still be recognised. For instance, garnet may include scapolite and calcite, or wollastonite, plagioclase and calcite. Furthermore, plagioclase may completely enclose scapolite blasts.
Figure 2-12: Photomicrographs of the calcite-marbles (crossed polars): a) typical weakly schistose mosaic texture of calcite-marbles (SC53) containing calcite (Cc), diopside (Di), grossular (Grs), scapolite (Sep), plagioclase (PI) and wollastonite (Wo), b) schistose mosaic texture of a calcite-marble at lower metamorphic conditions (BC18) containing calcite, quartz (Qtz) and phlogopite (Phi).
Of particular interest are the mineral phases of the green nodules, which are found embedded within the calcite marbles (Figure 2-11d), parallel to the main schistosity. The nodules display a granoblastic to polygonal texture predominantly formed by clinopyroxene and to a minor amount by plagioclase, garnet, scapolite, small amphibole needles, chlorite and biotite (Figure 2-13).
Figure 2-13: Photomicrograph of mineral assemblages of Cr-rich calc-silicate nodules within calcite marbles in the centre Cr rich garnet (plane polars) and its solid solution components.
The ophicarbonatic veins and pockets are randomly dispersed throughout the ultramafic body (Figure
2-11b), with abundant outcrops close to the tectonic contact between metapendotites and amphibolites.
They are more deformed than the host metapendotites, showing a distinct material competence.
Compositionally, there are no differences observed between pockets and veins. Similar to the marbles, the 16 Geological overview
ophicarbonates also display different colours depending on the mineral phase present in their matrix. The contact between the ophicarbonate material and the host ultramafic rocks varies from sharp to gradual.
The central part of the ophicarbonate samples drilled from veins and pockets has a slightly schistose granoblastic texture consisting of olivine and calcite (Figure 2-14). The grain size of this olivine varies from small roundish blasts to rather large slightly elongated olivines, locally completely crosscut and overgrown by mesh-textured serpentine veins. Calcite forms a granoblastic matrix A sharp contact separates this dominantly calcite-olivine matrix from domains made up of large, elongated serpentinised olivines accompanied by subparallel magnetite bands, where the amount of calcite decreases drastically. Other ophicarbonates are separated from the host rock by a black wall border composed of elongated diopside blasts or by a border of slightly elongated diopside and magnetite blasts associated with green spinel bands, where carbonate material is absent. 17 Metamorphism "lt*IEillll
3 Metamorphism
3.1 Problem Statement
Because of its geotectonic position, the Chiavenna unit is affected by all the major geological events characterising the tectono-metamorphic evolution of the Central Alps (section 2.1). The distinction between the single events is difficult because their occurrence was concentrated in a relative short time.
The main goals of this study are:
To gather new analytical data to unravel the greenschist to upper amphibolite metamorphic evolution
previously described by Schmutz (1976).
To estimate the thermodynamic conditions characterising the metamorphic evolution.
To identify the main driving force of metamorphism within the region.
The detailed microtextural and petrologic analyses presented in this study partially contrast with previous studies and introduce new elements of discussion related to the late Alpine history of the unit and consequently of the Central Alps. It is proposed that the unit underwent metamorphism at two different stages with different thermodynamic conditions and deformation styles. In the following sections the evidence of the multistage metamorphism are presented. They are organised considering the different rock-chemistries (ultramafic, mafic and calcareous rock) and the observed metamorphic modal and compositional changes. These changes are described by increasing metamorphic grade and decreasing distance from the Gruf contact, where the temperatures of the thermal event reached their peak conditions.
3.2 Metamorphism in ultramafic rocks
3.2.1 Temperature conditions
The ultramafics constitute the best chemical system demonstrating with modal changes the metamorphic temperature range observed in the Chiavenna Unit. The area comprised between 750-756 longitude and 129-134 latitude (Swiss geographical co-ordinates) offers the best continuous outcrop exposure. In this area a section is observed with metamorphic grade increasing from the diopside-out isograd in the north to the enstatite-in and spinel-in isograds in the south towards the Gruf contact (Figure 3-1).
In the North, antigorite, olivine, magnetite, tremolite and chlorite form the main mineralogical assemblage, locally associated with tremolite and olivine blasts replacing former diopside. Moving towards the south, talc-bearing rocks progressively substitute the antigorite-dominant assemblages. Further towards the south, the talc-assemblages are replaced by anthophyllite- or cummingtonite-assemblages 18 Metamorphism
associated with olivine, chlorite, magnetite and tremolite. Outcrops close to the Gruf contact contain enstatite and metamorphic Al-rich spinel. Figure 3-2 shows the simple petrogenetic CMSH system with the stability fields of these assemblages, the observed phase-changes in the metamorphic rocks and their feedback with distance from the Gruf contact.
Figure 3-1: Isograds in metapendotites of the Chiavenna unit between the Swiss geographical co¬ ordinates 750-756 longitude and 129-134 latitude (geologic map modified after Schmutz (1976)).
The low-pressure path (p < 10 kbar) of the reactions in Figure 3-2 is extrapolated (dashed line), but all modal changes are essentially temperature-controlled reactions where pressure plays only a secondary role. The temperature range of interest is comprised between values of approximately 470°C and 700°C, corresponding to the diopside-out and the enstatite-in reactions, respectively.
Figure 3-2: Phase-diagram for the ultramafic CMSH system. Reaction curves after Ulmer and Trommsdorff, (1999). The increase of metamorphic grade from the North (Tambo nappe) to the South (Gruf unit) is evidenced by dashed lines representing the extrapolated PT- condition for the reactions.
300 400 500 600 700 800 900 T("C)
The represented reactions are dehydration or hydration reactions depending on the path followed during metamorphism. A comparison between the phase-diagram of Figure 3-2 and the computed phase- diagram in the CFMASH system (Figure 3-3) shows that even if the main topology of the stability fields is 19 Metamorphism
identical, substantial shifts in the reaction-position may be caused by the presence of cations such as Al and Fe.
50 3032' 3-3: 91 Figure Computed phase-diagram P(kbar) Chl+Tr+OI+En - ml perplex for I (program (Connolly, 1990)) I the Chiavenna ultramafic 42- I system Tr+Chl+Di+a+Atg (CFMASH system) assuming water as the 9699 volatile and CaO: FeO:
. only phase MgO: „ Chl+Di+01+Atg+Brc 34k' , Ah03: SiC-2 = 0.09:0.14:1.18:0.05:1 and at Tr+Cbl+OI+Atg Chl+Tr+Oh-Ath 9 6 I I ' < 5 The 6;9 53 3 2-3 4 low pressure conditions (p kbar). 59/"/-/37 i Tr+O+En Chl+Di+OI+Atg Tr+Chl+Tlc+OI+Atg 26- arrow indicates the temperature range of interest for the metapendotites of the Chiavenna unit. 1 8 35 I h Chl+Tlc+OI+Tr
1 0 300 400 500 600 700 800 T(°C)
The phase-diagram of Figure 3-3 is obtained computing the thermodynamic data basis of (Holland and
Powell, 1998) with the computer program perplex (Connolly, 1990), assuming water as the only volatile component, i.e. neglecting the role of CO2, and considering the typical ultramafic bulk-rock composition
(Ca0.Fe0:Mg0:Al203:Si02 = 0.09:0.14:1.18:0 05:1) of a representative serpentinised metaperidotite from the Chiavenna unit.
20 Figure 3-4: Stability fields of different in an iron-free CMASH 18- peridotites system (dashed lines) and in a 16- CFMASH system (solid lines) after (Jenkins, 1980) calculated for enstatite ro .a and forsterite in excess. &12-
Chlorite-pendotrte 3 10' (/> £ 8 Spinel amphibole " pendotite
Tn-spiä* po*En*An»* r lagiociase-«^. '"•--86*Ä- .'*}*'* 'herzollte «
500 600 700 800 900 1000 1100 Temperature (°C)
In the CFMASH-system, chlorite is an important phase over the considered temperature-range.
Chlorite remains stable at temperatures between 650°C (p < 4 kbar) and 700°C (p > 4 kbar), crosscutting the anthophyllite-out reaction. In the Chiavenna peridotites, diopside is completely replaced by tremolite and olivine at temperatures above 470°C while the upper temperature limit is determined by the occurrence of enstatite in a completely anhydrous assemblage. The amount of water involved in these reactions varies between 6.3 and 9.5 in weight %. The diopside-enstatite anhydrous assemblage was never observed in the Chiavenna metapendotites, thus the estimated maximum temperature conditions reached during metamorphism do not exceed 750°C. Compared to the reaction-positions in the general 20 Metamorphism «tiiiiBlli
CMSH petrogenetic grid (Figure 3-2), the computed reactions partially overlap with the above predicted temperature-range except for a shift of the diopside-out reaction towards lower temperatures and minor shifts for the antigorite-out and the talc-out reactions. This temperature range corresponds to greenschist and upper amphibolite to granulite fades conditions (Figure 3-4), whereas the Chiavenna metapendotites compositionally fall within the chlorite peridotite and the spinel-amphibole peridotite stability fields.
3.2.2 Metamorphic evolution
The reactions in Figure 3-3 are reconstructed considering the distribution of the different mineral assemblages in the field and the thermodynamic calculation based on the real bulk-rock composition of the Chiavenna metaperidotites. The boundaries between the different assemblages represent metamorphic reactions where mineral-phases are consumed and/or replaced by other phases. The evolution followed by the rocks during metamorphism is identified by phase-relics of "previous" stable assemblages, by their composition and by textural features related to former phase-equilibria.
The metamorphic reactions documented in the field and summarised in Figure 3-3, are listed below, starting over with the low temperature reactions and ending with the enstatite in reaction occurring near the contact with the Gruf complex.
Diopside-out reaction: Along this metamorphic boundary, diopside disappears and is completely
replaced by tremolite and olivine. These tremolite and olivine pseudomorphs preserve the granular
shape of the rotated clinopyroxene and the typical pyroxene cleavage, locally replaced by magnetite
bands (Figure 3-5). Aggregates of tremolite, olivine and magnetite are also found in schistose chlorite
bearing assemblages. The reaction describing the disappearance of diopside is:
antigorite + diopside
Figure 3-5: Photomicrographs of pseudomorphic tremolite and olivine after diopside. a) Atg+OhTr+Chl+Tlc+Mag assemblage: Aggregate of tremolite and olivine blasts intergrowing with thick fine-grained magnetite-bands (sample BU41, crossed polars), b) Chl+OI+Atg+Mag±Tlc assemblage: Granular shaped tremolite and olivine after clinopyroxene (sample BU6, plane polars).
1 (Trommsdorff and Evans, 1972) 21 Metamorphism
Talc-in reactions: Talc assemblages progressively substitute antigorite or chlorite assemblages, by
reaction (2) or by reaction (3), respectively:
antigorite <-> olivine + talc + chlorite + H2O (2)11
5 chlorite <-> 2 talc + 7 olivine + magnetite +18 H20, (3)'"
Figure 3-6 shows elongated porphyroblastic olivine, embedded in a schistose fine-grained antigorite-
matrix being substituted by talc. The irregular borders of olivine document the instability of this
reacting phase. Several olivine blasts contain small magnetite inclusions. Magnetite is locally
surrounded by small sheets of talc. Similarly, the schistose antigorite is replaced by randomly
oriented talc-sheets. Because of the elevated H2O content of antigorite, large quantities of water
(between 2.8 and 6 % by weight for the considered system) are liberated during the formation of talk,
enhancing a relatively rapid recrystallisation of the rock. Similarly, the chlorite substitution by talc
involves large amounts of water and induces recrystallisation and changes in the composition of
chlorite and spinel (section 3.2.3).
Figure 3-6: Photomicrographs of Atg+Tlc+OI+Tr±Chl+Mag assemblage from the Chiavenna metapendotites (sample BU3): a) elongated porphyroblastic olivine immersed in a schistose fine-grained antigorite-matrix (crossed polars), b) irregular mineral-borders of olivine (plane polars), c) magnetite- inclusions in olivine (plane polars), d) magnetite crystals surrounded by talc, e) antigorite substitution by talc.
=> Talc-out reaction: The talc-breakdown and its substitution with Mg-amphibole occurs gradually. For
this reason the mapped isograd (Figure 3-1) mainly represents a zone where talc progressively
II (Trommsdorff and Evans, 1974) III (Evans and Frost, 1975) 22 Metamorphism
disappears and becomes replaced by Mg-amphiboles (anthophyllite or magnesio-cummingtonite).
Sample BU28 is exposed in this zone, where Mg-amphibole occurs beside olivine, talc and tremolite-
needles. Sample BU28 represents the northernmost occurrence of Mg-amphibole together with
tremolite (Figure 3-7). Electron microprobe (emp) analysis revealed the presence of magnesio-
cummingtonite associated with tremolite. Magnesio-cummingtonite is the monoclinic equivalent of
anthophyllite and unlike anthophyllite it can coexist with tremolite. Anthophyllite and cummingtonite
are produced by the reaction:
4 forsterite + 9 talc <^ 5 anthophyllite + chlorite + H20, (4)IV
which is a dehydration reaction liberating, in the considered system, between 0.5 and 1.9 wt. % of
water.
Figure 3-7. Photomicrographs of Chiavenna metaperitotites containing Mg-amphiboles. a) sample BU28 contains beside tremolite Mg-cummingtonite (crossed polars); b) sample SU76 contains only anthophyllite (crossed polars).
Figure 3-8: Photomicrographs of enstatite bearing high temperature assemblages: a) Coarse-grained enstatite, olivine and chlorite (crossed polars), b) fine grained neoblastic polygonal olivine texture (plane polars).
N (Greenwood, 1963) 23 Metamorphism
=> Enstatite-in reaction: This isograd corresponds to the highest metamorphic conditions that can be
observed in the ultramafic system. Enstatite essentially replaces anthophyllite or Mg-cummingtonite
while chlorite leads to the formation of enstatite and spinel both reactions occur above 650°C:
anthophyllite + forsterite <-» 9 enstatite + H2O (5)v
chlorite <^ forsterite + 2 enstatite + spinel + 4 H20. (6)^
Enstatite is associated with green-spinel, chlorite, olivine and tremolite (Figure 3-8). The enstatite
blasts are well-crystallised and contain relics of Mg-amphibole as well as chlorite and olivine
inclusions. The rock matrix generally consists of olivine, tremolite, chlorite and spinel. The grain size
of the minerals varies between coarse and fine grained, enstatite locally is present as large blasts
overgrowing the other minerals and their micro-fabric, or as small blasts embedded in a matrix formed
by fine grained polygonal olivine blasts. Tremolite is the stable Ca-phase while chlorite and spinel are
the stable Al phases in all analysed thin-sections. These characteristics indicate that the highest
temperature conditions reached during metamorphism did not exceed approximately 700°C and that
the high temperature diopside-in reaction was never reached.
3.2.3 Mineral chemistry
Electron microprobe (emp) analyses were carried out on each mineral phase of the considered ultramafic rocks. Representative mineral compositions are listed in a Table 3-1.
The mineral-phases which characterise the ultramafics recrystallised during Alpine metamorphism.
Mantle relics are absent. The elongated olivine porphyroblasts associated with antigorite and chlorite of the northernmost outcrops are formed during metamorphism. The Xm9 value of olivine varies between 0.87 and 0.92 from north to south towards the Gruf contact. Mantle olivines are usually characterised by high
XMg values of approximately 0.92 and by high Ni contents (between 0.008 and 0.009 cations per formula unit) (Müntener, 1997). Compared to any other mantle phase, olivine constitutes a preferential reservoir for Ni. In equilibrium with antigorite, olivine shows different features. The Ni content of the analysed olivine is significantly lower than in mantle olivine and antigorite.
Figure 3-9 shows the Ni versus Mn distribution for olivine, antigorite and enstatite in antigorite- and enstatite-bearing parageneses. Olivine in antigorite-bearing assemblages displays lower Ni contents than olivine recrystallised in equilibrium with enstatite. The calculated average distribution coefficients (Kd) for
Mn/Mg and Ni/Mg for antigorite/olivine are 0.35 and 0.68, respectively and for the same element pairs for orthopyroxene/olivine 1.86 and 0.22, respectively. These Kd values are in agreement with element ratios for analogue mineral-pairs in spinifex-like rocks of SE Spain (Trommsdorff et al, 1998), where similar Ni and Mn distribution among olivine, antigorite and orthopyroxene are observed.
v (Greenwood, 1963) Vl (Fawcett and Yoder, 1966) 24 Metamorphism
Table 3-1: Representative emp analyses of selected mineral phases characterising the metamorphic ultramafic rocks of the Chiavenna unit (the phases are not necessarily coexisting).
Composition in wt% Mineral Amphibole Olivine Enstatite Spinel Antigorite Talc Chlorite Tremolite AnthoDhvllite Cumminatomte Si02 54 95 56 41 58 29 39 66 58 35 0 00 42 55 6128 28 81 Ti02 017 004 0 05 0 00 0 05 0 53 002 0 02 013 M2O3 3 03 128 0 05 0 01 02 0 23 187 0 02 18 8 Cr203 014 0 06 0 02 0 00 0 06 17 07 018 0 05 1 1
FeaOs 185 155 0 85 - 0 00 5115 0 00 146 31 FeO 0 86 5 59 7 00 1127 5 67 28 29 3 72 0 00 164 MnO 0 07 0 24 0 32 017 015 04 0 08 0 01 0 00 NiO 0 07 011 01 04 0 08 0 58 022 0 25 023 MgO 23 11 3131 2945 48 53 35 91 164 38 47 30 93 32 08 CaO 12 66 034 0 68 0 00 0 01 0 00 002 0 01 0 03 Na20 0 50 0 00 004 0 02 0 01 0 01 001 0 02 0 03
K2O 0 05 0 01 000 0 01 0 01 - 001 0 00 0 06
H2O 218 2 20 2 20 - - - 12 67 4 66 12 45
Total 99 63 9914 99 06 100 09 100 5 99 89 99 83 98 72 9848
Cations (v 1 u ) 15 + A 15 + A 15 + A 3 4 3 10 7 10 Si 7 56 7 68 7 95 0 98 199 000 4 03 394 2 76 Ti 0 02 0 00 0 01 000 0 00 0 02 000 000 0 01 Al 049 0 20 0 01 000 0 01 0 01 021 000 213
Cr 0 01 0 01 000 0 00 0 00 0 51 0 01 000 0 08 Fe3 019 016 0 09 0 00 145 000 0 07 0 22 Fe2 010 064 0 80 0 23 016 0 89 029 000 013 Mn 0 01 0 03 004 0 00 0 00 01 001 0 00 0 00 Ni 0 01 0 01 0 01 0 01 0 00 0 02 002 0 01 0 02 Mg 4 74 6 35 5 99 178 183 0 09 543 2 97 4 61 Ca 187 0 05 0 099 0 00 0 00 0 00 000 0 00 0 00
Na 013 0 00 0 01 0 00 000 0 00 000 0 00 0 01 K 0 01 0 00 0 00 0 00 0 00 000 0 00 0 01 H 200 2 00 2 00 8 00 200 8 00 XMg(Fetot) 094 0 89 0 87 0 88 0 92 0 09 0 95 0 98 0 93 Fe3+/Fe(tot) 0 66 0 20 0 01 000 0 00 100 0 63 Forsterite 0 88 Enstatite 0 91 FeAI204 0 01
FeFe204 0 614 FeCr204 0 246
0 02- Figure 3-9: Nickel versus manganese o o (a.p.f.u.) contents for antigorite, olivine o ;0015- " and enstatite in Antigorite antigorite-olivine o bearing parageneses and in enstatite- Ol in olivine bearing parageneses (atoms per :£ 0 01- sp-pendotites (Müntener, 1997) 18 oxygens, 4 oxygens and 6 oxygens,
Ol in « Ol in respectively). En-pa ragenesis Atg-paragenesis -2.0 005-
Enstatite
00- 0 0 0 002 0 004 O 006 0 008 0 01 Mn (cations p f u )
Systematic Al increase (Figure 3-10) with increasing temperatures characterises the chlorite compositions, while the spinel compositions (Figure 3-11) vary between ferrite, ferrite-chromite and AIMg spinel compositions in agreement with the schema proposed by Evans and Frost (1975). 25 Metamorphism
Figure 3-10: Electron microprobe Enstatite of chlorite from different assemblages composition distances from the Gruf unit showing a Anthophyllite progressive increase in Al content with 2.0- ;ummington assemblages increasing metamorphic grade and decreasing distance from the Gruf (a.p.f.u.). 1.5-
1.0-
1.0 1.5 2.0 2.5 3.0
Al + Fe3+
Metamorphic reactions lead to the formation of distinct amphiboles (Table 3-1) such as tremolite, anthophyllite and magnesio-cummingtonite (section 3.2.2). Magnesio-cummingtonite is an end-member
phase of the cummingtonite-grunerite series defined by Xm9 ratio higher than 0.7 [Xmq = Mg/(Fe2+ + Mg)]
(Leake, 1978). In metamorphosed ultramafics it is optically indistinguishable from tremolite.
Compositionally it is similar to anthophyllite but it has slightly higher Ca contents and a lower number of Al
cations per formula unit (approximately < 0.2) than anthophyllite, even if the analysed metapendotites also contain low Al-anthophyllite (< 0.2 Al c.p.f.u.) similarly to other compositions reported in literature (Deer, et al, 1997). Distinct from anthophyllite, Mg-cummingtonite coexists in nearly all cases with tremolite (Evans
and Medenbach, 1997). Thermodynamic studies (Evans and Ghiorso, 1995) calculated for Mg-enriched
cummingtonite (XMg = 0.85-0.92) compositions such as the studied Mg-cummingtonites the maximum
possible temperature in a FMSH system of 720°C. This temperature is provisionally estimated at about
40°C lower in a CMFH system with Ca-saturated cummingtonite and anthophyllite (Evans, et al. in press).
Figure 3-11: Electron microprobe composition of spinel grains recovered in metapendotites from distinct temperature zones. The compositions are plotted in the Fe3*-Cr-AI (a.p.f.u.) diagram proposed by (Evans and Frost, 1975) for spinel compositions in progressive metamorphism: A) diopside- antigorite-olivine, B) tremolite-antigorite- olivine, C) talc-olivine, D) anthophyllite- olivine, E) enstatite-olivine.
Cr1 0.8 0.6 0.4 0 2 0 26 Metamorphism
3.2.4 Microtextural features
The main differences between the microtextural characteristics of the low grade and the high-grade metamorphic ultramafitites may be summarised as following:
In the northern outcrops where antigorite-bearing assemblages are still preserved, it is possible to
recognise a well developed E-W schistosity (S2) marked by the elongation of antigorite, large olivine
blasts and chlorite. Aggregates of tremolite and olivine preserve the granular shape typical of
pyroxene as well as its rotated texture within the antigorite-matrix.
The deformed rotated magnetite bands associated with the tremolite and olivine aggregates also
suggests a pre-deformation growth and may represent an old schistosity (S1) (prior to the main E-W
schistosity), whereas the magnetite bands within the rock-matrix, similar to the elongated olivine and
antigorite blasts, are parallel to the S2 schistosity. The tremolite and olivine aggregates replacing
clinopyroxene seem not to be influenced by these events, as documented by the polygonal texture
(Figure 3-5) of these olivine and tremolite aggregates. The tremolite-in reaction is likely to have
occurred after the deformation-phase (D3) that folded the antigorite, olivine, chlorite and magnetite
schistosity (S2) as shown by the presence of randomly oriented tremolite needles (Figure 3-12).
Figure 3-12: Photomicrograph of folded antigorite, olivine and magnetite associated to randomly tremolite blasts in ultramafic rocks (BU3B crossed polars).
The substitution of low grade metamorphic phases by high grade metamorphic phases during
increasing temperature is accompanied by the release of relatively large amounts of water, generally
accelerating reaction-kinetics and leading to the growth of new minerals such as talc, Mg-amphibole
and enstatite. The rapid growth of these minerals induced textural changes causing the substitution of
the former schistose textures by coarse-grained granoblastic textures. In fact, the formation of
metamorphic high-grade minerals such as enstatite is normally associated with coarse-grained
textures where fan-shaped large enstatite together with randomly dispersed Mg-amphibole and
tremolite blasts dominate the whole rock microfabric. By contrast, at lower metamorphic grade, the
growth of new minerals such as talc occurs through replacement of antigorite and olivine by fine¬
grained talc sheets overgrowing antigorite and olivine but not necessarily their schistose texture. 27 Metamorphism
Therefore, talc may maintain an erroneous vestige of planar fabric linked to an older deformation
phase involving antigorite and olivine.
Preserved from textural changes are those minerals such as olivine, chlorite and magnetite, which
are stable through the entire temperature range considered by the above listed reactions.
Consequently, these minerals are preserved from textural changes and maintain their textural
features also at high-grade metamorphic conditions.
3.2.5 Interpretation
In the metaultramafic rocks of the Chiavenna unit, higher-grade assemblages progressively substitute low-grade metamorphic phase-assemblages. A broad area characterised by talc assemblages and relatively narrow zones characterised by the presence of Mg-amphibole and enstatite can be distinguished. Similar metamorphic zones were described for contact-metamorphosed serpentinised peridotites of the Malenco unit and have been regarded as common features of ultramafic contact aureoles (Trommsdorff and Evans, 1972; Evans, 1982).
Microtextures in the Chiavenna metapendotites suggest the occurrence of two distinct metamorphic events, characterised by different thermodynamic conditions.
Evidence of a first dynamic event are:
An E-W antigorite, olivine, magnetite, diopside and chlorite schistosity.
The presence of rotated diopside pseudomorphs, completely replaced by tremolite and olivine
aggregates.
The occurrence of deformed and elongated magnetite bands, large elongated olivine blasts and
tremolite and olivine aggregates.
A second temperature-controlled event is characterised by:
The growth of high-temperature minerals such as talc, anthophyllite, magnesio-cummingtonite and
enstatite.
Systematic changes of single mineral compositions such as the Al content in chlorite.
Textural features such as the development of coarse-grained granoblastic textures suggest a strong
influence of temperature and fluid on recrystallisation and only minor effects of pressure and
deformation on the formation of the new mineral-phases.
Relationship among the mineral-phases indicating a younger age for this near static thermal
metamorphism compared to the dynamic event. The thermal overprint occurred after the deformation-
phase (D2) and probably after the deformation phase D3, which folded the main S2-schistosity.
The estimated minimum temperature conditions for the Chiavenna ultramafics are between 470°C and
500°C, which are slightly below the temperature values (470-550°C) estimated for the neighbouring 28 Metamorphism
southern part of the Tambo nappe (Baudin and Marquer, 1993, Marquer et al, 1994). The temperature range estimated in this study corresponds to the diopside-out and the antigorite-out reactions and is interpreted as the temperature conditions acting during the synkinematic greenschist faciès metamorphism which accompanied the formation of the E-W antigorite and chlorite rock-schistosity described above. Towards the south, close to the Gruf unit, the schistosity is progressively overprinted by a near static thermal event, which is responsible for the growth of higher temperature-phases such as talc,
Mg-amphiboles, enstatite and spinel. This event is characterised by a steep temperature gradient with temperatures increasing from less than 500°C to approximately 700°C across the 2.5 and 2 kilometres wide unit in the west and east, respectively.
3.3 Metamorphic evolution of the mafic rocks
Unlike the ultramafic system, the mafic rocks are petrographically less sensitive, over the given temperature interval, to temperature changes. Most of the reactions occurring in the mafic system are multivariant, thus they do not involve the appearance or disappearance of phases, but they display a high variance of solid solution products in minerals such as amphibole and plagioclase. The mineral- assemblages occurring along the N-S section, can be distinguished in three groups: epidote-bearing, hornblende-bearing and clinopyroxene-bearing assemblages.
3.3.1 Mineral assemblages
A complete description of the mineral assemblages characterising the amphibolites and their textural relationships is presented in section 2.2.2 of this study.
Table 3-2: Mineral assemblages of the Chiavenna amphibolites listed according to their metamorphic grade and their distances from the Gruf unit.
Mineral assemblages TAMBC)NAPPE Greenschist to Ab+Act/Hbl+Ep/Czo+Bt+Tit+Ilm±Qtz±Rt±Zrn±Ap±Chl A/c RTH ( > < 1 1 > 2450 m Amphibolite faciès Metamorphic index minerals Ep-group, Amph-composition a) o
Mineral assemblages LU ai to I ' Intermediate Pl+Hbl±Ep+Tit+Ilm±Qtz±Rt±Zrn±Ap±Chl o « JO u Amphibolite faciès Metamorphic index minerals IE o 900 m a 'oi Hbl- and Pl-compositions, E ro < a.
( Mineral assemblages a> -a Upper Amphibolite Pl+Hbl+Di+Tit+Ilm±Qtz±Rt±Zrn±Chl
Table 3-2 summarises the observed mineral assemblages according to their metamorphic grade and their
distances from the Gruf unit. With increasing temperature the epidote-bearing assemblages of the
northern outcrops are progressively substituted by hornblende-bearing assemblages through the reaction:
Ep + Hbli + Ab + Qtz <-> Hbl2 + PI + Qtz (7)VI1.
Diopside is produced through reaction consumming hornblende (Figure 3-13) as documented by the
amphibole inclusions within diopside. Several reactions forming clinopyroxene starting from hornblende
paragenesis are described in the literature (Spear, 1981; Russ-Nabelek, 1989):
Hbh + Ph <-> Hbl2 + Pl2 + Cpx ± Fe-Ti oxides + H20 (8)VI",
0.203 Amph, + 0.193 PI <-> 0.155 Amp2 + 1.135 Qtz + 0.063 Cpx + 0.048 H20 (9)IX
0.646 Amph2 + 0.101 Qtz + 0.03 PI <-> 0.606 Amp3 + 0.227 Cpx + 0.04 H20 (10)x.
Reaction (8) is a general reaction describing the first appearance of clinopyroxene after amphibole and
plagioclase. Reactions (9) and (10) involve quartz and operate over a wide temperature range. The
majority of the amphibolites of the Chiavenna unit contain minor amounts of quartz in their mineral-
assemblages. Amphibole and plagioclase are the mineral phases stable over the entire temperature range
characterising the metamorphic path of the Chiavenna amphibolites.
Figure 3-13: Photomicrograph of diopside blast enclosing amphibole in amphibolites (plane-polars).
3.3.2 Metamorphic related Compositional Changes in amphibole and plagioclase
The first appearance of clinopyroxene is preceded by a series of exchange-reactions modifying the amphibole composition and occurring over a widespread temperature range. Early experimental
studies on the stability of amphibole in rocks of tholeiitic compositions (Spear, 1981) showed that during
its breakdown reaction amphibole composition changes, becoming enriched in AI, Na, K and Ti and
v» (Apted and Liou, 1983) »(Spear, 1981) lx (Russ-Nabelek, 1989) x (Russ-Nabelek, 1989) 30 Metamorphism
depleted in Si. These observations are in agreement with petrological studies (Ferry, 1984; Russ-Nabelek,
1989) on amphibole reactions occurring at high-temperatures and in contact metamorphosed basic rocks.
The compositional changes are accompanied by a decrease in proportion of amphibole compared to the
other phases with increasing metamorphic grade as well as by small compositional variations in phases
other than amphibole (Russ-Nabelek, 1989).
Electron microprobe analyses of selected amphibole compositions carried out on several amphibolite
samples from the Chiavenna area (Table 3-2) showed that:
Amphibole solid solution phases are present in the entire amphibolite body of the Chiavenna unit and
are accompanied by a decrease in proportion compared to the other phases.
The compositional range of the measured amphiboles varies between actinolite and magnesio-
homblende and rarely pargasite (Table 3-3 and Figure 3-14).
Compositional changes in the mineral phases other than amphibole are limited to plagioclase.
Table 3-3: Representative emp chemical analyses of amphiboles from the Chiavenna mafic rocks.
Amphibole in Amphibolites (wt.%)
RM1 RM1 RM13 SM44 KM?R SMfin
SiO> 51.47 48.71 44.92 38.52 45.49 42.73 Ti02 0.22 0.27 0.44 1.01 1.22 0.95 AkOs 5.3 7.44 8.99 15.22 10.22 11.35 Fe?Oi 3.29 2.98 1.97 1.55 4.33 4.59 FeO 9.25 10.62 14.83 14.23 8.30 15.4 MnO 0.24 0.22 0.29 0.26 0.22 0.22 MqO 15.41 13.84 9.83 9.12 13.06 8.65 CaO 12.36 12.34 11.54 11.9 11.53 11.87 Na20 0.81 1.12 1.24 1.97 1.43 1.26 K,0 0.17 0.22 0.39 1.87 0.18 0.85 H20 2.1 2.06 1.95 1.94 2.04 1.99
Total 100.59 99.85 96.4 97.57 98.03 99.9 Cations (p.f.u. l CALCULATED ON THE BASES OF 23 OXYGENS Namp or Ramp norm with (Fe3+/Fetot) = 0.3
Si 7.33 7.07 6.9 5.96 6 697 6.45 Ti 0.02 0.03 0.05 0.12 0.13 0.11
Al 0.89 1.27 1.63 2.77 1.77 2.02 Fe3 0.35 0.33 0.23 0.18 0.48 0.52 Fe2 1.10 1.29 1.91 1.84 1.02 1.94 Mn 0.03 0.03 0.04 0.03 0.03 0.03
Ma 3.27 2.99 2.25 2.10 1.87 1.94 Ca 1.89 1.92 1.9 1.97 1.82 1.92 Na 0.22 0.31 0.37 0.59 0.41 0.37 K 0.03 0.04 0.08 0.37 0.03 0.16
H 1.99 2.00 2.00 2.00 2.00 2.00
Fe3+/Fe(tott 0.24 0.2 0.11 0.09 0.32 0.21 Tschermaks 0.62 0.72 0.86 1.14 1.22 1.19 Edenite 0.15 0.28 0.35 0.93 0.26 0.45 Plaaioclase 0.11 0.08 0.1 0.03 0.18 0.08 Al(IV) 0.67 0.93 1.1 2.04 1.30 1.56 Airvn 0 22 0.34 0.53 0.73 0.47 0,46
3.3.2.1 Amphibole breakdown reactions
The chemical evolution of these amphiboles is linked through exchange reactions to metamorphism and may be summarised into a set of amphibole-component reactions as defined in Thompson et al.,
(1982). According to this schema, the hornblende composition is described in terms of additive component tremolite and exchange components Tschermak, edenite and plagioclase: 31 Metamorphism !>mws
tremolite component (tr): Ca2Mg5Sis022(0H)2
- Tschermak component (Ts): AbMg.iSLi
edenite component (ed): NaAISUD-i
plagioclase component (pi): NaSiCa.iALi.
These components are used to describe the hornblende breakdown reaction in relation with other phases
such as clinopyroxene, albite, anorthite and orthopyroxene, as a sum of single tremolite, Tschermak and
edenite component-reactions:
Ca2Mg5Si8022(0H)2 (Tr) = 2 CaMgSi206 (Di) + 1.5 Mg2Si206 (En) + Si02 (Qtz) + H20 (11 )XI
CaAI2Si208 (An) = AI2Mg.iSLi (Ts) + CaMgSi206 (Di) + Si02 (12)XI1
NaAISi308 (Ab) = NaAISLi (ed) + 4 Si02 (13)XI".
I.U Figure 3-14: Compositional fields of hornblende showing how the amphibole 0.8- composition controls its breakdown products. -~- •--._. Black diamonds indicate the edenite versus 0O.6- Tschermak components of typical hornblende a) :> compositions of selected amphibolite samples 'S 0.4- from the Chiavenna unit.
0.2-
""---- + Qte\* + produced ^^—-^ I I I I 0.25 0.50 0.75 1.00 1.25 1.5C ts-vector
Tremolite, edenite and Tschermak components may also be used to graphically represent, the
hornblende breakdown reactions and the relationship between the hornblende composition and its
breakdown products (Figure 3-14). The diagram shows the edenite and Tschermak compositions of
hornblende from selected amphibolite samples situated at different distances from the Gruf contact. The
solid line represents the locus of hornblende compositions in which the reactions 11 and 12 are balanced
for silica, that is quartz is either produced or consumed by the breakdown of amphibole. The breakdown of
amphiboles plotting below the solid line will lead to the production of quartz, while compositions above the
line will consume quartz (Tracy and Frost, 1991). The majority of the analysed amphiboles plot above the
indicated line, thus consuming quartz during their breakdown reactions.
3.3.2.2 Amphibole compositional changes
Electron microprobe analyses of amphiboles in the Chiavenna amphibolites reveal an increase in Al
content with increasing metamorphic grade positively correlating with amphibole components such as AIVI,
XI (Thompson et al., 1982) XII (Thompson et al., 1982) Xl" (Thompson et al., 1982) 32 Metamorphism
Ti, Fe3+ and Na (Figure 3-15). The systematics of the chemical variations observed in the Chiavenna
amphibolites may be summarised as follows:
Systematic increase of Al in the octahedral site (AIVI) with increasing AIIV along a trend subparallel to
a pure Tschermak-substitution (slope 1.0).
Ti-increase with increasing temperature along a trend distinct from a pure Ti-Tschermak exchange
(slope 0.5).
Positive correlation between the M2 site occupancies of Fe3+, Ti and AIVI and the tetrahedrally co¬
ordinated AIIV, which is a combination of several substitution vectors such as ferri-Tschermak
(AIFe3+Mg.iSi-i), Tschermak and Ti-Tschermak (TiAl2Mg.iSi.2) substitutions.
a) « 0 o°«& / / *$ / • 0.6 - / A % Ô 0.15- **! <& • $/ o •* - #/ cOo 0.4 F 0.1- •• ° < / ! ^ / o
0.2 - 0.05-
/ /
0.0 - i i i I I o.o- I I I 1 0.0 0.5 1.0 1.5 20 25 0.0 0.5 1.0 1.5 20 25
AI'V AIIV
0.8- c) •
1.25-
*A# <# * 0.6- y 1.00- > < < + 0.75- F «04- si A, •SJfrp* + ß^h y
a BM1 1850 m • SM60450 m oSM28920m oSM44250m
* BM13 900 m A SM71 at the contact
Figure 3-15: Variation of site occupancies and AIIV (a.p.f.u) content in amphiboles from the Chiavenna unit. Different symbols are chosen for samples at different temperature conditions and from different distances to the Gruf contact, a) AIVI versus Afv shows a systematic correlation with increasing metamorphic grade. The line of slope 1.0 corresponds to pure Tschermak substitution from tremolite at the origin, b) Ti versus AIN, line with slope 0.5 is the Ti- Tschermak substitution, c) Fe3+2Ti+Alvl versus Allv displaying the M2-site occupancy in amphibole, the two lines with slope 1.0 and 0.5, respectively indicate the Ferri-Tschermak and Tschermak and the Ti-Tschermak substitutions, d) Na content in the A- site occupancy versus Altv. The substitution vectors to the edenite and pargasite compositions and the slope of trend obtained for the Chiavenna amphiboles are outlined. 33 Metamorphism
Systematic increase of Na in the A-site occupancy with increasing AIIV. The amphiboles plot along a
trend, which is subparallel to the pargasite vector, indicating that the charge balance is obtained
combining a Tschermak with an edenite substitution. This results in a positive correlation between the
edenite exchange component and AIIV (Figure 3-16), suggesting that edenite is the dominant
component responsible for AIIV and Na enrichment, whereas the plagioclase exchange (NaSiCa.iALi)
is less important.
These amphibole composition changes are directly or indirectly linked to the temperature increase in the studied area. The amphibole breakdown occurs through a dehydration process, where continuous reactions enrich the remaining amphiboles with AI, Fe3, Na and Ti components, whereas the typical pyroxene components (QUAD-components) Ca, Mg and Fe become unavailable. In the case of the
Chiavenna amphibolites, the correlation between amphibole composition and distance from the Gruf contact is evident, and the amphibole is progressively enriched in AIIV, edenite, Ti, AIVI and Fe (Figure
3-16). Sample SM71 represents the only exception, being very close to the Gruf contact but displaying relatively low AIIV and Na contents. This feature may be due to the presence of calcite-marbles directly adjacent to the amphibolite outcrop, enhancing CO2 metasomatism and Na-depletion in the mafic rocks, as observed in amphibolites from other localities (Frost and Frost, 1989).
125- Figure 3-16: Edenite and plagioclase GBM1 versus AIIV in BM4 • SM60 components (a.p.f.u) 100- Edenite is the dominant O SM28 « BM13 amphiboles.
SM80 O SM44 component responsible for AIIV and Na il 075J ASM71 enrichment, whereas the plagioclase 0) o is less q o> ^ exchange (NaSiCa.iALi) important. u^osoH
025- cUD e$°o plagioclase /component Htt^B o* " * 000- 3l i l^ 00 05 10 15 20 25 Al IV
3.3.2.3 Plagioclase compositional changes
Similar to the amphiboles, plagioclase compositions change along the N-S section of the Chiavenna amphibolites (Figure 3-17). The compositional variations of the analysed plagioclases apparently contradict the Ca-enrichment trend normally observed at high metamorphic grade. In the Chiavenna unit, samples closer to the Gruf contact not necessarily display higher anorthite contents than samples collected in lower temperature zones. This is due to the presence of clinopyroxene in the same amphibolite sample. In fact, clinopyroxene-bearing samples (i.e. samples: SM60 and BM13) display anorthite contents comparable to plagioclase compositions of epidote-bearing assemblages (sample 34 Metamorphism
BM4). By contrast, plagioclases from amphibole- or epidote-assemblages systematically increase their anorthite content with increasing temperature conditions and with decreasing distance from the Gruf contact. This indicates that the appearance of clinopyroxene not only involves the breakdown of hornblende but also affects the Ca-concentration in plagioclase. Plagioclase decreases its anorthite content as described in reaction (12), causing an inversion of the Ca enrichment trend in plagioclase as soon as clinopyroxene appears and similar to amphibole it increases its non-QUAD components.
* BM13 û SM71 0.8- o BM4 a) SM28 o SM44 • SM60 SM80 0.6- \ 0.4- >
0.2-
0.0- —i r 0.0 0.2 0.4 0.6 0.8 xAn
Figure 3-17: Plagioclase compositions of amphibolites samples from the Chiavenna unit: a) Na versus anorthite content, b) (Al + 27/ + Fe3+) in amphibole versus anorthite content (atoms p.f.u.).
3.3.3 Summary
Microscopic and field observations of amphibolite samples from the Chiavenna unit show that the temperature conditions increase along a prograde metamorphic path where diopside-bearing assemblages replace hornblende assemblages without any significant change in the texture of the rock.
The textural and petrographical differences between samples from the regional greenschist fades zones and samples outcropping in metamorphic higher grade zones are not significant because most of the reactions occurring in the mafic system are multivariat and do not imply the crystallisation of new minerals.
The study of reactions involving amphibole solid solutions provides a set of compositional data documenting a systematic amphibole compositional change in response to the temperature increase. The observed compositional changes correlate with distance from the Gruf contact. Amphibolites close to the contact display the highest AIIV, Ti and Na, contents and the highest edenite component. Similarly, the plagioclase chemistry is characterised by systematic changes as a function of distance from the Gruf contact. In this case however, the presence of clinopyroxene may considerably reduce the anorthite content of plagioclase. 35 Metamorphism
3.4 Metamorphosed calcareous rocks
The metamorphosed calcareous rocks are represented by calcite marbles outcropping in a few
localities. The scarcity of these rocks makes it difficult to define isograds in the field. However, it is
possible to distinguish between metamorphic low-grade and high-grade assemblages.
Figure 3-18: Photomicrographs of metamorphic phase-equilibria within the calcite-marbles of the Chiavenna unit: a) scapolite and calcite inclusions in garnet, b) scapolite surrounded by plagioclase.
Calcite, quartz and minor amounts of phlogopite, amphibole, plagioclase and diopside characterise the low-grade assemblages dominated by a schistose granoblastic microfabric where especially quartz is strongly orientated parallel to the main E-W foliation. Close to the Gruf contact, the metamorphic high- grade assemblages display a large number of mineral-phases (Figure 3-18), which is the result of the metamorphic high temperature conditions as well as of the compositional differences among the rocks.
The study of the textural relationship between the different phases and their compositions supplies information on the temperature and pressure conditions as well as on the composition of the fluid phase during metamorphism. The calcite-marbles are chemically described in the simplified system CaO-Ab03-
SiÛ2(H20-C02) (Figure 3-19), in which the compositions of the main mineral phases are plotted on the
Ca0-Ab03-H20-C02 surface, projecting data from the Si02 apex. The real system would comprise also other components such as K2O, and MgO, but the metamorphic important phases can easily be represented using the quaternary system Ca0-Al203-H20-C02.
CO2 Cc Cao Figure 3-19: Chemographic representation of the + S1O2 main mineral phases forming the metamorphic K20 high-grade marbles of the Chiavenna unit. The MgO chemical data are displayed in the quaternary CaO-Ah03-H20-C02 system, projecting the /ys compositions from the quartz vertex. H20 AI2O3 36 Metamorphism tffî'ÊtâSMM
3.4.1 Phase stability calculations
The metamorphic assemblage formed by calcite, scapolite, grossular, wollastonite and plagioclase characterising some calcite marbles near the Gruf unit formed at definite temperature and pressure conditions and under the influence of a specific fluid composition. On a PT-diagram, this 5-phases mineral assemblage is represented by an invariant point of fixed fluid chemistry and whose position is controlled by the composition of each phase. A phase diagram was computed with the thermodynamic calculations program perplex written by Connolly (1990). The calculations were performed applying the thermodynamic data set of Holland and Powell (1998), the activity for scapolite indicated by Oterdoom,
(1980) and the mineral compositions as measured by electron microprobe analysis and listed in Table 3-4.
The results are shown in a PTXCO2 diagram (Figure 3-20) where the observed five mineral assemblage falls exactly at the invariant point. The position of this invariant point is intimately coupled with the specific composition of each phase.
CC^ cc Sep Grs Cc •—•-• •
Cc + Grs + Wo + Sep + An
639 661 683 705 T(°C)
Figure 3-20: PTXCO2 diagram for the metamorphic high-grade assemblage calcite, scapolite, anorthite, wollastonite, grossular and quartz using program perplex (Connolly, 1990), the thermodynamic data set for mineral phases of Holland and Powell (1998) and the activity for scapolite given in Oterdoom (1980). The figure shows the reaction-topology for the specific mineral composition of the calcite-marbles and the stable mineral phases in a binary system for each PT field.
The invariant point is determined by the topology of 4 main reactions: Metamorphism
Calcite (Cc) + Anorthite (An) <-» Grossular (Grs) (14)XIV
Calcite <-» Wollastonite (Wo) (15)
Anorthite + Calcite <-> Scapolite (Sep) (16)XVH
Wollastonite + Anorthite <-> Scapolite (17)"
Anorthite + Wollastonite <-> Grossular (18)XIX
The reaction topology depends on pressure as well as on mineral and fluid composition. T-XCO2 diagrams
(Figure 3-21) for different pressure conditions show how the reaction topology for the same chemical system is sensitive to pressure changes.
1 1 1 1 1 1 1 1 1 T(C|) a) T(Ci)
847 < - 847 + Q. Ü CO
767 - 767
687 {2Ü^^ Sep 687 An + Cc . — -y-—And=S|11- - ^^--Z-^-Grs = An + \Ato X(1) 607 - 607
/-~J^Cc = Wb
_
/r^-Cc + An = Grs
lli 1 1 1 1 1 ' 1 ' Fi97 527 0.0 0.2 0.4 0.6 0.8 1.0 0.0 xco2 xco2
Figure 3-21: T- XCO2 diagrams computed at different pressures using the program perplex (Connolly, 1990) based on the thermodynamic data of Holland and Powell (1998) and scapolite activity of Oterdoom, (1980) and specific mineral compositions: a) pressure 3 kbar, b) pressure 5 kbar. The positions of the univariant points on the diagrams and the fluid composition rapidly change with increasing pressure. The invariant point characterising the five minerals equilibrium of the calcite marbles indicates the temperature and pressure conditions and fluid composition at which the invariant points (1), (2) and (3) would collapse together in one single point represented in Figure 3-20.
xw(Gordon and Greenwood, 1971) »(Greenwood, 1967) m Scapolite forms solid solution products between the endmember phases marialite (Na4[Al3Si9024]CI,C03,S04) and meionite (Ca4[AleSi6024] CI.CO3.SO4). »»(Goldsmith and Newton, 1977) »'»(Newton and Goldsmith, 1975) xlx(Goldsmith and Newton, 1977) 38 Metamorphism
3.4.1.1 Implications
The results exposed above indicate that the equilibrium of the described phases is reached at temperature and pressure values of approximately 670°C and 4.3 kbar and for a fluid composition of 0.17
XCO2. Despite the fact that the available thermodynamic data on scapolite are still insufficiently precise, the calcite marbles provide pressure and temperature estimates which are in agreement with the temperature estimates obtained for the ultramafic system (section 3.2.1) and the pressure estimate computed for an independent tonalitic system (section 3.5).
Table 3-4: Representative emp analyses on selected solid solution phases of metamorphic high-grade calcite marbles.
REPRESENTATIVE COMPOSITIONS OF SOLID SOLUTION PHASES (WT.%) Cations per formula unit Scapolite Plagioclase Garnet Scapolite Plagioclasio Garnet S1O2 43.54 61.77 38.25 Si 6.79 2.77 2.92 T1O2 0.01 0.02 0.07 Ti 0.00 0.00 0.00 AI2O3 28.33 24.29 21.08 Al 5.21 1.29 1.89
P2O5 0.22 - 0.38 P 0.03 - 0.02
Fe203 - 0.00 3.16 Fe3 - 0.00 0.18 FeO 0.03 0.22 0.00 Fe2 0.00 0.01 0.00 MnO 0.00 0.00 0.71 Mn 0.00 0.00 0.05 MgO 0.01 0.00 0.11 Mg 0.00 0.00 0.01 CaO 19.35 5.58 35.72 Ca 3.23 0.27 2.92 Na20 2.79 7.56 0.02 Na 0.84 0.66 0.00 K2O 0.17 0.08 0.01 K 0.03 0 00 0.00 S03 0.03 0.01 0.00 S 0.003 0.00 0.00
CI 0.13 - 0.00 CI 0.03 - 0.00 Total 94.60 99.53 99.51
Meionite 0.79 An 0.29 Grossulana 0.88 Equiv. An 0.74 Ab 0.71 Andradite 0.09
3.5 Estimated pressure conditions in the Chiavenna unit
The near static thermal metamorphic Alpine event documented in the Chiavenna unit is well constrained in its temperature evolution (section 3.2.1), but none of the chemical systems, except for the calcite marbles, yields substantial information which could be used for the pressure conditions.
3.5.1 Ultramafic and carbonate rocks
The absence of typical low-pressure minerals such as plagioclase and cordierite hinders precise determination of pressure for the ultramafic rocks. Nevertheless, the chlorite and anthophyllite breakdown reactions seem to fix at least one point in the calculated phase-diagram (Figure 3-3 in section 3.2.1) for the ultramafic system. The two breakdown-reactions (5) and (6) intersect at a pressure of approximately 4 kbar, inverting the breakdown-sequence. For pressures below 4 kbar, the chlorite-breakdown reaction precedes the anthophyllite-out reaction, whereas at pressures above 4 kbar anthophyllite is consumed before chlorite. In the Chiavenna ultramafics, chlorite is consumed after anthophyllite, hence the pressure conditions during the considered high temperature thermal event should not be lower than approximately 39 Metamorphism
4 kbar. Pressure values of approximately 4 kbar are also obtained considering the five-phases equilibrium of the metamorphic high-grade calcite marbles (section 3.4.1).
3.5.2 Tonalitic system
Independent pressure computations for tonalitic intrusions crosscutting the Chiavenna amphibolites
(section 2.2.2) supply pressure estimates for the intrusion-stage of the tonalité dikes and indirectly for the amphibolites at the moment of the intrusion. Field relationships between the schistose amphibolites and the tonalité dikes show that the magmatic dikes are syn- to post-S2 schistosity. Microtextural evidence indicates that the overprinting thermal event occurred after the E-W syn-collision extension phase (D2) responsible for the S2 foliation (section 3.2.5). Consequently, the pressure conditions affecting the amphibolite host-rock at the time of intrusion are identical to the pressure conditions recovered in the tonalitic rocks. Therefore, it is possible to estimate the pressure affecting the Chiavenna amphibolites immediately before or at the beginning of the "static" thermal event by performing barometric calculations on tonalitic samples.
The tonalitic dikes composition (Table 3-5) and its typical mineral-assemblage consisting of hornblende, biotite, plagioclase, orthoclase, quartz and Fe-Ti oxide allow the application of the geobarometer developed by Hammarstrom and Zen (1986) and successively by Schmidt (1992) for amphibole compositions in calc-alkaline plutons.
Table 3-5: Bulk rock composition (wt.%) of a tonalitic dike from the Chiavenna unit compared with a typical tonalité composition of the Masino-Bregaglia tonalite-granodiorite intrusions (Reusser, 1987).
Tonalité Chiavenna Masino-Bregac Si02 58.24 58.95 Ti02 0.67 0.71 AI2O3 18.62 16.8 Fe203 5.34* 2.36 FeO 0 3.75
MnO 0.07 0.11 MgO 3.79 3.15 CaO 3.93 6.48 Na20 4.49 2.83 K20 2.91 2.42 P205 0.26 0.23 H20 0.95L°' 1.11 Total: 99.27 98.93
* Fe203 is total iron LOI s loss on ionition
The geobarometer is applied to the total Al content (AIT) in hornblendes from intrusions in the pressure range from 2.5 to 13 kbar with a precision of ± 0.6 kbar (Schmidt, 1992). The equation proposed for the link between pressure and AIT in hornblende is:
P = - 3.01 + 4.76 AIT (Schmidt, 1992).
Figure 3-22 shows the Al content of hornblende from tonalitic dikes of the Chiavenna unit and the compositional range of hornblende from different plutonic complexes compiled by (Hammarstrom and 40 Metamorphism
Zen, 1986). The chemical composition of the hornblende speciemens analysed in this study plot within the suitable compositional field indicated by (Hammarstrom and Zen, 1986).
The calculated pressure for the intrusion of the tonalité dikes within the Chiavenna unit, is obtained considering an average AIT value of 1.47 and is equivalent to 4 ± 0.6 kbar.
This value represents a reliable estimate of the pressure condition during the high temperature thermal event which induces the recrystallisation of the entire unit, and is in agreement with the values estimated from the ultramafic and carbonate rocks, as described above.
20- Figure 3-22: Total Al (AIT) versus AF contents (a.p.f.u.) of hornblendes from a tonalitic dike from the Chiavenna 1.5- crosscutting amphibolites unit dotted field = .;;.*";•:• (grey diamonds), compositional range of hornblende from ^_ 1.0- different plutonic complexes (Hammarstrom < compositional-range of and Zen, "" 1986). hornblende data from Hammarstrom&Zen 1987 0.5-
0.0- "i r 0.0 0.5 1.0 1.5 20 2.5 3.0 ait
3.6 Summary and Interpretation
The petrological and compositional analyses of the Chiavenna unit demonstrate that the metamorphic greenschist to upper amphibolite granulite faciès evolution of the unit is linked to Alpine metamorphism.
Pre-Alpine features are hardly preserved (section 2.2.1) and consist of a pre Alpine mantle layering within ultramafic rocks crosscut by a approximately N-S oriented schistosity discordant to the main S2 foliation and interpreted as a possible S1 Alpine schistosity related to the nappe staking during collision.
Petrological and microtextural analyses indicate the coexistence of two distinct metamorphic events.
Both metamorphic events are Tertiary in age (Alpine) but are characterised by distinct deformation styles, temperature conditions and space distribution. The first metamorphic event is distinguished by:
=> in the ultramafics:
A well developed antigorite or chlorite schistosity where large olivine blasts together with magnetite
bands are elongated parallel to the schistosity.
Rotated mineral aggregates consisting of tremolite and olivine represent former diopside involved
in the deformation phase (D2) forming the antigorite schistosity (S2).
Preserved relic magnetite schistosity (S1) within the tremolite and olivine aggregates. 41 Metamorphism
=> In the mafic rocks:
Schistose epidote, actinolite and low Ca plagioclase bearing amphibolites.
=> In the calcite marbles:
The schistosity displayed by the metamorphic low-grade carbonate assemblages.
Outcrops of the northern part of the Chiavenna unit show best relics of this metamorphic event these features become less evident towards the south.
The second metamorphic event overprinted the first dynamic event with mineral assemblages of
progressively increasing temperature. The major metamorphic features of this event in the different rock
systems are:
The occurrence of metamorphic reactions of increasing grade along a prograde path where new
mineral phases are formed. The inferred temperature increase is approximately 200°C, rising from
500°C to approximately 700°C.
The development of coarse-grained granoblastic textures replacing former schistose microfabrics.
Exchange reactions involving solid solution phases such as hornblende and plagioclase. During the
temperature increase the amphiboles of the mafic system modify their compositions towards
progressively higher edenite, AIIV, AIVI and Ti contents while plagioclase increases its anorthite
component until clinopyroxene appears. This compositional replacement of phases occurs without
inducing any significant change in the microfabric.
The temperature evolution of the Chiavenna unit is well constrained, whereas the pressures are not
precisely constrained. Nevertheless, independent pressure estimates obtained for calcite marbles and tonalitic dikes yield values of approximately 4 kbar. This pressure refers to conditions achieved during the
high temperature metamorphism and consequently, represents the pressure acting during the near static thermal event described above.
The structural features characterising the dynamic event are comparable with S2 structures observed
in the neighbouring Tambo nappe (Marquer et al., 1994) and interpreted to be linked to the Tertiary syn- collisional E-W extension phase D2 occurred after the nappe-stacking phase (D1). This study proposes to consider the dynamic event the greenschist faciès metamorphism associate with the E-W syn-collisional deformation phase. For the thermal event, the narrow isograd zones and the temperature increase of approximately 200°C along a distance of less then 2.5 kilometres argue for a local metamorphic event
possibly induced by the rising of a neighbouring hot body. The isograds distribution is subparallel to the
present tectonic contact between the Chiavenna and the Gruf unit and suggests that the emplacemet of the migmatitic Gruf unit may have provided the necessary heat for the thermal overprinting of the southern
Chiavenna unit. This emplacement would be accountable for the petrologic modifications described in this chapter.
43 Dating u-mmimm
4 DATING OF AMPHIBOLES FROM METABASIC ROCKS
4.1 INTRODUCTION
Amphibole has been recognised as a useful monitor of polyphase metamorphism (Laird and Albee,
1981). This feature is due to its compositional sensitivity to change in P, T and bulk rock composition.
According to some authors (Villa et al., 1996), the radiogenic isotopes preserve the record of the conditions under which their host amphibole was formed. Amphiboles, dated by ^Ar/^Ar stepwise-heating method, supported by chemical information and structural observation, offers the possibility to interpret polyphase growth of minerals.
In this study, isotopic compositions of the amphiboles help constrain the timing of metamorphic events in the Chiavenna unit during Alpine metamorphism, as the pre-Alpine metamorphic events in this unit are almost completely obliterated by the intense tectonic-metamorphic transformations of the Alpine phase.
More specifically, amphibole separates from metabasic rocks of the Chiavenna unit were dated using the
^Ar/^Ar method, in order to:
Define the crystallisation age of the different amphibole phases present in the mafic rocks.
Constrain the age of the main metamorphic event correlated with the recrystallisation of the
amphiboles.
Combine the age spectra and the isotopic compositions of the minerals, to define possible
geochronological relationships between amphiboles from surrounding areas.
Correlate, the metamorphic event responsible of the amphibole-recrystallisation with a regional Alpine
event. The Alpine thermal events of major regional importance in this area are represented by the
Masino-Bregaglia intrusions, dated at 32-30 Ma and by the successive Novate granite emplacement
at 25 Ma. Through the comparison of the age data, it may be possible to associate amphibole
recrystallisation with one of these two major events.
4.1 39Ar/40Ar METHOD
Samples consist of the 125-177 pm fraction of amphibole separates, obtained by separation of several grain size fractions using standard high density liquids, magnetic separation (Frantz) and handpicking.
The separates were irradiated in the TRIGA reactor in Pavia and subsequently analysed at the laboratory of the Isotopengeologie, Min. Pet. Institut Universität Bern" using the Ar-Ar stepwise heating method. A detailed description of the analytical procedure can be found in Villa et al. (2000) and Villa (1992).
The technique allows the measurement of signal intensities produced by all five isotopes of argon: 40Ar,
39Ar, 38Ar, 37Ar and 36Ar. The age of the samples is calculated on the base of the ratio 39Ar/40Ar, where the 44 Dating
39/1 amount of atmospheric Ar is corrected by measuring Ar. The irradiation of K, CI and Ca produces Ar,
38Ar and 37Ar, respectively. Thus, from the argon isotope data it is also possible to obtain the relative concentration of these elements.
The 39Ar/40Ar method involves heating the amphibole separates to progressively higher temperatures.
Single temperature-values represents the heating steps at which argon isotopes were measured. During each heating step, the measured gas is that released by breakdown of phase or end-member phase present in the measured separate. To distinguish between the gas contributions of the different phases, it is necessary to support the argon method with microprobe analyses on the same amphibole-specimen. In this way, the microprobe data are used to calibrate the chemical information provided by the Ar isotopic data.
4.2 SAMPLE DESCRIPTION
Samples were chosen to obtain information on the age of the metamorphic overprint that affected the area close to the Gruf unit. Amphibolites crop out in the whole area of the Chiavenna unit and a great
number of outcrops are present close to the Gruf Unit contact. The choice of dating this particular rock type by the Ar/Ar method was favoured by following considerations:
The amphibolites display an increase of metamorphic grade from greenschist to pyroxene
amphibolites fades, which in terms of temperature values, corresponds to a range approximately
between 500 and 700°C. The amphibole blocking temperature is considered to fall between 530°C ±
40°C (Heaman and Parrish, 1988) or between 550-650°C (Villa, 1998). The majority of the samples
chosen in this study underwent complete recrystallisation at temperatures above the quoted
amphibole blocking temperatures and consequently, they may record this event in the isotope
signature. For those samples, which recrystallised incompletely below or close to the blocking-
temperatures, the age-spectra probably display inherited age-components of older events.
o BM4 SM81 0.8- SM28 o SM60 SM44
0.6- SM71
0.4- distance In km Tambo1..7 0.7 0.5 0.25, Gruf • ' nappe contact 0.85 -Tl.15 0.2- o Ik
Figure 4-1: Na (M4) andAI (IV) diagram (a.p.f.u.) versus distance from the Gruf unit. 45 Dating
Chemical analyses of amphiboles of the Chiavenna amphibolites show that the Al content increases
progressively towards the Gruf unit contact in contrast to Na (M4) (Figure 4-1) which is constantly
low. The increase in Al in the amphiboles is also accompanied by the increase of the edenitic
component (Figure 4-2) and is related to an increase in temperature at a relatively constant pressure.
These constraints suggest that the composition of the amphiboles in the Chiavenna unit is mainly
controlled by thermal metamorphism (section 3.3) overprinting older Alpine metamorphic
assemblages and structures, which are widespread in the adjacent nappes.
0.50 1.00 Edenite component
Figure 4-2: Edenite versus Na (M4) component (a.p.f.u) of amphiboles, inside: edenite versus Tschermak components, symbols as in Figure 4-1.
A detailed description of the samples used for analyses is given in Table 4-1, where each sample is characterised by its mineralogical assemblage, structural and textural features and distance from the Gruf
Unit contact.
7ab/e 4-1: Short description of the samples chosen for the 39Arf°Ar methods with their main micro- macroscopic features and distance from the Gruf Unit.
SAMPLE MINERAL ASSEMBLAGE STRUCTURE AND TEXTURE DISTANCE Locality
BM4 AMPH+PL+QZ+EP+BT+TNT+OX FINE GRAINED; FOLIATED 2450M V. Bregaglia SM28 AMPH+PL+QZ+TNT+OX FINE GRAINED; BANDED, FOLIATED 850m V. Schiesone SM81 AMPH+PL+QZ+CHL+BT+TNT COARSE GRAINED, POLYGONAL; MASSIVE 700m V. Schiesone SM60 AMPH+CPX+PL+QZ+TNT+EP+OX FINE TO COARSE GRAINED; FOLIATED 500m V. Schiesone
SM44 AMPH+PL+CPX+TNT+OX±CHL±SCP MEDIUM-COARSE GRAINED; POLYG. FOLIATED 250m V. Schiesone SM71 AMPH+PL+QZ+OX±BT COARSE GRAINED, POLYGONAL; FOLIATED 150m V. Schiesone
Table 4-2 lists the compositions of amphiboles determined by electron microprobe analyses (emp).
Noteworthy, is the chemical difference between sample BM4 collected about 2.5 kilometres from the Gruf contact and the other samples collected closer to the contact. BM4 separates mainly consists of Mg- hornblende, while the amphibole-separates of the other samples display a more edenitic composition.
Amphiboles of SM71 also compositionally differ from the other amphiboles and do not correlate with the edenite-trend. This amphibolite sample was taken from a mylonite zone where the deformed amphibolites are intimately associated with metacarbonates rocks close to the Gruf contact. Metasomatic exchanges 46 Dating
between the two lithologies likely modified the bulk-rock composition and consequently also the mineral composition.
7ab/e 4-2: Representative chemical compositions of amphibole separate gained by electron microprobe measurements.
Amphibole in Amphibolites (wt.%) Sample BM4 SM28 SM81 SM60 SM44 SM71 S1O2 49.150 45.49 44.51 42.73 38.52 46.28 T1O2 0.14 1.22 1.24 0.95 1.01 1.45 AI2O3 7.77 10.22 10.70 11.35 15.22 9.47 Fe203 4.34 4.33 3.99 4.59 1.55 3.68 FeO 9.11 8.30 8.37 15.4 14.23 8.63 MnO 0.65 0.22 0.22 0.22 0.26 0.23 MgO 14.53 13.06 13.32 8.65 9.12 13.37 CaO 10.49 11.53 11.97 11.87 11.9 11.64 Na20 1.05 1.43 1.44 1.26 1.97 1.15 K2O 0.15 0.18 0.38 0.85 1.87 0.40 H2O 2.08 2.04 2.03 1.99 1.94 2.05 Total 99.46 98.03 98.18 99.9 97.57 98.68
Cations (p.f.u ) calculated on the basis of 23 oxvaens FÎAMP (Fe3+/Fe == 0.3) or NamD norm Si 7.09 6.697 6.56 6.45 5.96 6.76 Ti 0.02 0.13 0.14 0.11 0.12 0.16 Al 1.32 1.77 1.86 2.02 2.77 1.63 Fe3 0.47 0.48 0.44 0.52 0.18 0.44 Fe2 1.10 1.02 1.03 1.94 1.84 1.04 Mn 0.08 0.03 0.03 0.03 0.03 0.02 Mg 3.12 1.87 2.92 1.94 2.10 2.95 Ca 1.62 1.82 1.89 1.92 1.97 1.82 Na 0.29 0.41 0.40 0.37 0.59 0.33 K 0.03 0.03 0.07 0.16 0.37 0.07 H 2.00 2.00 2.00 2.00 2.00 2.00
Fe3+/Fe(tot) 0.30 0.32 0.30 0.21 0.09 0.30 Tschermaks 0.91 1.22 1.15 1.19 1.14 1.17 Edenite 0.16 0.26 0.38 0.45 0.93 0.24 Plagioclase 0.16 0.18 0.10 0.08 0.03 0.16 AI(IV) 0.91 1.30 1.44 1.56 2.04 1.24 AI(VI) 0.41 0.47 0.42 0.46 0.73 0.39
4.3 39Ar-40Ar DATA PRESENTATION
Each analysed sample is presented and described together with its argon isotopes data and its
chemical features below. The chemical compositions gained independently by electron microprobe
analyses (Table 4-2) and the back-scatter electron-images (bse) furnish an essential chemical monitor and are indispensable for the interpretation of the irregularly-shaped spectra. The argon-data are
presented starting with spectra of samples recrystallised at higher temperatures, and concluding with
samples from relatively lower temperature-zones outcropping further away from the Gruf contact. The
interpretation of the Ar data follows the approach presented by Belluso et al. (2000) and Villa et al. (2000).
All amphibole separates may contain a minor amount of impurities such as biotite, clay minerals and
plagioclase. The chemical control gained by analysing all argon isotopes allows the age-spectra of these
phases to be distinguished from amphiboles and consequently it is possible to correct the age spectra for these impurities. In this study, the first heating steps are generally characterised by the Ar-degassing of
impurities. Therefore, these steps are not included because they clearly do not contribute to the determination of the amphibole age. 47 Dating
4.3.1 Sample SM71
Sample SM71 represents the amphibolite closest to the Gruf contact and therefore, according to the thermal history of the Chiavenna unit, it recrystallised under high temperature conditions probably above
650°C (section 3.3). The Ar/Ar age spectrum obtained from heating of this sample resulted in an irregularly shaped age-spectrum (Figure 4-3). The two depressions are separated by a marked peak (step
9), characterised by a small Ar-degassing rate, with an apparent age of 68 Ma years. Sixteen heating steps were applied to the sample. Approximately 50% of the 39Ar gas was removed during a restricted temperature range between 1020 and 1060°C given by steps 6 and 7. The apparent ages correlating with these steps are 33.5 ± 0.45 Ma and 35.4 ± 0.33 Ma years, respectively. Remnant gas is released in small aliquots during previous and following steps.
2-3 9 16 Figure 4-3: Age spectrum of sample SM71. J t 50-
45-
100
The Ca/K versus Cl/K diagram diagnoses the presence of different reservoirs (Figure 4-5). This is supported by the microchemical data (Figure 4-5), which clearly show that at least two, and probably 3, amphiboles are required to explain the large variability.
0 070 SM71 0 068- o5 4o c 10 0.066-
0 06 | 064-I 8° 13
0.062-|
03 0.060 J6
0.058 I i 27 29 31 33 35 Ca/K
Figure 4-4: Detail diagram of Cl/K versus Ca/K ratios for sample SM71. The diagram comprises only heating steps (3-16) which are related to the degassing of amphibole. Dating
0.40- a) SM71 b) + EMP analyses 0.35- O Ar isotopic analyses 0.20-
o; 0 30- V 'c 0 15-
010- 0.20- 0
Figure 4-5: a) Cl/K versus Ca/K ratios measured in sample SM71. The measurements were obtained with microprobe and Ar isotopic analyses, b) Edenite versus Cl/K ratios of single amphibole grains of sample SM71.
The comparison between microprobe analyses and Ar isotopic data shows that heating steps between 3 and 16 are chemically comparable with amphibole compositions obtained by electron microprobe analyses.
The age versus Ca/K plot (Figure 4-6) indicates that the heating steps (between 3 and 16) do not plot along a line. The electron microprobe data reveal minor differences in the Cl/K and edenite contents
(Figure 4-5b): It appears clear that the amphiboles must not be considered completely recrystallised, and that the inherited Ar resides in relics of previous pre-mylonitic metamorphic events. Nevertheless,
important information can be extracted from sample SM71 by using the approach of identifying chemically
distinct amphibole reservoirs (Villa et al., 2000 and Belluso et al., 2000). One reservoir has Cl/K ratios of
approximately 0.060 and Ca/K ratios of approximately 28, while another has Cl/K of approximately 0.068
and Ca/K of 34. A possible third reservoir has Cl/K of approximately 0.064 and Ca/K of 34. The first one
corresponds to a step age of 33.5 ± 0.45 Ma.
20 25 30 35 Ca/K
Figure 4-6: a) Detail diagram comprising only amphibole heating steps, as described in text. Age scale in Ma; b) Age versus Ca/K plot of sample SM71. 49 Dating
4.3.2 Sample SM44
The next closest amphibolite to the Gruf contact is sample SM44 (Table 4-1). It forms a breccia within metacarbonate rocks next to a contact zone between amphibolites and metacarbonates. Metasomatic exchange with the abutting metacarbonatic rocks generates noticeable compositional differences in the bulk-rock composition of SM44 presents compared with the typical amphibolitic composition. These differences are especially evident in mineral assemblages that include Ca and Cl-rich minerals such as calcite and scapolite, which are typical of the neighbouring metacarbonates (section 2.2.3).
Figure 4-7: Composite age spectrum of sample SM44.
10 35- 3 7 •= 1 32.5-
5 J= 1 a) 30- , ^ < 4 1
27.5- 1 8] run B run A 25-
20- i i 25 50 75 100
"/o-^Ar recalculated
Because a part of the gas of the first run, A, was lost, a second run, B, with a few leftover handpicked grains was repeated, but with a lower precision and a lower temperature resolution. The age spectrum shown in Figure 4-7 is a composite, obtained by combining the lower temperature steps measured during run B with the extant steps of run A. In all diagrams, the term B corresponds to the sum of the lower temperature steps of run B, while A represents all recovered steps of run A.
a) b)
9 10
0.007- 60- 8j
0.006- 9
"(9 40- o O 0.005- B A
3 20- B A 0.004- 1 4 7J 7
1 2 3 e 5 4 5 r 2 I 6 0- 0.003- I I 1 25 50 75 100 25 50 75 100 "/o^Ar recalculated o/o^Ar recalculated
Figure 4-8: Composite compositional spectra of sample SM44 obtained combining run A with run B: a) composite Ca/K versus %39Ar, b) composite Cl/K versus cumulative %39A.
Although the variation in Cl/K ratios (Figure 4-8) indicates that the degassing-phase is not homogenous in its chemical composition, the resulting age spectrum shows considerably less internal discordance that the preceding one, with step-ages mostly limited to an age interval between approximately 28 and 33 Ma. 50 Dating
4.3.3 Sample SM60
The age spectrum of sample SM60 (Figure 4-9) is comprised of 13 Ar-degassing steps.
Figure 4-9: Age spectrum of 1-3 11-12 t t sample SM60. Age in Ma 50 years. 13 45 10
40-
£,35 < 30
25-
20 ~i i' 1r 1 r1 r 20 40 60 80 100 "/o^Ar
For sample SM60, comparison between electron microprobe analyses and isotopic compositions
permits to distinguish between heating steps related to the Ar-degassing of amphiboles and heating steps
related to other mineral-phases (Figure 4-10). This method allows to focus the Ar-age discussion of
sample SM60 only on those heating steps which are clearly related to the degassing of amphiboles and to
exclude those steps such as 2,11,12 and 13 which display a different chemical composition.
0 05 EMP Argon isotopes 0 04 Non amphiboles
g003i
12 O 0 02
0 01 I I I I I I 8 16 24 32 40 48 56 64 Ca/K
Figure 4-10: a) Comparison between Ar data and independent gained microprobe analyses of sample SM60. In particular, the diagram shows the chemical ratios of the Cl/K versus Ca/K of sample SM60, measured independently with the Ar/Ar method and the electron microprobe, b) Back-scatter electron image of the amphibole-separate SM60. It shows the presence of non-amphibole phases such as plagioclase and titanite, beside amphibole.
Based on the Ca/K ratio of these steps (Figure 4-11), it is possible to distinguish between two groups of steps with different Ca/K ratios. A first group (steps: 3-7) has Ca/K ratios comprised between 12 and 14 and a second group displays (steps: 8-10) Ca/K ratios between 16 and 18. Hence it is possible to distinguish between two generations of chemically distinct amphiboles or hornblendes (Table 4-2). These 51 Dating
chemical differences correlate with different ages. The Ca/K versus age diagram (Figure 4-12) shows that with approximately 47 Ma years, the hornblende with relatively higher Ca/K ratios is older than the hornblende characterised by lower Ca/K ratios, which has an age of approximately 31 Ma.
0.023- —I 0 023- 5 11 o 9 o 0.022- 0 022-|° O O 10 12 O10 o 4 O «0 021- $3 * 0.021 - o 6 08 13 6 *7 O Ö 0 020H O O 7 0.020 H o 0 019- 10 20 30 40 50 60 70 Ca/K 0.019-
12 14 16 18 20
Ca/K
Figure 4-11: Cl/K versus Ca/K ratios of sample SM60 obtained with the Ar/Ar method.
The coexistence of different hornblende generations is furthermore confirmed by electron microprobe analyses (Figure 4-13). The emp analyses (Figure 4-13) indicate that lower Ca/K hornblende-ratios correlate with higher edenite content and that higher Ca/K hornblende-ratios correlate with lower edenite content. It can be conclude that in sample SM60 two hornblende generations of distinct edenite contents and distinct ages coexist and therefore their composition is related to distinct metamorphic events.
50- SM60
45- T 4 -L 10 40-
< 9 5* 35- 6*
30- 10 2p 30 40 50 60 70 Ca/K 25- 10 15 20 25 30 Ca/K
Figure 4-12: Age versus Ca/K ratios of selected heating steps of sample SM60. The represented heating steps are related to the Ar-degassing of two distinct amphibole generations. The two amphibole generations have distinct Ca/K compositions and correlate with distinct Ar ages.
Beside hornblende, sample SM60 also contains prograde clinopyroxene (after amphibole and plagioclase see section 3.3.2.3) the presence of which permits to estimate the temperature conditions at approximately 670°C (cpx-in temperature after Russ-Nabelek, (1989)). It is clear that despite reaching this high temperature, the low-edenite hornblende has retained both its chemical and isotopic signature (cfr.
Villa et al., 1996; Villa 1998). However, it is possible that this temperature condition was not experienced 52 Dating
long enough to reset the Ar system completely. Therefore, inherited older amphibole compositions could
be distinguished from younger amphibole compositions.
Figure 4-13: Edenite versus Ca/K 0.70 ratios of amphibole grains from sample SM60. The electron microprobe (emp) analyses shows the presence of two amphibole generations characterised by distinct edenite and Ca/K ratios.
4.3.4 Sample SM81
The age spectrum of SM81 (Figure 4-14) has an internally discordant stage. The resolution provided by this sample is rather limited, as more than 50% of the entire gas was released during step 3, at a very unusually low temperature of 1009 °C.
Figure 4-14: Age spectrum of sample SM81
T 40 60 100 %39Ar
The microprobe data (Figure 4-15) demonstrate that sample SM81 is chemically quite heterogeneous.
It contains amphiboles which Ca/K ratios vary between 20 and 35. This compositional variability is also confirmed by the Ar isotopic compositions (Figure 4-16). With Ca/K values between 22.6 and 30.7, the steps between 2 and 11 are heating steps related to the degassing of amphiboles.
The age value correlated with step 3 is 31.2 ± 0.1 Ma, which is consistent with the ages obtained for the other samples. All other steps display scattered age values ranging between a minimum age of 16.2 ±
3.3 Ma and a maximum age of 63.1 ± 8.3 Ma. In particular, the majority of these steps correlate with Ar ages below 40 Ma. 53 Dating
1 D-
1.4-
1.2-
1- I I I 10 20 30 40 50 Ca/K
Figure 4-15: Electron microprobe analyses of sample SM81. a) Altv and b) edenite component versus Ca/K.
Similar to the other samples also sample SM81 clearly preserves relic amphibole compositions related to different metamorphic events. Nevertheless, also this sample indicates that the majority of the amphibole grains recrystallised during a metamorphic event occurred between approximately 35 and 30 Ma, which is consistent with the age information obtained from the other samples.
0,009 SM81 O7
0,008
^ 0,007 °6 « o O 0,007 5 °03 OO o 10 11 0,006 09
0,005
22 24 26 28 30 32
Ca/K
Figure 4-16: Cl/K versus Ca/K ratios of sample SM81. Steps 1 and 12 were excluded from the diagram because they do not represent amphibole heating steps.
4.3.5 Sample SM28
Sample SM28 has an irregular age spectrum (Figure 4-17a) where more than 50% of 39Ar-gas was released at a temperature condition of 1022°C represented by step 5. The apparent age linked to this step is 34.9 ± 0.6 Ma years. The emp data on this amphibole separate indicate the presence of optically indistinguishable amphiboles with Ca/K ratios ranging from 75 to 90 (Figure 4-18). Except for the first 54 Dating
three steps that represent impurities, all other steps show a significant spread of Ca/K ratios (85-105) with
scattered ages (Figure 4-17c).
The Cl/K versus Ca/K ratios (Figure 4-17b) shows that the relationship between K, CI and Ca is not
simple, in particular amphibole phases with similar Ca/K ratios have different Cl/K ratios. On the other side
back-scatter images (Figure 4-18b) taken of single amphibole grains from sample SM81 indicate
chemically homogeneous amphibole, whereas the electron microprobe analyses show a compositional
variability in their Ca/K and AIIV component. Apart from steps 6, a possible relationship (Figure 4-17c)
exists between the Ca/K ratios and their apparent ages. Amphibole compositions characterised by higher
Ca/K ratios correlate with older ages, while amphiboles with lower Ca/K ratios are display younger ages.
02
-1 100 110 120 130 Ca/K
90 100 110 120 130 Ca/K
Figure 4-17: Ar data relative to sample SM28: a) Age spectrum; b) Cl/K versus Ca/K ratios, steps 1 and 6 plot outside the represented range; c) Age (Ma) versus Ca/K ratios. The increase of Ca/K ratios correlates with a general increase of age.
The argon age-data and the isotopic information of sample SM28 show the existence of different
optically indistinguishable amphiboles, which display a complicated isotopic composition. Furthermore the
mineral assemblage and the distance of sample SM28 from the Gruf contact indicate that this sample
recrystallised under only moderate temperature conditions compared to sample SM60 (section 4.3.3).
Therefore this sample is probably characterised by a relatively higher concentration of relic older amphiboles, related to older metamorphic events. 55 Dating .:z:y*ymm
a) 1 ^ -
1.4-
1.3- >
1.2-
1.1-
10- 1 1 r 50 75 100 125 150 175 Ca/K
Figure 4-18: a) EMP-analyses in sample SM28: AIN versus Ca/K contents; b) BSE-imagine of amphiboles from separate SM28.
4.3.6 Sample BM4
Sample BM4 is the last amphibole-separate of the transept and corresponds to the "coldest" amphibole
sample among the separates. It crops out in the northern area of the Chiavenna unit, approximately 2.5
kilometres away from the Gruf contact. Temperature conditions at this distance are estimated to be
approximately 500°C. The amphibolites exposed at this distance contains amphiboles, which
compositionally are actinolitic Mg-hornblendes characterised by a low edenitic component. The influence
of the thermal overprint is interpreted to be weak or almost absent.
80 Figure 4-19: a) Age spectrum 13. of sample BM4. Age in Ma 70- 14 years. 60- 15 10 50- 12 g 40- 30- "l 20-
10-
0- —i— 20 40 60 80 100 o/o^Ar
The age spectrum of sample BM4 (Figure 4-19) has the most irregular shape of all analysed spectra.
The cumulative 39Ar-degassing rates of each single step (15 total steps) are relatively small. The electron
microprobe analyses show that the amphibole compositions extend over a large Ca/K range and the different Ca/K ratios correlate with distinct AIIV contents (Figure 4-20a), while the Na(M4) (Figure 4-20b) is
nearly constant. 56 Dating
fa) 1 75 0 30-
0 25- 1 50-
0 20-
1 25-
015- \ * 1 00- 010-
0 75- o os- i i i i 30 40 50 60 70 80 90 30 40 50 60 70 80 90 Ca/K Ca/K
Figure 4-20: Emp analyses of sample BM4: a) Altv versus Ca/K; b) Na(M4) versus Ca/K.
The BSE-images of BM4 show that the amphiboles contain domains of distinct Al compositions (Figure
4-21). The age versus Ca/K diagram (Figure 4-22) shows that steps with higher Ca/K ratios correlate with older ages and steps with lower ratios are linked to younger ages. The comparison of the electron
microprobe data with the Ar-isotopic results suggests that the amphibole reservoir with Ca/K ratios higher than 60 do not degassed alone in any step (cfr. Villa et al., 2000, Fig. 3d). An interpretation of such an age spectrum must be in agreement with the observation that this sample was not completely overprinted by the thermal metamorphic event and therefore it contains amphibole relics from older metamorphic events.
Figure 4-21: BSE-images of amphibole grains of sample BM4. 57 Dating
4-22: Ca/K of 100 Figure Age versus spectrum sample BM4.
4.4 Discussion and Interpretation of the results on a regional scale
4.4.1 Interpretation of the 39Ar/40Ar data
The compositional and textural analysis of the Chiavenna unit lithologies (Chapter 3) suggests the existence of a thermal event to account for the recrystallisation of the rocks under high temperature and relatively low-pressure conditions. This metamorphism changed the textural and the compositional features of former Alpine fabrics.
The 39Ar/40Ar age data obtained for selected amphibole separates of the Chiavenna amphibolites yielded irregular age spectra with complex interpretation. Nevertheless, taking into account the results of all spectra together, they support the metamorphic evolution proposed in this study of the Chiavenna unit
(Chapter 3). The results of investigating amphiboles from amphibolites using the 39Ar/40Ar dating method may be summarised in the following main points:
The comparison between Ar-data of samples extensively recrystallised at high temperature such as
SM71 and SM44 with relic-rich samples such as SM28 and BM4 showed that the intensity of
recrystallisation controls the Ar-Ar ages. Less discordant age-spectra are found for samples re-
equilibrated under high temperature conditions. The age spectra obtained for amphiboles
incompletely recrystallised at lower temperature are more irregular, due to the presence of inherited
amphiboles, chemically distinguishable from the younger amphibole compositions. In all samples, it is
possible to identify different amphibole compositions, which correlate with distinct ages.
The most coherent age-spectra support the occurrence of a metamorphic event changing the
compositional features of the amphiboles between 31 Ma and 33 Ma. None of the amphiboles provide
evidence for a metamorphic event at 25 Ma. 58 Dating
Sample SM71 and SM44 recrystallised rather extensively and older relics play the role of trace
phases. In contrast the EMP and the age data of sample SM60 allow to characterise amphibole relics
with ages around 45-47 Ma that show compositions distinct from the younger amphiboles. The
occurrence of compositionally distinct amphiboles in the separate is confirmed by emp analyses. Two
generations of amphiboles appear to coexist. The first amphibole generation is characterised by high
Ca/K ratios and a low edenitic component. The second generation has low Ca/K ratios and a high
edenitic component. The 39Ar/40Ar dating provides ages of approximately 45-47 Ma for the poor
edenitic amphiboles and ages of 31-33 Ma for edenite-rich amphiboles (Figure 4-23).
Ar-isotopes Emp Figure 4-23: Schematic simplified measurements analyses representation showing the link / \ \ / between isotopic argon Ca/K data, emp for the edenitic of LOW analyses component EDENITE the amphiboles and the obtained ages. AMPHIBOLES Y ARE OLDER V X X / / X
Older High// I nw High „ \ ^ ^ * •" AGES Ca/K EDENITE Ca/K \ \ \ \ ni IMGFR Low High ^ ^ I ow ^ ^ - " AGES Ca/K>\ /EDENITE Ca/K \ / HIGH EDENITE AMPHIBOLES ARE YOUNGER
Similarly, in those samples where the presence of inherited amphiboles components dominate the
whole age-spectra, it is possible to recognise age-trends between 30-35 Ma.
Although, the presence of "excess" argon cannot be completely excluded as opposed to the well
documented "inherited" Ar, the relatively uniform age information obtained for all samples suggests
that, if "excess" argon is present, then its influence is of minor importance in the Oligocène amphibole
generation.
4.4.2 Overview of the existing isotope data on the surrounding areas of the Chiavenna
unit
The Chiavenna unit, similar to other nappes and units of the Central Alps, underwent important Alpine tectono-metamorphic transformation. One important Alpine event was the Tertiary tonalite-granodioritic
Masino-Bregaglia intrusions (MB). An extensive age data collection on mineral-samples documenting the cooling and exhumation history of the MB intrusions can be found in the literature. Its intrusion age was established by von Blanckenburg (1992). The tonalite-granodioritic intrusions of the MB body were dated 59 Dating -""«'dgÊÊÊÊm
using different geochronometers. The different methods yielded identical crystallisation ages with a resulting mean age of 30.15+0.12 Ma. Data from the older tonalité intrusion are not so well constrained. In the western part of the MB intrusion, age-determinations performed on allanite and zircons document a time-range between 32.9 and 28 Ma (von Blanckenburg, 1992). Recent studies (Wiedenbeck and Baur,
1986; Villa and von Blanckenburg, 1991) investigated an E-W profile through the MB intrusions, using various dating methods, which provided more distinct results. According to these authors, the amphibole- ages are cooling ages. Hence, the metamorphic peak temperatures along the E-W extension of the MB intrusions were reached at different times. In the east, the thermal peak of metamorphism occurred before the MB tonalité intrusion at 32 Ma. In the west, where the deeper levels of the intrusion are exposed, the regional metamorphic peak was coincident with the Novate granite genesis.
For the eastern and northern area of the MB body, the relationship between the regional metamorphism and the MB intrusions is well established through the MB-contact aureole, which overprinted a Cretaceous greenschist fades metamorphism (Trommsdorff and Nievergelt, 1983). In the western area, next to the Lepontine dome, high metamorphic conditions up to amphibolite fades prevailed until 21-23 Ma (Koppel and Grünenfelder, 1975; Villa and von Blankenburg, 1991). The two regional events distinct in age and in metamorphic grade are thought to overlap within the MB area but it is still not clear. Tonalité pebbles found within the Oligocene-Miocene sedimentary sequence of the Gonfolite
Lombarda deposits (Bernoulli et al., 1993) suggest that, by this time part of the MB body had already reached the surface and had become eroded, while deeper parts of the intrusions (south of Bellinzona) were still above the solidus at depth of about 27± 4 km (Reusser, 1987).
Rb/Sr-age determinations on micas from gneiss of the Tambo nappe yielded ages around 30 Ma,
(Gulson, 1973) interpreted as intrusion ages of the MB body into the surrounding country rock. Age determinations in other units such the Gruf unit were unsuccessful and yielded uncertain Hercynian ages
(Gulson, 1973; Purdy and Jäger, 1976). On the contrary, the Novate intrusion was dated at 25 Ma (Liati and Gebauer, 2000).
4.5 Conclusions
The 39Ar/40Ar dating of Mg-homblende of the Chiavenna amphibolites confirms the occurrence of different Tertiary metamorphic events. One event is relatively well constrained in terms of ages and amphibole compositions. It is characterised by the degassing of a Ca/K low-ratio Mg-hornblende and in all separates analysed it was dated between 30-35 Ma. Edenite-samples, which were most extensively recrystallised at high temperature, yield the best age-spectra and refine the age of the above event to between 31-33 Ma. Electron microprobe analyses on amphiboles demonstrate that the change in amphibole composition is related to a metamorphic thermal event, which based on textural evidence, overprinted an older dynamic metamorphism. The thermal event is responsible for replacing former 60 Dating ____^
amphiboles characterised by low edenite and Al contents, as well as high Ca/K ratio compositions, with high temperature amphiboles of progressively higher edenite and Al values and low Ca/K ratios. Beside the Oligocène thermally recrystallised amphiboles, sample SM60, SM44 and BM4 preserve amphiboles with high Ca/K ratios, dated at approximately 45-47 Ma. These amphiboles are interpreted as index minerals of an Eocene regional medium temperature metamorphism.
The comparison of these dating results with the occurrence of Alpine regional metamorphism suggest that the high temperature overprinting (thermal event) of the Chiavenna amphibolites can be associated with regional thermal changes contemporaneous to the emplacement of the MB intrusions between 30-32
Ma, rather than to the thermal anomaly of the younger Novate granite intrusion dated at 25 Ma. 61 Thermal Model
5 Model of Thermal cooling
5.1 Introduction
As described in Chapter 3, the Chiavenna unit preserves evidences for two distinct Alpine metamorphic events. A first dynamic metamorphism in greenschist fades conditions and a second, essentially static, temperature controlled metamorphism evolving from lower amphibolite to upper amphibolite-granulite fades conditions. The analysis of mineral assemblages in the different lithologies indicates a temperature increase during the thermal event of approximately 170°C over a distance of only 2.5-3 kilometres. This temperature increase occurs along a prograde metamorphic path, where high-grade metamorphic assemblages progressively replace low-grade assemblages starting from the Tambo nappe in the north towards the Gruf contact in the south. The petrologic features describing this metamorphic event reveal strong analogy with typical features of contact metamorphosed terrain.
Alpine contact metamorphism in the Central Alps is not a rarity but has been documented for different units surrounding the Tertiary Masino-Bregaglia (MB) intrusions (Trommsdorff and Evans, 1972; Wenk et al., 1974). The Chiavenna unit is not directly in contact with the MB bodies, thus a metamorphic overprint related to the emplacement of these bodies appears rather improbable. In the present, the Chiavenna unit tectonically abuts the migmatitic Gruf unit. Field mapping of isograds in the Chiavenna unit (Schmutz,
1976), especially in the ultramafic system, showed that the isograds are essentially subparallel to the Gruf contact over much of the present exposure, suggesting the migmatitic Gruf unit as a possible heat source of thermal metamorphism.
In order to verify the hypothesis of contact metamorphism, numerical modeling of thermal cooling in the
Chiavenna contact aureole was computed assuming the emplacement of the hot Gruf unit as possible heat source. The comparison of the computed theoretical results with the natural metamorphic system of the Chiavenna unit evidences the analogies and differences existing between the model and the natural system. The model itself leads to discuss the geological implications assuming the Gruf unit as hypothetical contact metamorphic source and at the same time gives rise to consideration of other possibilities. The final proposition of the modeling consists of establishing the theoretical features necessary to generate a contact metamorphism in the order of the thermal range observed in the
Chiavenna unit and of comparing these features with the characteristics of the neighbouring bodies of the
Chiavenna unit.
5.2 Mathematical Model
The variables processed by the numerical model imply detailed knowledge of geometry, thermodynamic and structural conditions of both intrusion and country rocks. Metamorphosed peridotites of the Chiavenna unit were chosen as the chemical system hypothetically "intruded" by a "hot body", 62 Thermal Model
whose geometrical features, temperature and pressure conditions as well as position in respect to the
Chiavenna unit coincide with the features of the migmatitic Gruf unit.
Assuming the temperature conditions of the Gruf unit1, at the time of its emplacement to be at 800°C, it
would thermally overprint the adjacent "colder" Chiavenna unit. The initial temperature of country rocks
prior to the thermal overprint is assumed to be between 470°C and 520°C. The emplacement would result
in the development of metamorphic isograds with increasing metamorphic grade in the Chiavenna contact
aureole parallel to the Gruf contact, as observed in the real system. In the model of thermal cooling of the
aureole, the migmatitic Gruf unit (assuming 30% of rock partial melting) is considered to be a magmatic
intrusion with similar thermodynamic parameters as a normal magmatic body. Its initial temperature of
approximately 800°C is sufficiently high compared to minimal temperature required for incipient meltjng of
granitic magmas (Luth et al., 1964).
Table 5-1: Thermodynamic and geometrical parameters computed for modeling the thermal evolution of the Gruf aureole in the Chiavenna metapendotites.
Symbole Meaning Units CHARACTERISTIC VALUE
OuFIT SPECIFIC HEAT. GRUF UNIT J/m3/K 2X10«
C ULTRAMAFICS SPECIFIC HEAT. TLC-BEARING ROCKS J/m3/K 4.5x10«
Kmelt THERMAL CONDUCTIVITY, GRUF UNIT W/m/K 2.0
Ktlc THERMAL CONDUCTIVITY, TLC-BEARING ULTRAMAFICS W/m/K 2.25
Katg THERMAL CONDUCTIVITY, ATG-BEARING ULTRAMAMFICS W/m/K 1.75 Hm SPECIFIC ENTHALPY OF CRYSTALLISATION, GRUF UNIT J/m3 2.2 x10s
Hath SPECIFIC ENTHALPY OF DEHYDRATION ATH-ISOGRAD J/m3 16.2x10»
Htlc SPECIFIC ENTHALPY OF DEHYDRATION TLC-ISOGRAD J/m3 0.14X10»
Htr SPECIFIC ENTHALPY OF DEHYDRATION TR-ISOGRAD J/m3 32.7X10»
p PRESSURE bar, kbar
t TIME, (IN MILION YEARS) Ma
T TEMPERATURE °c
To PRE-INTRUSIVE AMBIENT T °c 470.500.520
X HORIZONTAL DISTANCE COORDINATE m
p GEOMETRICAL PARAMETRE 2
The specific heat of the Gruf rocks is assumed to be similar with the specific heat values of pure magmatic
sources (i.e. granite or tonalité), as indicated in Trommsdorff and Connolly (1996). The shape of the Gruf
unit, based on the present distribution of the outcrops, is approached to as a sphere of 5 kilometres radius.
The calculations performed are based on the typical peridotite composition from the Chiavenna unit
(CFMASH), neglecting the possible influence of CO2 in the fluid phase (Figure 3-6). The specific heat and
enthalpy values of anthophyllite, talc and tremolite isograds at pressure of 4 kbar are calculated on the
basis of the petrological results constrained in sections 3.2 and 3.5. All computed parameter values are
listed in Table 5-1.
1 On the basis of thermo-geobarometric results (Bucher-Nurminen and Droop, 1983; Droop and Bûcher, 1984) suggested for the sapphirine-granulites of the migmatitic Gruf unit maximum temperature and pressure conditions of 830°C attained at approximately 10 kbars. The age of migmatisation is contrivers, although Droop and Bûcher (1984) hypothesised the uplift of the Gruf complex between 38 and 30 Ma and recent studies (Davidson et al., 1996) indicated a late Alpine age for the emplacement of the Gruf unit, detailed geochronological studies remain to be done. 63 Thermal Model
The thermal evolution of the aureole, applying the parameters listed in Table 5-1 is described by the heat conductivity equation:
dT/dt = (1/xp) (5/Ôx) [xpK(ÔT/Ôx)] (1 ), whereby X is the horizontal distance from the centre of the body, P is a geometrical parameter (P-values of 0, 1 and 2 respectively correspond to lateral, cylindrical and spherical geometry), K is the thermal conductivity of the host rocks.
Equation (1) is solved by the Crank-Nicolson finite difference method applying no-flow boundary conditions at the centre of the sphere as well as isothermal boundaries at 25 km. The obtained data may be computed together with additional petrologic database to formulate diverse constraints on the effects of the heat transport in the aureole zone, such as:
• Maximum temperature values versus distance from the contact
• Temperature-time patterns
• Isograds distribution in the thermally overprinted area.
5.3 Results
The results obtained using the simple heat conduction equation (1) (Figure 5-1), reproducing the thermal cooling of the hypothetical Gruf contact aureole, are summarised in profiles describing the distribution of maximum temperatures as a function of distance from the heat source.
Figure 5-1: Computed thermal profile of maximum temperatures in the metamorphic aureole around the Gruf unit as a function of the contact distance. The solid curve represents the profile obtained for an initial country rock temperature of 520°C, the dashed curves are profiles obtained for initial temperatures of 470°C and 500°C. The horizontal grey fields show the calculated distances of the talc-in and anthophyllite-in isograds. The Talc in horizontal bars indicate the isograd "--.. .. ~ 500jC ~ - 470jC .... locations as observed in the field. _
1000 2000 3000 4000
Distance in metres
These results indicate that:
1. The computed temperatures within the contact aureole are comparable with the temperature values
observed for the thermally overprinted Chiavenna metapendotites;
2. The maximum temperatures obtained from the heat conductivity equation (1) are sufficient to stabilise
talc and anthophyllite parageneses; 64 Thermal Model
3. The extension of the thermal aureole indicated by the model is less than 3 kilometres and is in
agreement with the isograd distances observed in the field;
4. The computed talc- and anthophyllite-isograds (grey fields in Figure 5-1) are located slightly nearer to
the contact than the respective locations in the field.
5. The computed thermal model does not reach the temperature conditions necessary to stabilise
enstatite-bearing assemblages (685-700°C), which are present in the field.
Figure 5-2 summarises the thermal evolution in the contact aureole as a function of time, assuming an initial country rock temperature of 520°C. At the contact, temperatures above approximately 620°C are maintained over a time of 1.6-1.7 million years (Ma), whereas in the aureole, maximum temperatures are reached between 0.3 and 0.5 Ma after the intrusion. At 1.7 Ma, the most distant outcrops from the contact are still above 530°C, a temperature sufficiently high to stabilise talc. Each temperature-profile, independently from the distance from the contact, may be divided in two sections with distinct slopes:
During the first 1.7 Ma after the intrusion, the temperature within the aureole evolves relatively slowly.
After this period, the temperature profile falls relatively rapidly to pre-intrusion conditions. This marked change in the slope of the temperature curve at 1.7 Ma, within the aureole, is dictated by thermal changes in the source. The relatively slowly cooling of the aureole during the first 1.7 Ma is probably in relationship with the crystallisation of the source, at 1.7 Ma the source reaches solidus conditions heating the surrounding rocks significantly slower. The heating or cooling process in the surrounding rocks may be considered finished after 3 Ma, when with the temperatures falling below 520°C at the contact.
692 5-2: evolution ~i r Figure Computed temperature at different distances from the contact versus time (in million years). * 658 'So, N
^623 - /500m"- / -/' ! -'1000 m" 589 | r :
;.' /'Î500m~
554 " ,' /" 2000 m ."'-"-C^- 1.7 520 i; i I i_ i i i 0.6 1.2 18 2.4 Time (Ma) 65 Thermal Model ,-~~$m.
5.4 Discussion
5.4.1 Was the Gruf Unit responsible for the thermal overprinting event in the chiavenna
ROCKS?
By applying a simple heat conduction equation of type (1) it was possible to compute the thermal and the geometrical features of an area ultramafic in composition directly in contact with a spherical heat source, whose thermodynamic and geometrical values coincide with the characteristics of the Gruf unit.
The results indicate that the temperature boundary conditions coincide with the temperature differences between the regional and the thermo-metamorphic overprinting observed in the Chiavenna
metapendotites. The computed data are in the order of the real metamorphic data observed in the field and the location of the calculated isograds are consistent with the position of isograds in the field, with the only exception of the enstatite-bearing assemblages, which were not reproduced by the computed model.
However, the differences between the model and the natural system are related to the unavoidable simplifications that have to be made for the computed parameters and inevitable in any model. In particular, following aspects were not included in the mathematical model:
The minimum temperature required by the enstatite-in reaction is assumed to be 700°C. This latter
assumption is true for a pure CFMASH system, which is considered representing bulk rock
composition of peridotites from the Chiavenna unit. The influence of other possible chemical
participants such as CO2 is neglected. In particular, the presence of CO2 in the fluid phase would
stabilise the anthophyllite field and the enstatite-in reaction at lower temperature. The occurrence of
carbonate-bearing rocks in the Chiavenna ultramafic system (section 2.2.3) is documented,
consequently these rocks may reasonably affect the fluid chemistry, shifting it to mixed H2O-CO2
compositions and favouring the dehydration reaction of Mg-amphiboles (enstatite-in) at lower
temperatures than the calculated values.
The thermal model used in this study is a static model whereas the emplacement of the Gruf unit
occurs through tectonic movements which dynamically "feed" the surrounding country rocks with new
heat coming from rocks of deeper crustal levels. Therefore, the real temperature conditions at the
contact are likely to be enough high to stabilise enstatite.
Structural features, such as fractures and textural anisotropy, as well as heat advection related to the
presence of fluids, may influence the thermal development within the contact aureole and may
enhance or hamper the heat transmission in the rock system.
The most important factors directly controlling the thermal conditions in the contact aureole are the
geometrical and thermodynamic parameters characterising the heat source. In the case of the Gruf
Unit, the thermal conditions of approximately 800°C may be considered a minimum estimate of the
real temperature, which probably was above (Droop and Bûcher, (1984) suggested 830°C). 66 Thermal Model
Furthermore, all assumptions made in the order to assimilate the migmatitic Gruf unit to a spherical
intrusive body are complicated and imply significant simplification problems that could be minimised
with specific field and experimental work on the Gruf unit, which would exceed the goals of this study.
5.4.2 Alternative heat-sources
Despite the fact that alternative heat sources may have furnished the enthalpy necessary for the thermal overprinting in the Chiavenna Unit, these scenarios seem to be rather improbable, because they are not able to explain the subparallel character between the observed ultramafic isograds and the Gruf
Unit contact in the field. However, some considerations may be brought forward, in order to evaluate objectively all possible causes.
5.4.2.1 Masino-Bregaglia related intrusions
The Ar/Ar dating of amphiboles from the Chiavenna amphibolites (Chapter 4) indicate that the thermo- metamorphic event responsible for the recrystallisation of the amphiboles took place around 30 Ma ago.
This time approximately coincides with the age of the Masino-Bregaglia intrusions (32 Ma for the tonalité and 30 Ma for the granodiorite). The comparison of the thermal conditions of the Chiavenna ultramafics with the thermal conditions reached in ultramafic rocks adjoining the eastern border of the MB intrusions
(Malenco ultramafic contact aureole (Trommsdorff and Connolly, 1996)) reveals lower temperature conditions in the aureole as well as a slightly narrower aureole of approximately 2 kilometres thickness.
Hypothesising the presence of a tonalité "sheet" along the present Chiavenna-Gruf contact, the anthophyllite assemblage of the Chiavenna ultramafic rocks would require either an emplacement temperature higher than 700°C or a larger extension of the tonalitic sheet.
Dikes of tonalitic composition crosscut the Chiavenna unit in several outcrops and may genetically be related to the MB early tonalité intrusion, but none evidences of clear tectonic or intrusive contacts between the Chiavenna unit and the MB intrusion directly exist. By contrast, intrusive relationships between the southern Gruf outcrops and the southwester MB intrusions are described in several studies
(Trommsdorff and Nievergelt, 1983; Davidson et al., 1996; Schmid et al., 1996b). According to recent studies (Davidson et al., 1996; Schmid et al., 1996a)], the Masino-Bregaglia intrusions was originally emplaced along a former nappe boundary corresponding to the North-Penninic suture zone, between the
Adula and the Tambo nappes. The emplacement of the pluton is interpreted to be part of a regional deformation phase, tentatively dated at 30-35 Ma. This geotectonic scenario would coincide with the tectonic position of the Chiavenna unit. Consequently it cannot be excluded that, in an early stage, the MB bodies might have been in contact with the Chiavenna ophiolitic rocks and the tonalité dikes crosscutting the Chiavenna unit may be considered as a relic occurrence of this particular stage. 67 Thermal Model
5.4.2.2 Deep seated bodies
Other heat sources might be related to deep-seated high temperature bodies not visible on the surface. However, this hypothesis is not supported by seismological data (Schmid et al., 1996b), which revealed that there is no evidence of deep-seated magmatic bodies in the specific area.
5.4.3 Geological implications of gruf unit enhanced contact metamorphism
The Ar/Ar data on amphibole (chapter 4) indicate that the thermo-metamorphic event in the Chiavenna unit can be dated at approximately 30 Ma. At this time the Central Alps, especially the area around the
Masino-Bregaglia intrusions, were dominated by complex tectono-metamorphic conditions related to the emplacement of the magmatic bodies and to vertical movements related to the general uplift of the northern block along the Insubric line. Assuming the Gruf unit as possible heat source of contact metamorphism, following geological implications on a regional scale have to be taken into consideration:
The emplacement of the Gruf unit in its present position has to be contemporaneous to the
emplacement of the MB intrusions.
At the time of intrusion of the MB bodies, the Gruf unit still underwent high-grade metamorphism
(amphibolite-granulite facies conditions). The absence of contact aureole in the Gruf unit related to
the emplacement of the Bregaglia intrusions suggests that temperatures in the Gruf unit at the time of
intrusion were high (Droop and Bûcher, 1984) and sustains the hypothesis of a Gruf unit enhanced
contact metamorphism in the region.
The uplift of the Gruf rocks along the existing tectonic contact between the Gruf and the Chiavenna
units (steeply north dipping fault) had to be rapid, in order to drive the amount of heat necessary for
the thermal overprint in the Chiavenna rocks. The pressure conditions of 10 kbar estimated for the
Gruf unit during granulite facies metamorphism (Bucher-Nurminen and Droop, 1983; Droop and
Bûcher, 1984) may be overestimated, but not necessarily if the uplift was isothermally. Distinctly lower
pressure values at the moment of intrusion were calculated for the rocks of the Chiavenna unit
(approximately 4 kbar, section 3.5).
The tectonic contact existing between the two units is interpreted to be the south-western continuation
of the nothern Engadine line (Schmid et al., 1996b). In the north, field relationship between the
Engadin line and the Bregaglia granodiorite suggests pre-intrusion ages for the tectonic discontinuity
(Davidson et al., 1996). This should also be the case for the Gruf-Chiavenna tectonic contact, which
should not be younger than the contact metamorphic event. 68 Thermal Model
5.5 Conclusions
In this chapter, the results of a simple mathematical modelling were presented. The main goal of these calculations were to verify if the Gruf Unit may have theoretically provided sufficient enthalpy to enhance the thermal overprinting observed in the rocks of the Chiavenna Unit. The parameters used for the calculations were chosen to simulate the physical and morphological characteristics of both Gruf and
Chiavenna units, as observed in the field and as obtained from experimental data on similar rocks.
Although specific features of the system considered, such as anisotropies, fractures or heat advection caused by percolating fluids were not taken into account, the computed data are in reasonable agreement with the petrologic situation observed in the field. The results do not demonstrate that the Gruf unit acted as a sole heat source for the thermal overprint in the region, but indicate that the heat furnished by a similar migmatitic body, may be sufficient to enhance prograde metamorphic reactions typical for contact metamorphism, as the reactions described for the ultramafic system in this study. 69 Geochemistry
6 Geochemistry
6.1 Introduction
This chapter focuses on the geochemistry of the Chiavenna Unit and on its possible "ophiolitic" nature.
Former studies [Schmutz, 1974 #47] interpreted the Chiavenna ultramafic-mafic complex as remnants of a metamorphic ophiolitic body while other studies (Huber and Marquer, 1998) advanced the hypothesis to consider the Chiavenna unit, in analogy to studies on the Val Malenco ultramafics (Trommsdorff et al.,
1993) as relics of exhumed pre-rift subcontinental mantle with associated lower crust rocks. The
Chiavenna unit underwent an intense Alpine pT-evolution (Chapter 3). Pre-Alpine metamorphic or former textural and mineral assemblages are hardly preserved. In this study chemical investigations in high- metamorphic terrain such as the Chiavenna unit are considered a complementary and useful approach to answer to the above problematic. In the following sections the geochemical data on the Chiavenna unit and possible interpretations are presented and discussed.
6.1.1 Aims of geochemical analysis
The following sections attempt to:
Present the bulk chemistry of the different lithologies building up the Chiavenna unit.
Discuss the possible genetic history and origin of the Chiavenna unit by integrating field observations
and chemical characteristics lacking in previous studies.
Suggest possible pre-metamorphic scenarios for the Chiavenna unit taking into account regional
geological aspects.
The bulk rock chemistry of each lithology is presented in separate sections. The data are recalculated on an anhydrous basis and discussed with respect to their major, minor and RE element trends. Diagrams used for the data presentation contain data of the Chiavenna ultramafic-mafic and carbonate rocks and the compositional fields of similar rocks from different Alpine regions or from areas outside the Alps. This comparison helps to recognise similarities and differences among the rocks due to genetic processes or latter magmatic and metasomatic features. Detailed major element, transition metal and trace element abundances of all lithologies are listed in the Appendix.
6.2 Bulk rock chemistry of metaultramafic Rocks
6.2.1 Major element composition
The major elements concentrations of the Chiavenna metapendotites (Figure 6-1) are similar to those found in other metamorphosed peridotite bodies from different areas of the Alpine orogen, from the 70 Geochemistry 'Willi
Northern Apennine area, from outside the Alpine edifice, from present day oceanic serpentinites and from abyssal peridotites.
12
SU6 SU15 47- 11 o <> SU8
45- 10-n CD CM O O § CO 9
43
/x SU8 0 SU°^SU75 41 Q
020- 45
40
O 35
30
25
Al203
Peridotites
} Chiavenna Unit (this study)
0 Adula-Cima Lunga nappe Estimates of primary mantle composition: Jagoutz et al ,1979 f Hofmann, 1988 R,ngwood,1975 • ^^MS Compositional ranges of different peridotites areas: H Ronda, Horoman, and Lanzo Al203 \Z\ Oceanic serpentinites 1 I External Ligunde EL I I Internal Ligunde IL
' l Malenco Unit
Figure 6-1: Major element distribution of the Chiavenna metapendotites compared with other orogenic peridotites and oceanic serpentinites: (Lanzo: (Bodinier, 1988); Ronda: (Frey et al., 1985); Horoman: (Frey et al., 1991) (LRH); Malenco unit: (Müntener, 1997) (M); Adula-Cima Lunga nappe: (Pfiffner, 1999) (CdG)), External and Internal Liguridi (Rampone et al., 1995; 1996; 1998) (EL, IL), Atlantic serpentinites (Cannât et ai, 1995; Gillis et al., 1993)), abyssal peridotites (Dick, 1989).
The AI2O3 content has been chosen as a convenient measure of rock-depletion in basaltic components.
In the Chiavenna metapendotites it varies between 2.7-3.9 wt.% and correlates with most of the other major elements CaO (0.02-3.9 wt.%), MgO (36.9-43.7 wt.%) and Ti02 (0.04-0.14 wt.%) in agreement with the above mentioned trends of typical orogenic peridotites or oceanic serpentinites. On the other hand, 71 Geochemistry
SiÜ2 (43-47 wt.%) and FeOtot (7-11.2 wt.%) contents show scattered values plotting outside any compositional field of other peridotites.
High AI2O3 values are comparable with estimates of primary mantle composition (Jagoutz et al., 1979;
Ringwood, 1975; Hofmann, 1988), while lower contents of AI2O3 are comparable with compositions of abyssal peridotites (Dick, 1989). The most primitive metaperidotite composition of Chiavenna is given by sample SU6 while the metaperidotite most depleted in basaltic component is sample SU8. The remaining samples display fertile or slightly depleted Iherzolitic composition and only sample SU8 with an AI2O3
content of 1.26 wt.% can be considered harzburgitic in composition.
The S1O2 content of the rocks is high and the data are scattered but still in the compositional range of
Alpine peridotites. The FeOtot values plot within the Lanzo, Ronda, Horoman, (LRH) Malenco (M) and
Cima di Gagnone (CdG) peridotites fields. Rocks with high AI2O3 contents overlap with the range of the
External Ligundi (EL) peridotites and are chemically similar to primitive mantle compositions. In the TiÛ2 versus AI2O3 diagram sample SU8 plots outside any field and most of the other metapendotites mainly fall within the compositional field of oceanic serpentinites and in the range of the LRH peridotites.
The MgO content documents some differences among the Chiavenna peridotites. It is possible to
distinguish between two groups of rocks: low MgO (<40 wt.%) metapendotites with fertile Iherzolitic
compositions comparable to primitive mantle and high MgO (>40 wt.%) metapendotites which show a
positive correlation with AI2O3. Similar features can be observed in the CaO distribution. High CaO
contents are comparable with CaO values of primitive mantle estimates and fit well the trends described
for the other Alpine peridotites. However, some samples with equal or similar AI2O3 have extremely low
CaO conttnts. The Ca-depletion is in agreement with Ca-trends observed in oceanic serpentinites and in
Alpine peridotites which still retain oceanic serpentine (i.e. Malenco unit: (Miintener, 1997)).
Figure 6-2: Ca/AI ratio compared to chondrite composition: Ca/AI versus MgO contents, filled diamonds = metapendotites with chondrite composition, white diamonds = metapendotites with depleted chondrite composition. Chondrite range for primitive mantle fixed at Ca/AI ratio 1.09, for massif peridotites = 1.18 (after McDonough, 1994). A schematic illustration of the effects of partial melting on the element concentration of a primitive mantle source after (McDonough and Sun, 1995) is illustrated.
MgO
It is known that most peridotite xenoliths and some massif (Alpine) peridotites have experienced early melt extraction and later incompatible element enrichment. Recent work on the composition of the "Silicate
Earth" (McDonough, 1994; McDonough and Sun, 1995) showed that for most peridotites from different 72 Geochemistry
areas of the world, the proportional depletion of Al, removed from the peridotite due to partial melting, is greater than the Ca-depletion, consequently the Ca/AI ratio increases with increasing degree of melting.
Therefore, the high Ca/AI ratios observed in depleted peridotites are likely to be caused by melt depletion and are not an original feature of the mantle. The analysed Ca/AI ratios (Figure 6-2) of the metapendotites do not show partial melting effects. On the contrary, the Ca depletion in the rocks is higher than the decrease in Al. This feature is interpreted as the result of secondary Ca-leaching processes, possibly caused during hydrothermal alteration and ocean floor weathering (serpentinisation) rather than a consequence of partial melting.
Although, it is reasonable to regard the Ca-depletion as an effect of alterations in an oceanic environment on the basis of the above mentioned features, the contents of T1O2, and especially of MgO, are not affected by the same amount of depletion as other typical serpentinised Alpine peridotites
(Malenco and CdG harzburgites). In contrast, the Chiavenna serpentinised metapendotites (Figure 6-1)
(open diamonds) display similar Ti and Mg abundances as unaltered fertile metapendotites (filled diamonds). Harzburgitic peridotites are defined as former Iherzolites depleted in basaltic component.
Hence, with relatively high AI2O3 content (3.1-3.9) the Chiavenna Ca-depleted metapendotites can not be considered harzburgitic in composition but represent rather depleted Iherzolites. The modest Mg-increase and Ti-depletion occurring with increasing refractory character of the metapendotites can be explained with small degree of partial melting (see below) whereas where similar Mg and Ti-behaviour characterise
Ca-depleted Iherzolitic samples (open diamonds), serpentinisation may be a better explanation of the compositional signal.
6.2.2 Transition elements
The Chiavenna metaperidotite transition element trends (of Cr, Ni, Sc, and V) are in agreement with compositional ranges of other peridotites from the Alpine belt (Figure 6-3), as observed for the major element compositions.
The compatible element Cr (2928-5168 ppm) displays values similar to primitive mantle compositions
(data from (McDonough and Sun, 1995)), slightly higher than the Malenco and LRH peridotites and similar to values observed for oceanic serpentinites. The Ni data (1795-2189 ppm) coincide with the fields of the
M, LRH and EL. In both element diagrams, an overlap with the Iherzolites trend of the CdG can be
recognised. Sc (10-20 ppm) and V (52-138 ppm) contents decrease with increasing refractory character of the metapendotites. Both elements show positive trends similar to the other represented peridotites. The best overlap occurs with the CdG and LRH fertile Iherzolite samples and with the EL peridotites. 73 Geochemistry
3000
Peridotites Compositional ranges of different peridotites areas:
Ronda, Horoman, and Lanzo # } Chiavenna Unit (this study) _i \7\ Oceanic serpentinites o Adula-Cima Lunga nappe fol External Liguride EL Estimates of primary mantle composition [TO Internal Liguride IL Jagoutzetal., 1979 I Malenco Unit Ringwood, 1975
Figure 6-3: Transition element abundances of the Chiavenna peridotites (filled diamonds) compared with the compositional fields of peridotites from other Alpine and not Alpine regions: grey field - peridotites from the Ronda, Horoman and Lanzo areas (Frey et al. 1985; 1991; Bodinier, 1988), dotted fields = peridotites from the External and Internal Liguridi (Rampone et ai, 1995, 1998), striped field = oceanic serpentinites (Cannât et al., 1995; Gillis et al., 1993), dashed field = peridotites from the Malenco unit (Müntener, 1997), triangles = estimates of primary mantle composition (Jagoutz et al., 1975; Ringwood, 1975). Sample SU6 has the highest Al203 content.
Only two samples display compositions that differ substantially from the compositions of most analysed
Chiavenna metaultramafic rocks. These variations are due to differences in the protholitic composition and to metasomatic processes:
The presence of chromite-rich layers in sample SU15 is largely reflected in the Cr, FeO tot
and V contents.
The high V and very low Ni and Cr contents of sample BU35 reflect element-mobility during
metasomatism, caused by the particular setting of this rock-specimen exposed next to
metamorphosed rodingite boudins (section 6.5.4).
The fertile character retained by the Chiavenna metapendotites is also recorded in the Ni-Co-Sc (Figure
6-4) (normalised to primitive mantle) composition diagram. 74 Geochemistry ^SaÊÊÊÊÊÊÊtBÊÊÊÊÊÊÊÊÊÊÊÊÊÊ^ÊÊ^KÊÊÊÊÊÊÊIÊ^^^^^^^KÊÊÊÊÊ
As a matter of fact, the Ni/Co ratio, and other compatible elements in peridotites, is unlikely to change significantly as results of partial melting (Ottonello et al., 1984). In contrast, Sc belongs to the moderately incompatible elements and partitions into the liquid phase. The continuos increase of partial melting of a peridotitic source during "equilibrium" or "fractional" melting would decrease the Sc in the residual rock at approximately constant Ni/Co ratio. Figure 6-4 indicates that the Chiavenna metapendotites have fertile
Iherzolite features affected by a slight depletion that drives some rock compositions away from the Sc corner. The Malenco and CdG peridotites comprise samples that are more depleted in Sc and occupy essentially different Ni and Co fields than the studied metapendotites. This Sc-depletion mirrors a higher degree of partial melting occurred for the M and CdG peridotites than for the Chiavenna metapendotites.
NiN Figure 6-4: Ni-Co-Sc contents normalised to mantle abundance after (Jagoutz et al., 1979): black filled diamonds = Chiavenna metapendotites, open diamonds = CdG peridotites (Pfiffner, 1999), grey diamonds = Malenco peridotites (Müntener, 1997).
CoN ScN
Distinct transition element patterns according to rock type can also be observed using spider diagrams, where the element concentration is normalised to the primitive mantle abundance (Figure 6-5). The Al-rich metalherzolites have rather flat patterns, with most sample/primitive mantle ratios close to one. The slightly Al-poor refractory metalherzolites are variously depleted in incompatible elements such as Sc, Ti,
V and slightly enriched in Cr, Mn and Zn relative to the primitive mantle abundance. The Chiavenna metapendotites accommodate high contents of incompatible elements similar to the Lanzo peridotites. The
Ti/V ratios (ratio-values: 9-5) for Iherzolitic and refractory samples are lower than the Lanzo fertile (ratio value: 12) and refractory (ratio value: 7) Iherzolites, but the Chiavenna Ti N ratios are still higher than the values (ratio values: 0.17-6.2) of the Lanzo harzburgites ((Bodinier, 1988) and references therein).
Furthermore, the analysed metapendotites display values that are close to the primitive mantle value of
7.9 (McDonough and Sun, 1995). In summary, it can be concluded that the major and trace element patterns of the Chiavenna metapendotites are consistent with typical compositions of rather fertile
Iherzolites, which have undergone varying degrees of serpentinisation. 75 Geochemistry
bulk rock/primitive mantle Figure 6-5: Transition elements of the Chiavenna 10: peridotites normalised to primitive mantle composition (Jagoutz et al., 1979). O metaperidotites • meta-pyroxenite
0 1- ~i 1 1 1 1 1 1 1 r~ Sc Ti V Cr Mn Fe Co Ni Zn
6.2.3 Re- and trace element composition
Chondrite and primitive mantle normalised REE patterns provide a useful information to distinguish between different Iherzolite types and reveal information on depletion mechanisms.
The chondrite normalised REE patterns of the Chiavenna metaperidotites are outlined in three different diagrams (Figure 6-6a-c). Based on their specific REE-features, three different types of metaperidotite trends (Figure 6-6d) can be recognised:
=> Type A metaperidotites display a fertile character in their heavy REE (enrichment factor<2 x
chondrite), are slightly fractionated in their middle REE SiTWYbN=0.48-0.69 and are strongly
fractionated in their incompatible light rare earth elements (CeN/SmN=0.06-0.17).
=> Type B metaperidotites are characterised by relative flat REE-segments, are slightly more depleted in
their HREE compositions relative to type A patterns (and similar to the latter group) they show slight
MREE fractionation (SmN/YbN=0.49-0.62). By contrast, the incompatible light elements
(CeN/SrriN=0.46-0.68) are less depleted than those one of group A.
=» The third group of metaperidotites is sketched out by type C patterns. They have lower HREE and
MREE concentrations and are more fractionated (SrriN/YbN=0.21-0.32) than group A, while they have
similar fractionation values in the LREE (CeN/SmN=0.09-0.27). In addition type C metaperidotites
display a systematic negative Eu anomaly.
Type A and C metaperidotites coincide with the fertile metalherzolites and the Ca-depleted refractory metalherzolites, respectively, already defined on the basis of their major and trace element behaviours.
Beside these two groups the chondrite normalised REE patterns allow the recognition of a third group of rocks, defined as type B enriched metalherzolithes, which in their major and minor element distributions are not clearly distinguishable from the other groups of metaperidotites.
The convex downward normalised chondrite pattern shown by the type A and C metaperidotites is characteristic for many peridotites from the Alpine belt (harzburgites in particular present similar LREE depletion). 76 Geochemistry
bulk rock/chondrite bulk rock/chondrite
\U - a
r--&^c5 ^I^P^^-* tT=V'-* W''-'
's/ " metalherzolites enriched metalherzolites
0.1- ^^SU6 0.1- -- SU77-B-- SU51 [ ^7 --B--SU4 —o— BU35 —•— BU28 B --SU15 - V - BU27 ---*--- BU41 —0—BU6
0.01- I I I I I I I I I I I I é Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu bulk rock/chondrite bulk rock/chondrite
0.1-
0.01 i—i—i—i—i—i—i—i—i—i—i—r Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ce Pr Nd Sm Eu dd Tb rjy hlo Er Tm Yb Lu
Figure 6-6: REE chondrite normalised patterns of the Chiavenna peridotites. Chondrite composition after (McDonough and Frey, 1989). a-c): single sample REE patterns, d) REE trends summarised in different compositional fields: A = metalherzolites, B = enriched metalherzolites and C = refractory metalherzolites.
The Iherzolitic peridotites of the Internal Liguridi (IL) are also characterised by depleted LREE patterns
(Rampone et al., 1996). Similar to the IL peridotites, the Chiavenna A metalherzolites are depleted in their
LREE and show a Tb/Yb ratio between 0.16-0.22. The Tb/Yb ratios of type C refractory metalherzolites are 0.1-0.12, indicating a fractionation of the middle to the heavy REE. In Figure 6-7, the Tb/Yb ratios versus Yb are represented.
Figure 6-7: Tb/Yb versus Yb (in ppm) fractionation Chondrititc ratio A diagram. 0.20- %
* 0.15-
oo * 0.10- o Metaperidotites A-type 0.05- B-type o C-type
0.00- i i i 0.0 0.1 0.2 0.3 0.4 Yb Geochemistry
Type C refractory metalherzolites display distinct REE concentrations (Figure 6-6) from type A metalherzolites. The SrrWYbN fractionation range of these rocks approaches values between 0.21-0.32, which are considered low for fertile Iherzolites concentrations. Nevertheless, beside the fact that type C metaperidotites display Al contents typical for Iherzolites, similarly SrrWYbN ratios of 0.28-0.31, characterising ultramafitites from the IL, are regarded to be amongst the most depleted compositions of ophiolitic Iherzolites from the Western Mediterranean area (Rampone et al., 1996).
The peculiarity of type C patterns is the slightly negative Eu anomaly exhibited by these metaperidotites. The Eu anomalies can be visualised in a Eu/Eu* versus CaO contents diagram where
Eu/Eu* = EuN/((SmN)*(GdN))1,2](Rollinson, 1993 and references therein)) (Figure 6-8).
1.2- Figure 6-8: CaO (in wt%) versus Europium BlÄ%J35 anomalies (Eu/Eu)* in metaperidotites of the Chiavenna unit; [Eu/Eu* = EuN/((SmN)*(GdN))1/2 as 1.0- * defined in Rollinson, 1993 #32].
3 ^ SU16 J*. * LU 0.8- & o Sn Susi 3 111 rBU27 i . Mg13 0.5^ SU75 metalherzolites Ô refractory metalherzolites &SU76 # enriched metalherzolites
0.2- -1 r~ —\— 0.0 1.0 2.0 3.0 4.0 5.0 CaO
The Eu anomaly distribution among the metaperidotites of the Chiavenna unit shows that metaperidotites with low Ca content are characterised by high values of negative Eu anomaly (Eu/Eu* < 1) while metaperidotites with relatively high Ca contents display only slightly negative anomalies (Eu/Eu* < 1). If the Ca-loss in peridotites is considered as a convenient measure of serpentinisation it is evident that this hydrothermal alteration also influenced the Eu content of the Chiavenna metaperidotites. During serpentinisation the peridotites, interacting with the seawater are subject to a chemically highly reducing environment where divalent Eu is leached from the system.
The primitive mantle normalised incompatible trace elements of the Chiavenna metaperidotites (Figure
6-9) exhibit variously scattered trends. The plotted elements are ordered by increasing peridotite-basaltic melt distribution coefficient, as suggested by Sun and McDonough (1989) and McDonough and Sun (1995).
Amongst the elements integrated in the normalised spider plots the high field strength elements (e.g. Ti, Hf and Zr) with distinct concentrations for different metaperidotites are of particular interest. Titanium generally attains similar concentrations to the adjacent Gd and Eu elements. Positive Ti-anomalies characterise the refractory type C metalherzolites, while type B metaperidotites display positive and slightly negative anomalies. 78 Geochemistry
Negative Zr-anomalies characterise all three types of metaperidotites with significantly low values for the type C rocks. The refractory metalherzolites also show low Hf-values, while the fertile (A type) and the enriched (B type) metalherzolites have comparable Hf concentrations.
The strontium concentrations are more scattered than most other trace elements. Positive anomalies relative to the neighbouring elements can be observed in A and in most B type metaperidotites. A strong positive Sr anomaly is present in an ultramafic rock crosscut by various carbonate veins (sample SU51).
Contrasting positive and negative Sr anomalies characterise type C refractory metaperidotites. Related to
Ca-depleted ultramafic rocks, negative Sr-anomalies are also described in serpentinised peridotites from the Malenco area (Miintener, 1997) and in some harzburgites from CdG (Pfiffner, 1999).
bulk rock/primitive mantle bulk rock/primitive mantle 100 a
10-1
I \ 111 /ö\ 01- ' GQpjS SU77 o-~ BU35 —— SU4 St / —o-- BU27 xJ° -... SU6 0 01. 0 01 —&— SU51 ---o-- SU15 --B- BU28 —tx— BU6 ---»-- BU41
0 001 - 0 001 1 1 1 1 1 1 1 1 1 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Rb Th Nb La Pr Nd Hf Eu Gd Dy Er Yb Rb Th Nb La Pr Nd Hf Eu Gd Dy Er Yb Ba U Ta Ce Sr Zr Sm Ti Tb Ho Tm Lu Ba U Ta Ce Sr Zr Sm Ti Tb Ho Tm Lu
bulk rock/primitive mantle bulk rock/primitive mantle 100 100
0 1-
SU75
- -«•- SU76
0 01 —o- SU8 0 01
—a— Mg13
--»- SU16
0 001 0 001 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Rb Th Nb La Pr Nd Hf Eu Gd Dy Er Yb Rb Th Nb La Pr Nd Hf Eu Gd Dy Er Yb Ba U Ta Ce Sr Zr Sm Ti Tb Ho Tm Lu Ba U Ta Ce Sr Zr Sm Ti Tb Ho Tm Lu
Figure 6-9: Mantle normalised trace elements of the Chiavenna metaperidotites. Figures a, b and c: trace elements and REE patterns of single samples; Figure d: Trace and REE profiles grouped in three different compositional fields: A = metalherzolites, B = enriched metalherzolites and C - refractory metalherzolites.
All metaperidotites are characterised by constant negative thorium anomalies. The niobium and especially the uranium abundance vary with different rock-chemistry. Positive U-anomalies are present in type B metaperidotites while the niobium values are slightly higher but still comparable to the adjacent La
concentration. The U concentration decreases in the fertile Iherzolitic metaperidotites where two samples 79 Geochemistry show an unusually positive U-anomaly. The refractory metalherzolites have positive U-anomalies similar to the B type metaperidotites. Finally the more fertile and the refractory metalherzolites are enriched in Nb compared to the adjacent La contents.
The heterogeneity of these element-anomalies makes difficult to verify the hypothesis of element mobility during metamorphism more over when taking into account that the processes related to the abundance of elements such as U, Nb, Ba, Ta and Th in mantle peridotites are generally poorly known.
6.3 Trace element distribution in ultramafic mineral phases of metamorphic
parageneses: a laser ablation icp-ms study
6.3.1 Introduction
The amount of data published on trace and rare earth element abundance and distribution in mineral phases of peridotites is increasing with the analytical advances achieved during the last ten years. The scientific interest concerns especially the abundance of incompatible and compatible trace elements hosted in mantle minerals such as olivine, pyroxene and garnet but also mantle hydrous phases such as amphibole (O'Reilly et al., 1991, Bedini and Bodinier, 1999; lonov, 1992;Eggins et al., 1998;Norman,
1998; Chazot et al., 1996; Blusztajn and Shimizu, 1994). Olivine and orthopyroxene seem to play a minor role in terms of bulk rock REE budget while clinopyroxene and garnet incorporate the majority of these elements. The information attained through the knowledge of element distributions in mantle phases is of interest for the characterisation of lithospheric and upper mantle compositions, as well as for the recognition of cryptic mantle metasomatism, and to infer melting processes. All these studies have been done on primary mantle phases. The amount of data available in the literature on purely metamorphic minerals of ultramafic rocks is scarce (Rampone, in prep.).
This study presents a collection of REE abundances in single mineral phases typical of metamorphic ultramafic parageneses and compares them with the REE content of primary mantle phases.
The goal is to evidence similarities and differences between metamorphic and primary REE patterns of distinct mineral phases and to investigate possible element mobility during secondary processes.
The REE measurements presented were performed using a laser ablation procedure combined with inductively coupled plasma mass spectrometry (LA-ICP-MS) at the "Institut für Isotopengeologie und
Mineralische Rohstoffe" of the ETH in Zürich. This technique represents a useful tool for the measure of trace and RE elements in single mineral phases. The detection limits achieved for most elements are 1-10 ppb.
6.3.2 Samples description
The LA-ICP-MS technique was applied to mineral phases of four metaperidotites of the
Chiavenna unit. The metamorphic mineral assemblages of the samples are listed in Table 6-1. 80 Geochemistry SS
Table 6-1: Mineral assemblages of metaperidotites samples investigated with the LA-ICP-MS method for REE determination
SAMPLE MINERAL ASSEMBLAGE
SU6 TYPE A METAPERIDOTITE OLIVINE+TREMOLITE+CHLORITE+MAGNETITE
SU76 TYPE C METAPERIDOTITE OLIVINE+MG-AMPHIBOLES+CHLORITE+MAGNETITE
BU27 TYPE B METAPERIDOTITE OLIVINE+TREMOLITE+MG-CUMMINGTONITE+TALC+CHLORITE+MAGNETITE
BU28 TYPE B METAPERIDOTITE OLIVINE+TREMOLITE+MG-CUMMINGTONITE+TALC+CHLORITE+MAGNETITE
The samples analysed consist of at least one example from each chemically distinct metamorphic group, as defined in section 6.2.3.
The analyses have been done by measuring each mineral phase with approximately ten laser- shoots for each sample. The ablated material was measured and visualised with a multi-element intensity versus time spectrum. The obtained element-intensities have been recalculated in terms of concentrations
(ppm) using program lamtrace of Jackson (1996).
6.3.3 Results
The REE elements of amphiboles from selected metaperidotite samples of the Chiavenna unit are normalised to primitive mantle composition after McDonough and Frey (1989) and plotted in spider diagrams (Figure 6-10). Element abundance of other phases such as olivine, talc, chlorite and magnetite are not shown because they contain concentrations lower than the machine detection limit (>10 ppb). All investigated samples are completely recrystallised (section 3.2) and only display metamorphic parageneses without any relic mantle phases.
From the comparison of the amphibole REE abundance with analogue bulk rock analyses of the same samples (Figure 6-11) it can be recognised that amphibole controls the entire bulk rock REE-composition of the analysed metaperidotites.
For comparison, in Figure 6-10a the REE contents of mantle clinopyroxenes (cpx) from the IL
(Rampone et al., 1996) and the compositional range of clinopyroxenes from abyssal peridotites (Johnson et al., 1990) are plotted. It can be recognised that the trace element profile of tremolite is parallel to the compositional pattern of clinopyroxene from the IL. The vertical shift displayed by the tremolite compared to cpx profiles is justified by the low Ca content of tremolite (ca. 12 wt%) compared to the approximately double content of Ca in cpx (ca. 22 wt%). This compositional difference results in a general dilution effect of the REE budget in tremolite. Despite the fact that tremolite essentially grows during metamorphism
replacing former diopside, the trace element content of this phase remains most probably close to the original element abundance of the primary clinopyroxene. These data suggest that during serpentinisation and metamorphism no significant rare earth element mobility occurred.
Anthophyllite represents the hydrous phase of sample SU76 (Figure 6-1 Ob). Contrary to the other samples, SU76 is interpreted to be a refractory metalherzolite in which the bulk rock composition is controlled mainly by serpentinisation (sections 6.2.1 and 6.2.3). 81 Geochemistry «sJllf
amphibole/primitive mantle amphibole/primitive mantle 10^ clinopyroxenes clinopyroxenes in peridotites from IL in peridotites from IL
1-
M
0.1- orthopyroxenes in peridotites from IL iälf clinopyroxenes : in abyssal peridotites clinopyroxenes in abyssal peridotites 0.01-
SU76 bulk rock
>. elements under SU6 detection-limite SU76 i I I I I I I I I I I I I "I—l—T" 1—I—I—I—T Ce Pr Nd Sm Eu Gd Dy Ho Er Tm Yb
10-
clinopyroxenes in abyssal peridotites 0.1^ clinopyroxenes in abyssal peridotites
clinopyroxenes rin peridotites from IL W 0.01- clinopyroxenes in peridotites from IL
BU27 BU28 0.001- T i i i i—i i r^ ~i—i—i—i—r Ce Pr Nd Sm Eu Gd Dy Ho Er Tm Yb Ce Pr Nd Sm Eu Gd Dy Ho Er Tm Yb
BU27 and BU28 contains also magnesio-cummingtonite
Figure 6-10: Mantle normalised spider diagram of La-ICP-MS analysed metaperidotite samples from the Chiavenna unit: grey field = compositional range of clinopyroxenes in abyssal peridotites (Johnson et al., 1990), dark grey field = compositional range of clinopyroxenes from the IL and dotted field = compositional range of orthopyroxenes from the IL, data from (Rampone et ai., 1996). a) REE abundance in tremolites of sample SU6, b) REE profiles of anthophyllites sample SU76, dashed line = elements under detection limit, c) and d) REE contents of tremolites, respectively in sample BU27 and BU28.
The plotted trace element contents are limited to the heavy rare earth elements. Elements with incompatible character higher than dysprosium display concentrations below detection limit. The fractionated amphibole patterns are similar to typical mantle orthopyroxene pattern (Figure 6-1 Ob) as for example to the IL orthopyroxene. The bulk rock composition of sample SU76 plots with slightly lower values parallel to the orthopyroxene REE abundance.
The last two samples (BU27 and BU28, respectively Figure 6-1 Oc and d) represent enriched metalherzolites with similar bulk rock compositions to normal fertile metalherzolites but with higher content in light REE. Similar to the tremolites of the metalherzolite SU6, the tremolites of these rocks represent the only important reservoir of trace elements of the entire rock and their REE abundance may confidently be used to represent the bulk rock composition. Compared to mantle clinopyroxene compositions, they fall approximately in the compositional trend of middle and heavy REE of SU6 tremolites and can be 82 Geochemistry
interpreted as replacing-phase of clinopyroxenes. The enriched light earth signature of these rocks
contrasts with the IL clinopyroxene and also with the signatures observed for the metamorphic tremolite of
SU6.
bulk-rock/primitive mantle 10^r Figure 6-11: Mantle normalised spider diagram of bulk rock composition of BU28 metaperidotite samples successively SU6 analysed for REE abundance in their single mineral phases. Dotted fields = compositional ranges of fertile
0.1- metalherzolites (A-type) and refractory metalherzolites (C-type), grey field = enriched metalherzolites (B-type), single A: fertile metalherzolites bulk rock of selected LA-ICP- 0.01- REE-patterns B: enriched metalherzolites MS analysed samples from the Chiavenna C: metalherzolites refractory unit and from IL, EL (Rampone et al., 1995; et al., and Lanzo 0.001- Rampone 1996) (Bodinier, "I 1 1 1 1 1 1 1 1 1 1 T" Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb 1988)
6.3.4 Significance of metamorphic amphibole for mantle processes
The presented LA-ICP-MS data on trace element concentrations in mineral phases of entirely
metamorphic ultramafics from the Chiavenna unit confirm that the major part of the REE budget is
concentrated in the stable hydrous phase. All other mineral species play only a secondary role. The REE
budget of the Chiavenna metalherzolites is mainly controlled by the hydrous phase tremolite, while in
those metalherzolites where Mg-amphiboles are the only amphibole phases also mineral phases such as
olivine and chlorite contribute to the bulk rock REE budget.
The comparison of the Ca-amphibole phases with primary mantle clinopyroxene compositions suggests
that high-grade metamorphic metaperidotites maintain the trace element signature of the former mantle
pyroxene. Consequently the REE signature of these metamorphic minerals can be applied to infer the
degree of depletion related to the mantle processes.
Serpentinisation has been verified on the basis of bulk-rock chemical changes in several
metalherzolites of the Chiavenna unit. Sample SU76 represents a serpentinised metaperidotite
characterised by a strong Ca loss where the low content of Ca may stabilise Mg rich phases rather than
Ca amphiboles. The comparison with typical mantle compositions shows that the bulk rock composition of
SU76 approximately reproduces the compositional range of the IL mantle orthopyroxenes. Compared to the other samples, the anthophyllite of SU76 is not directly replacing the former pyroxene phase as tremolite, but the Mg amphibole is growing through the metamorphic reaction involving talc and olivine.
Talc and olivine are replacing antigorite-bearing assemblages, which might have built after former orthopyroxene, consequently the present Mg amphibole maintains the REE budget of the former mantle orthopyroxene, if any significant element mobility during metamorphism occurs. 83 Geochemistry
In the case of the Chiavenna metaperidotites the particular REE distribution of the metamorphic amphibole phases can be used in inferring primary and secondary mantle processes.
6.4 Summary and discussion on the Geochemistry of the Chiavenna metaperidotites
The major and transition elements patterns (sections 6.2.1 and 6.2.2) indicate that the Chiavenna metaperidotites consist mainly of fertile and refractory metamorphic Iherzolites characterised by a slight depletion relative to primitive mantle abundance. The analysed samples are similar to the compositional fields of other massif peridotites such as the Lanzo, Ronda, Horoman, Malenco, Cima di Gagnone and the
Liguridi ultramafic rocks, but they also overlap with serpentinised and abyssal peridotites compositional fields. The high AI2O3 contents and the relatively high FeOtot, MgO, TiÛ2, Ni, Cr and Sc abundances confirm the relatively fertile character displayed by these rocks.
The REE analyses of the Chiavenna metaperidotites and of single mineral phases (sections 6.2.3 and
6.3.3) lead to the recognition of three compositional groups marked by different degrees of depletion relative to chondrite and primitive mantle compositions. A comparison of the analyses with available data from the literature argues for a complex evolution of the Chiavenna metaperidotites that involves both magmatic and alteration processes on the seafloor. Some of these processes are discussed in the following sections.
Melting processes: The specific shape of REE-depletion relative to primitive mantle and chondrite abundances and other incompatible elements can be related to different degrees of partial melting. For instance, partial melting events in the spinel stability field are described for the Lanzo (Bodinier, 1988) and
CdG peridotites (Pfiffner, 1999), while partial melting for the IL peridotites presumably occurs in the garnet stability field (Rampone et al., 1996). The Malenco peridotites (Müntener, 1997) are also derived from partial melting involving combined processes: an early event in the garnet stability field followed by a second episode in the spinel filed.
According to some authors, evidence for melting in the garnet stability field is testified by strong fractionation of the MREE to the HREE (Bodinier et al., 1988; McDonough and Frey, 1989). Hence, the relatively low Tb/Yb ratios exhibited by all types of metaperidotites from Chiavenna can be considered indicative of partial melting of a fertile Iherzolite host in the garnet stability field. Similar conclusions are reached considering the melting model of Frey et al. (1985) (Figure 6-12). The chondrite normalised Yb versus La plots show different theoretical melting trends ("fractional" or "batch" melting) computed for the garnet (Figure 6-12a) and the spinel (Figure 6-12b) stability fields. In Figure 6-12a part of the type A and C fertile and refractory metalherzolites plot within the calculated "equilibrium" melting residue field and correlate with different degrees of melting. The refractory metalherzolites would represent residues after a higher degree of melting while the more fertile metalherzolites correlate with lower degrees of melting or 84 Geochemistry plot outside the "batch" field next to the "fractional" melting field. By contrast, in Figure 6-12b the same rocks plot between the fields for "fractional" and "equilibrium" melting, which according to Frey et al. (1985) is typical for residual peridotites formed by processes intermediate between "fractional" and "equilibrium" melting.
Fractional Fractional residues melting melting residues ^
1 o 1 0 :
Equilibrium _Q melting >- Equilibrium residues melting , residues' •
La N
Figure 6-12: Melting models of Iherzolites after Frey et al. (1985); a) Calculated-melting model for the garnet stability field; b) Melting model for spinel Iherzolites; symbols: black diamonds = metalherzolites, grey diamonds = enriched metalherzolites, open diamonds = refractory metalherzolites. Values within parentheses indicate degree of melting calculated for two extremes REE mineral/melt partition coefficients: High set with HREE D***" ~ 1 and HREE D9rtx/me,t - Wand low set with HREE DcPx/met' -0.2 and HREE D***"8» ~ 4.
The analysed residual metaperidotites seem to be the product of a melting process that occurred in the garnet stability field rather than in the spinel field. The precise melting mechanism ("fractional" or "batch") remains uncertain.
On the other hand, the occurrence of negative Zr anomalies in the trace element patterns of the
Chiavenna residual metaperidotites (Figure 6-9) are inconsistent with the Tb/Yb signature and the results of the melting model of Frey et al. (1985). The latter two suggest that melting initiated in the garnet stability field, while residual garnet would produce positive Zr anomalies. Similar incompatible element behaviours are also observed in the IL residual peridotites, where the garnet breakdown erases the Zr but not the
Tb/Yb signature of the garnet stability field. In this case, the negative Zr anomalies are explained through a second melting event in the spinel field (Rampone et al., 1996).
In addition, other melting models (McDonough and Frey, 1989) were applied to the analysed rocks
(Figure 6-13). According to these models, beside the Tb/Yb ratio also the AI2O3 content of peridotites may be applied in distinguishing between partial melting processes occurred in the spinel or garnet stability fields, but for the analysed metaperidotites, the discrimination between garnet or spinel field residues is difficult to do and highly speculative.
Significant information on mantle processes are provided by the REE profiles of single mineral phases.
The REE patterns obtained for metamorphic amphiboles of the Chiavenna fertile and refractory metalherzolites (Figure 6-10) overlap with typical compositions of mantle clinopyroxene of the IL
(Rampone et al., 1996) and with the compositional range of clinopyroxene (cpx) from abyssal peridotites 85 Geochemistry
(Johnson et al., 1990). In particular, Johnson et al. (1990) showed that especially depleted LREE patterns of some abyssal peridotite cpx are too extreme to be modeled by "batch" melting alone, while "fractional" melting or melting and segregation in < 1% increments fit better the observed REE concentrations. Similar conclusions may be extended to the Chiavenna trace element profiles for the fertile and refractory metalherzolites which may represent the products of combined fractional and incremental melting processes, commence in the garnet stability field and successively in the spinel field.
Figure 6-13: Melting model after (McDonough and Frey, 1989), Tb/Yb (in ppm) versus Al203(in wt%), the star represents the initial chondritic mantle- 0.3- source composition, the two arrows indicate the fractionation followed after in the Chondritic ratio trends melting , Jï 0.2- garnet or spinel stability fields, diamonds = Chiavenna black = metalherzolites, spinel field metaperidotites: = open - refractory metalherzolites, grey enriched 0.1- o Ï metalherzolites. garnet field
0- I I I I 0 12 3 4 5 ai2o3
Type B enriched-metaperidotites patterns are distinct from the other metperidotites whole rock REE budget as well as amphibole REE compositions. Especially for these rocks a simple one phase melting model would not explain the singular element enrichment (for explanation see below).
Serpentinisation: The exposure of ultramafic rocks directly on the oceanic floor induces chemical reactions between rock and seawater. This metasomatic process, which may produce hydrothermal and ocean floor weathering alterations of the rocks, is known as serpentinisation. Various experiments have been performed by different authors (Seyfried and Dibble, 1980; Kimball et al., 1985; Menzies et al., 1993) in an attempt to evaluate the chemical changes occurring within peridotites when seawater reacts with mantle rocks. According to these experiments, serpentinised peridotites show an overall increase in silica and decrease in aluminium, magnesium and alkalis relative to their initial composition. The incompatible trace elements are barely or not obviously affected. In detail, the LREE are not preferentially mobilised during serpentinisation of clinopyroxene bearing peridotites while they are generally incorporated into secondary minerals during the alteration of clinopyroxene free peridotites. Only Sr (and its isotopes) seems to be influenced by the interaction with seawater, becoming enriched compared to normal Sr contents of an unaltered peridotite. The increase in Sr has two main explanations: i) new growth of Ca-rich phases such as anhydrite (CaSO-») and ii) the survival of primary clinopyroxene, which is considered a major Sr repository (Menzies et al., 1993). 86 Geochemistry
For the analysed metaperidotites of the Chiavenna unit the most important effects clearly related to serpentinisation are: i) strong Ca leaching and ii) negative Eu anomalies. Both features are particularly evident in those rocks, which underwent higher degrees of serpentinisation.
I. The Chiavenna metaperidotites doe not contain Ca rich phases such as anhydrite or primary
clinopyroxene. On the contrary, major elements such as Ca and Mg Figure 6-14 demonstrate that
the refractory metalherzolites (type C) display significantly lower Ca contents than the other rocks
and are characterised by higher Mg concentrations. This strong Ca leaching is the most
important and clear effect related to serpentinisation. Indeed, partial melting would induce a
residual peridotite to have a Ca/AI ratio higher than its initial value (Figure 6-2) while the analysed
rocks all clearly retain lower Ca/AI ratios than chondrite samples. All other major elements such
as Al, Ti and Si are not obviously affected by serpentinisation and especially the Al content of
these rocks is still comparable to primitive mantle concentration.
II. Among the rare earth element the negative europium anomalies is the only significant effect
related to serpentinisation processes. Negative Eu anomalies especially characterise the Ca
depleted refractory metalherzolites (C type) (Figure 6-8). The europium anomalies are probably
linked to the chemically reducing conditions that are set up during serpentinisation.
Figure 6-14: CaO versus MgO in wt.% contents for the Chiavennna metaperidotites: black filled diamonds
= fertile metalherzolites, gray diamonds = enriched metalherzolites, open diamonds - refractory metalherzolites.
Serpentinisation also explains the trend directed towards the opx-ol join displayed by the Chiavenna metaperidotites in the normative compositional diagram for ultramafic rocks in Figure 6-15. According to
Bodinier (1988) the vertical trend of the Lanzo peridotites pointing towards the olivine apex is related to incongruent melting of pyroxene, which shifts the evolving residual peridotites progressively towards dunitic compositions. The Chiavenna, Cima di Gagnone and Malenco metaperidotites also show similar
"oblique" trends which are not interprétable alone with partial melting processes, but rather assert 87 Geochemistry ' :s;::lsSsïi(ililIliliH chemical changes like decrease in Ca and passive enrichment in Si02, due to hydrothermal alteration and ocean floor weathering.
Opx Cpx
Figure 6-15: Classification diagram for ultramafic rock-composition: The plotted fields and points are norms calculated in wt.% olivine (01) clinopyroxene (Cpx) orthopyroxene (Opx) and spinel or garnet, projected from spinel or garnet onto the Ol-Opx-Cpx plane. Legend: black diamonds = Chiavenna metaperidotites (this study); grey diamonds = Cima di Gagnone peridotites (Pfiffner, 199; Trommsdorff, unpublished data); open squares = Malenco peridotites (Müntener, 1997); black arrow = Lanzo peridotites trend (Bodinier, 1988); dotted field = abyssal peridotites (Dick et al., 1984).
Secondary processes: In addition to melting and serpentinisation, other mantle processes are probably involved in the singular chemical enrichment in light REE of type B metaperidotites. Light REE enriched profiles also characterise the amphibole phases of type B metaperidotites. In order to verify if these trends are the product of secondary enrichment processes or rather linked to the primary REE budget of fertile
Iherzolites, a comparison between the REE composition of the tremolites and typical REE patterns of mantle clinopyroxene from fertile Iherzolites such as the peridotites from the External Liguridi (EL)
(Rampone et al., 1995) is proposed (Figure 6-16). The comparison shows that the tremolites do not reproduce the composition of clinopyroxene from fertile Iherzolites. On the contrary, the tremolites of type
B metaperidotites probably replace former clinopyroxene, which was already fractionated in its
HREE/LREE composition similarly to the depleted tremolite profiles of type A metaperidotites, which with
AI2O3 contents between 3.2 and 3.7 wt.% are clearly the most fertile metalherzolites of the Chiavenna unit.
Furthermore, the fact that type B metaperidotites do not fit in any partial melting model (Frey et al., 1985)
(Figure 6-12) plotting for La composition outside of any field is probably best explained by mantle metasomatic processes of unsolved origin which enriched some of the already depleted Iherzolites with light rare earth elements. Geochemistry
amphibole/chondrite Figure 6-16: Chondrite normalised REE patterns measured in tremolite of type B enriched metalherzolites and in
10t clinopyroxene from the EL Iherzolites (Rampone et al., 1995) (chondrite tremolite average after and composition of composition (McDonough Frey, enriched metalherzolites 1989).
1 -
01 ~i i i—i—i i i i—i—i—i—r Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb
6.5 Metabasic rocks
The metabasic rocks consist of different lithotypes (section geol.), which were described and interpreted as gabbro layers overlain by volcanic material by Schmutz (1976) and by Huber and Marquer
(1998) while bulk chemical analyses published by Dürr et al. (1993) suggested for these rocks a tholeiitic trend with MORB affinity.
The proposals of the following sections are:
Compare the compositional field of the analysed rocks with typical compositional fields of ophiolitic
mafic bodies from the Alpine belt and from outside the Alps.
Identify the unmetamorphic protolithic source.
Link the rocks to a possible tectonic setting of emplacement.
This study also documents the first occurrence of metamorphosed rodingites in the Chiavenna unit.
These Ca-enriched basic rocks are exposed as small boudinaged mafic lenses in a few localities within the metaperidotites (see geology). Rodingite are Ca enriched mafic rocks hosted in ultramafic lenses and usually associated with the serpentinisation of the host peridotite (Coleman, 1966; Rice, 1983) (section
6.4). The occurrence of metamorphosed rodingites within the Chiavenna metaperidotites implies that the peridotites were formerly in an environment where serpentinisation took place, which supports the idea that the Chiavenna Unit may form a part of an ophiolitic sequence.
Three bulk chemical analyses were performed on samples of boudinaged mafic rocks suspected to represent metarodingites. Only one sample (BM34) can be considered a metarodingite sensu strictu. The other samples display only a modest Ca enrichment and do not fall within the compositional field of other
Alpine rodingites (see below). Their compositional features are plotted together with the amphibolites. 89 Geochemistry
6.5.1 Major and Minor element-chemistry of amphibolites
The compositional variation diagrams of major element oxides (Si02, CaO, AI2O3, TO2, MgO, Na2Ü and
FeOtot, in wt% versus the Ml) show that the Chiavenna amphibolites plot along approximately linear correlation trends (Figure 6-17). The mafic index (Ml) of the rocks is used as a proxy of the degree of igneous fractionation and varies between 0.4 and 0.7. The Ti02, Na2Û and FeOtot contents increase with increasing Ml values while AI2O3 and MgO decrease, other oxides such as SiÛ2, and CaO contents do not significantly change their contents by increasing Ml. The Ti, Fe and Na oxides behave as incompatible elements and become enriched in the residual melt.
The unusual CaO and TiÛ2 contents of sample SM43 are probably due to the fact that this sample forms an amphibolite breccia within metacarbonate rocks (section 2.2.2). Consequently, the Ca- enrichment and the Ti-leaching are best explained by the mobility of these elements during metasomatism.
A comparison of the data with the gabbros from the northern Apennines ophiolitic zone (NAO) (Tiepolo et al., 1997), from the Western Alps (Pognante et al., 1982) and from the Bellinzona-Dascio zone (Stucki, pers. comm.) reveals that these rocks are characterised by lower Ml values (MI=0.2-0.45). Their MgO and
AI2O3 contents are higher than in Alpine and fresh basalts, while the Si02, Na2Û, K2O and CaO contents are partially comparable with typical concentration of basalts. The FeOtot, and TiÛ2 seem to be less abundant in the gabbroic rocks with a certain overlap limited to more evolved gabbros with the compositional field of the basaltic rocks. However, the TiÛ2 and the FeOtot contents of the gabbroic rocks from the NAO zone are clearly higher than in the Chiavenna mafic rocks (Figure 6-18). Furthermore, the
Chiavenna amphibolites plot in agreement with the compositional fields of other basaltic rocks such as the metabasalts from the CdG (Pfiffner, 1999) and from the Forno Units (Puschnig, 1998) as well as with the field of basaltic dykes from the EL (Vannucci et al., 1993).
The analysed amphibolites plot in agreement with the compositional fields of Alpine basaltic rocks and with the field of fresh mid-ocean ridge basalts. A detailed comparison shows that the Chiavenna amphibolites have Na2Û contents slightly higher but comparable with the CdG amphibolites but significantly lower than the majority of the Forno amphibolites indicating a specific affinity to the CdG amphibolites but only restricted to the Na20 content.
The minor element distribution (Ni, Zr, Y, and Cr versus the Ml) exhibits analogue fractionation trends of the major elements (Figure 6-19). The Ni and Cr contents decrease with increasing Ml while Y, Zr and V correlate positively with increasing fractionation degree with the exception of sample BM11, where the extremely high Y and Zr concentrations are related to the presence of small zircons. On the contrary the high Sr variability (not plotted) between 5-300 ppm displayed by all samples suggests a possible mobility of this element during, for example ocean-floor metamorphism, or during any other successive metasomatic event. Geochemistry
Figure 6-17: Major element oxides distribution expressed in wt.% versus Ml: Ml = FeOiot I FeOtot + MgO, black filled diamonds = Chiavenna amphibolites, grey squares = Chiavenna metarodingites, open diamonds - CdG amphibolites (Pfiffner, 1999), half filled diamonds = amphibolites from the Forno Unit (Puschnig, 1998), open squares = basaltic dykes from the EL (Vannucci et al., 1993), black dots = Mg- and FeTi-gabbros from the Indian ocean (Engel and Fisher, 1975), open dots = Fe-gabbros from the NAO zone (Tiepolo et al., 1997) and the Bellinzona-Dascio zone (Stucki, pers. comm.), closed dotted field = compositional Fe-gabbros from the Western Alps (Pognante et ai, 1982), grey field = compositional field fresh mid ocean ridge basalts (Schilling et al., 1985), open dotted field - compositional field Mg-gabbros NAO zone (Tiepolo et al., 1997). 91 Geochemistry
Figure 6-18: FeOtot versus T1O2 contents (expressed in weight %) of the Chiavenna amphibolites (black diamonds) and metarodingites (grey squares) as well as of metabasalts from CdG (Pfiffner, 1999) (open diamonds) and Forno units (Puschnig, 1998) (half filled diamonds), of basaltic dykes from the EL (Vannucci et al., 1993) (open squares) and the Mg- and Fe-gabbros from the NAO zone (Tiepolo et al., 1997) (open circles).
The distribution of the major and minor element establishes the compositional differences and
analogies between the analysed amphibolites and the compositions of Alpine and other then Alpine
gabbro or basaltic rocks used as comparison. It can be concluded that the Chiavenna amphibolites plot in
agreement with the compositional fields of fresh mid-ocean ridge basalts (Schilling et al., 1985) as well as
of metabasalts and basalts from the Alpine belt.
Figure 6-19: Minor element concentration expressed in ppm versus Ml: Ml = FeOtot I FeOtot + MgO, black diamonds = Chiavenna amphibolites; grey squares = Chiavenna metarodingites, open diamonds = amphibolites from CdG (Pfiffner, 1999), half filled diamonds = amphibolites Forno unit (Puschnig, 1998), open squares = basaltic dykes EL (Vannucci et al., 1993), black circles = Mg- and FeTi-gabbros from the Indian ocean (Engel and Fisher, 1975), closed dotted field = gabbros from the NAO zone (Tiepolo et al., 1997) and from the Western Alps (Pognante et ai., 1982), open dotted field = Mg-gabbros from the NAO zone (Tiepolo et al., 1997) and from the Bellinzona-Dascio zone (Stucki, pers. comm.), dashed field = fresh mid-ocean ridge basalts (Schilling et al., 1985). 92 Geochemistry
By contrast, the gabbros exhibit distinct compositional features from the basalts, especially in their TO2,
MgO (Figure 6-20a) and Ni contents (Figure 6-20b) where the analysed amphibolites occupy an intermediate compositional field similar to other Alpine amphibolites and basalts and always outside the fields outlined by Ol-gabbros and Fe-gabbros.
1000 MgO
Figure 6-20: a) T/O2 versus MgO contents (in wt%); black diamonds = Chiavenna mafic rocks, black circles = FeTi- and Mg-gabbros from the Indian ocean (Engel and Fisher, 1975), grey circles = Fe-gabbros from the Western Alps (Pognante et ai, 1982), open grey dots and open circles = Ol-gabbros and 01- cumulus from the NAO zone (Tiepolo et al., 1997), open diamonds and triangle = amphibolites respectively, from CdG (Pfiffner, 1999) and Forno Unit (Puschnig, 1998). b) T1O2 versus Ni contents (in wt% and ppm, respectively); black diamonds - Chiavenna mafic rocks, black dots = FeTi- and Mg-gabbros from the Indian ocean (Engel and Fisher, 1975), grey circles = Fe-gabbros from the Western Alps (Pognante et al., 1982), open grey and open circles = Ol-gabbros and Ol-cumulus from the NAO zone (Tiepolo et al., 1997), open triangle and diamonds = amphibolites from CdG (Pfiffner, 1999).
6.5.2 Magmatic classification
Based on the hypothesis that the Chiavenna amphibolites represent former basalts rather than gabbros, it is proposed to discuss the data in terms of chemical classification for volcanic rocks. A first fundamental distinction among volcanic rocks can be made on the basis of the bulk alkaline contents (total alkali TA) and the SiÛ2 contents (S) used to subdivide basic rocks into highly alkaline, alkaline and subalkaline categories (Irvine and Baragar, 1971). The TAS content of the rocks suggests a subalkaline derivation for the analysed amphibolites (Figure 6-21).
~* ' Figure 6-21: Simplified TAS diagram which permits distinction between alkaline and subalkaline basic
rocks: black diamonds = Chiavenna mafic rocks, curve 1 = compositional line separating alkaline from subalkaline rocks after Irvine and Baragar (1971). 93 Geochemistry
Moreover, based on the sum of alkali, FeOtot and MgO (AFM) contents the subalkaline rocks may be subdivided into calcalkaline and tholeiitic series. The AFM content (Figure 6-22) of the Chiavenna amphibolites and metarodingites associates the rocks to subalkaline basalts of tholeiitic composition. The rock compositions plot approximately along a line parallel to the M-F axis and pointing towards the iron vertex (F). This line may approximate a bundle of possible subparallel silica liquid lines of descent followed by the melt/s during its evolution. Compared to the compositional field of fresh mid ocean ridge basalts the
Chiavenna amphibolites and metarodingites represent slightly more evolved former tholeiitic basalts. The tholeiitic affiliation is furthermore supported by the low K2O contents (0.09-0.73 wt %) and by the relatively variable T1O2 rates (1.02-2.16 wt %) of the Chiavenna mafic rocks, attested to be typical attributes of tholeiitic mafic rocks (Peccerillo and Taylor, 1976).
F Figure 6-22: AFM diagram for magmatic rocks with tholeiitic and calcalkaline evolution trends; A = Na20+ K20; F = FeOtot; M = MgO, black diamonds = Chiavenna amphibolites, dotted field = compositional field of fresh mid ocean ridge basalts, curves: 1 after Irvine and Baragar (1971), 2 after Kuno (1968).
A 80% 60% 40% 20% M
6.5.3 TRACE AND RARE EARTH ELEMENTS IN AMPHIBOLITES
The chondrite normalised rare earth elements abundance of the Chiavenna amphibolites show flat
patterns with REE contents in the range of 10 and 50 times chondrite except for an enrichment in La
(Figure 6-23). Also the metarodingites display flat patterns not distinguishable from the amphibolites. As suggested by the major element distribution both lithologies show a strong affinity to a tholeiitic composition. This property is also confirmed by the comparison of the REE abundance with typical trace element patterns of different MORB compositions such as N-type, E-type (Sun and McDonough, 1989) and T-type (Walker, 1991). The profiles are generally parallel to "normal" type MORB pattern displaying a constant heavy and middle REE enrichment accompanied by a slight depletion in LREE, which is characteristic for typical N-type MORB. Additionally, comparisons with REE abundance of pillow basalt and gabbro compositions of the Internal Liguridi (IL) (Rampone et al., 1998), establish a nearly perfect overlap with the REE abundance of the pillow basalts while correlation with the trace element pattern of the gabbro is lacking. Gabbros are generally characterised by distinct more depleted chondrite enrichment 94 Geochemistry trends usually marked by positive europium anomalies never observed in the Chiavenna amphibolites and metarodingites.
i—i—i—i—i—i—i—i—i—i—i—i—i—r 1~n—i—i—i—i—i—i—i—i—i—i—i—i—r La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr NdSmEuGd Tb Dy Ho Er Tm Yb Lu
Figure 6-23: a-c) Chondrite normalised REE patterns of the Chiavenna amphibolites and metarodingites. d) Chondrite normalised REE patterns of N-type, E-type (Sun and McDonough, 1989) and T-type MORB (Walker, 1991) as well as selected REE patterns of pillow basalts and gabbros from the IL (Rampone et al., 1998).
It is noteworthy that most samples display erroneously high La, Ce, Pr, Nd and Sm values. These values are related to analytical and technical problems and are not linked to primary compositional features of the rocks (see Appendix). Nevertheless, the systematic of these erroneously high values preserves the original shape of the N-type MORB patterns. In contrast with the N-type MORB trend is the high La enrichment affecting the majority of the samples. This positive La anomaly is interpreted to be of secondary origin, probably related to a contaminating pervasive La-enriched metamorphic fluid.
The primitive mantle normalised REE spider diagrams (Figure 6-24) indicate a general enrichment of all trace elements, with some marked negative anomalies for Zr, Hf, Th and Nb which are not of primary origin. The Zr and Hf anomalies are possibly related to the presence of not dissolved zirconium crystals in the acid-rock powder liquids used for the measurement of trace elements with the inductively coupled 95 Geochemistry plasma mass spectrometry technique (ICP-MS). In addition, the Th, Nb and Ta abundance displays high standard deviations because of analytical errors.
Despite the technical problems, the mantle normalised patterns of the Chiavenna amphibolites and metarodingites are marked by high U contents compared to N-type MORB compositions. High U contents are also observed characterising the REE budget of the Chiavenna metaperidotites (section 6.2.3). In both lithologies these anomalies may be accounted to secondary contamination processes linked to the interaction between the Chiavenna unit and the surrounding basement nappes probably during the Alpine tectonics.
Sample/Primitive mantle Sample/Primitive mantle 100 1000
100-7
10-
1-
0 1 'I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Rb U Ta Ce Sr Sm Hf Ti Tb Ho Tm Yb Rb U Ta Ce Sr Sm Hf Ti Tb Ho Tm Yb Th Nb La Pr Nd Zr Eu Gd Dy Er Y Lu Th Nb La Pr Nd Zr Eu Gd Dy Er Y Lu
Sample/Primitive mantle Sample/Primitive mantle 100^ 100
E-type MORB
10-
0.1- I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I l i i i i i i i i i i Rb U Ta Ce Sr Sm Hf Ti Tb Ho Tm Yb Rb U Ta Ce Sr Sm Hf Ti Tb Ho Tm Yb Th Nb La Pr Nd Zr Eu Gd Dy Er Y Lu Th Nb La Pr Nd Zr Eu Gd Dy Er Y Lu
Figure 6-24: Primitive mantle (Sun and McDonough, 1989) normalised trace element patterns of the Chiavenna amphibolites and metarodingites as well as typical N- and E-type MORB trace element compositions (Sun and McDonough, 1989).
The normalised chondrite (Figure 6-23) as well as the primitive mantle (Figure 6-25) profiles of the
Chiavenna mafic rocks are marked by scattered enrichment factors. To distinguish whether these features are related to magmatic differentiation processes or whether they are a consequence of additional partial melting mechanisms the data are investigated for single element ratios distribution such as (Ce/Yb)N and
(Sm/Yb)N (Figure 6-25). Differentiation processes increase the amount of incompatible elements by 96 Geochemistry increasing fractionation degree (Ml). Partial melting processes would determine a variation in the element concentration independently from the Ml value inducing a shift of the element ratio towards higher values at relatively low Ml ratios. The analysed (Ce/Yb)N and (Sm/Yb)N ratios distribution correlates positively with the Ml index where the increase of the ratios seems to be controlled by igneous fractionation rather than by partial melting. Sample BM4 with relatively high ratios is the only exception to this trend and needs to be explained with different processes. This sample is exposed close to the basement rocks of the Tambo nappe, which may be responsible of secondary element contamination between amphibolites and crust material such as gneiss.
1.60 1 .DU b) BM4
1.45-
1.30-
1.15- m 1|a 1.00-
0 85-
I i i i 0.3 0.4 0.5 0.6 0.7 0.8 Ml
Figure 6-25: Chondrite-normalised REE ratios versus Ml of amphibolites (black filled diamonds) and metarodingites (grey filled squares), a) Chondrite normalised Ce/Yb distribution versus Ml. b) Chondrite normalised Sm/Yb distribution versus Ml. The arrow shows the ratio trend followed during magmatic differentiation.
Comparison of the Chiavenna basic rocks with a collection of REE-data of basalts from the Alpine belt
(Venturelli et al., 1981) shows a nearly identical match of the fields (Figure 6-26). Therefore, it is proposed that the banded amphibolites of the Chiavenna unit are crystallised liquids and not gabbro cumulates.
However, the massive amphibolites, chemically identical to the fine-grained banded amphibolites may represent shallow depth dolerite bodies on the basis of textural evidences (section 2.2.2).
Bulk-rock/chondrite 100 Figure 6-26: Chondrite normalised compositional field of the Chiavenna mafic rocks (dotted field) and compositional field of basalts from the Alpine belt (Venturelli et al., 1981) (grey field).
-i—i—i—i—i—i—i—i—i—i—i—i—i—r La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 97 Geochemistry
6.5.4 Rodingites
The metamorphosed rodingites of the Chiavenna Unit display a general increase of CaO and MgO contents and a slight decrease of SiÛ2, AI2O3 and Na2Û. The other major element concentration displays compositional ranges similar to the amphibolites (Figure 6-17). The minor element contents of Y, Cr and
Zr (Figure 6-19) fall within the range of tholeiitic composition. By contrast, the Ni values are slightly more scattered compared to the amphibolites but still in agreement with typical basaltic composition. Sample
BM34 is the most rodingitic rock among the analysed samples. It is characterised by high CaO values
(25% in wt) compared to normal Ca content of mafic rocks (Ca around 15 wt %) and by a significant low
Na2Û content (Figure 6-27). During rodingitisation Ca is mobilised from the host peridotite towards the mafic intrusion while Na2Û and K2O are leached away from the mafic system (Coleman, 1966; Rice,
1983). In the studied metarodingites the alkalis are accommodated behind the rodingite in the blackwall- site as shown by sample BM8. The inverse correlation between Na2Û and CaO contents in rodingites best distinguishes this metasomatic alteration. Figure 6-27 shows a comparison between amphibolites and metarodingites of this study and typical metamorphosed rodingites from the CdG unit (Pfiffner, 1999).
Figure 6-27: Inverse correlation between Na20 and CaO (in wt%) in mafic rocks from Chiavenna and CdG: Symbols: black diamonds = Chiavenna amphibolites, grey squares = Chiavenna metarodingites, open squares = CdG rodingites (Pfiffner, 1999).
6.5.5 Tectono-magmatic interpretation
Based on bulk rock chemistry the Chiavenna banded and massive amphibolites represent former tholeiitic basalts with trace element patterns matching the compositional trend of N-type MORB, while detailed textural evidences demonstrate that the massive amphibolites are former shallow depth mafic bodies such as dolerite. The following diagrams represent a selection of discrimination diagrams applied to volcanic rocks in the attempt to affiliate them to a possible tectonic environment. The most classical and fully tested ones were proposed by (Pearce and Cann, 1973):
• The Ti-Zr-Y contents in basaltic rocks are proposed to discriminate between within-plate basalts,
ocean-island or continental flood basalts and other basalt types (Figure 6-28). The majority of the
Chiavenna basaltic rocks fall within field B (MORB and island arc tholeiites), with a few samples
plotting outside any field. 98 Geochemistry
Ti/100
Y*3
Figure 6-28: Ti-Zr-Y (in ppm) triplot: black diamonds=Chiavenna metabasic rocks; compositional fields of basalts after Pearce and Cann (1973), A = island-arc tholeiites, B = MORB, island-arc tholeiites and calc-
= alkali basalts, C - calc-alkali basalts, D within plate basalts.
• The Ti versus Zr diagram (Figure 6-29) distinguishes between distinct calcalkaline and tholeiitic basalt
scenarios. The majority of the Chiavenna amphibolites plot within the MORB field (excluding island
arc tholeiites), although a few samples fall outside the MORB field (D). However, high Zr values may
be related to the presence of small zircons in the samples.
15000-
10000-
5000-
—i— 50 ~ïoo i5o" 200 250 Zr
Figure 6-29: Ti-Zr (in ppm) discrimination diagram after Pearce and Cann (1973); Black diamonds and grey squares respectively Chiavenna amphibolites and metarodingites; Fields: A = island-arc tholeiites, B
= MORB, calc-alkali basalts and island-arc tholeiites, C - calc-alkali basalts, D = MORB.
Finally also the vanadium versus titanium plot (Figure 6-30) suggested by Shervais (1982) offers a further indication for a MORB or BAB (back-arc basin basalts) affinity of the Chiavenna metabasalts. In agreement with the discrimination diagrams of Pearce and Cann (1973) the common scenario to all three types of diagrams is an ocean ridge origin as confirmed by the REE distribution. 99 Geochemistry
Figure 6-30: V versus Ti/1000 (in ppm) after Shervais (1982). Black and grey diamonds respectively Chiavenna amphibolites (black diamonds) and metarodingites (grey squares).
6.6 Metacarbonates and Metaophicarbonatic rocks
The Chiavenna lithological sequence also contains small bands of metacarbonate rocks outcropping in a few localities (section 2.2.3). These rocks consist of folded bands of calcite-marbles of different grain size and locally, they contain centimetres long deformed green-nodules of Cr-rich calcsilicates. The ophicarbonate rocks are rare. Mostly, they form thin veins and pockets within the ultramafic rocks (section
2.2.1). The following sections deal with the bulk rock chemistry of these lithologies in order to constrain their nature. The carbonate and ophicarbonate chemistry is also compared with the metaperidotite and amphibolite compositions of the Chiavenna unit as well as with marbles and calcsilicates from the CdG
Unit (Neuenschwander, 1996; Pfiffner, 1999) and with fresh caldtised serpentinite breccias, probably of sedimentary origin drilled along the western Iberian passive continental margin (Sawyer et al., 1994b). All plotted data are recalculated on a volatile component-free basis.
6.6.1 Compositional Features and Discussion of the metacarbonates data
The compositional variability of the calcite-marbles and the ophicarbonates is outlined in variation- diagrams where the MgO content was chosen as chemical discriminate factor between the different lithologies and in addition, it may be useful in distinguishing between mixing features of distinct rock- chemistry.
The calcite-marbles of the Chiavenna unit are characterised by low (<10% in wt) MgO content, and they plot in the same major element compositional field (SiÛ2, Ti02, MnO, Na20, FeOtot and CaO in Figure
6-31) as the calcite-marbles from the CdG Unit. Distinct from the CdG-marbles, the analysed metacarbonates display higher and more scattered Si02 values comparable with Si-contents of the
Chiavenna amphibolites. The ÜO2 and the CaO contents of the marbles are always higher than the 100 Geochemistry corresponding values of other rock-types, while compared to ultramafic, mafic and calcsilicates compositions, the FeOtot and the Na2Û contents of the marbles are lower.
70 06
o 60 05
SU51C UM 04 SC12 ^-"*
^SO103 03 CM O
SO100 02
SU51C 01 S0103 UM SO103 - SO100 "• SC12 00 -* «•- i ^ i r^ n
30 SO100
O CN20
SO103 SU51C
UM 1 o
^-•SOKB--8012-^^ SU51C UM • SOI 03 SQiSQ12SO100250100^
00 T^ I SO103
10 UM 60 'àô^( 103 8 o SO103 SO100 G~ O 40 SU51C O m SO103 SC12 4 SC12 SU51C s 20 - SO100 * t • •W, SO103 $ > UM
-1— -1 1— —i— —i 1— 4& 10 20 30 40 50 10 20 30 40 50 MgO MgO
Figure 6-31: Major oxide composition (Si02, T1O2, Na20, FeOtot, CaO and MnO in wt%) of the calcite- marbles, calcsilicates and ophicarbonates of the Chiavenna Unit. Solid symbols correspond to rocks of the Chiavenna unit from this study, open symbols are samples from CdG (Neuenschwander, 1996; Pfiffner, 1999), the half solid symbols are samples from the Iberia margin (Sawyer et al., 1994a): diamonds = calcite-marbles, circles = calcsilicates, squares = ophicarbonates. The dotted fields represent the compositional variation of amphibolites and ultramafic rocks from the Chiavenna unit while the dashed field corresponds to the composition of the calcsilicates from CdG.
Among the measured minor elements, Y, Zr, Cr and Ni (Figure 6-32) are chosen to represent the rather simple rock-chemistry of the calcite marbles. From the diagrams it can be seen that the marbles are relatively poor in ultramafic components such as Cr and Ni compared to the other outlined rock-types. By contrast, the incompatible elements Y and Zr are present in high amounts and are scattered between 20-
70 ppm (Y) and 25-120 ppm (Zr). Compared to the CdG marbles, the Chiavenna marbles contain higher 101 Geochemistry amounts of Y and Zr comparable to similar element-concentration of the Chiavenna amphibolites. The Zr- abundance is also comparable with Zr-contents recorded in calcsilicates from the CdG Unit. However, samples with considerably high Zr content (approximately 100 ppm) may contain not dissolved zircon crystals or detritic zircons which were described for the CdG calcite marbles (Pfiffner, 1999). The bulk rock characteristics of the calcite marbles identify a group of rock with peculiar chemical features distinct from other rock compositions excluding for the formation of these rocks any mixing process between different rock chemistry. It is proposed to consider the calcite marbles former carbonates probably of sedimentary or biochemical origin.
80 - 175-
150-
60 - 125-
,_100- > 40 - *#03 N 75- SÖ103 50- 20 - SO100 A £ SU51G3C12 25- S0100^ SO103^ UM SU51CSC12* UM I l P "( 1 P sCl03 '
5000 - 2000- UM
4000 - UM 1500-
Ö 3000 - -z. SC12
SU51C 1000- • SU51c"sBl03 - SO103A«. 2000 • s5i2 500- 1000 - SÖ103 S°100 ^. «SO103 SO100 ^ 0 - ^* "^ I* I oH i i i i 0 10 20 30 40 50 c 10 20 30 40 5I MgO MgO
Figure 6-32: Y, Zr, Cr and Ni (in ppm) versus MgO (in wt%) plots. Black diamonds = Chiavenna calcite- marbles, black squares = Chiavenna ophicarbonatic rocks, black triangle = Cr-rich nodule, black points = Chiavenna calcsilicates, open diamonds = calcite-marbles CdG, half filled diamonds = caldtised serpentinite breccias (ophicarbonates) Iberian margin, UM signed field = compositional field of Chiavenna ultramafites, dotted field = compositional range of calcsilicates CdG, dashed field = compositional range Chiavenna amphibolites.
The processes involved in the formation of the ophicarbonates of the Chiavenna unit may have been different; metasomatic, tectonic or of sedimentary nature. For the ophicarbonatic rocks from the
Chiavenna Unit the original textural characteristics of the rocks are no more available. As a matter of fact, the intense Alpine deformation that overprinted the area and the relatively small centimetre scale of the fractures and pockets filled with ophicarbonatic material obliterate any primary texture. The major oxide element distribution (Figure 6-31) shows that the compositional signature of the ophicarbonates with MgO values higher than 25% in weight is intermediate between the Chiavenna calcite marbles and 102 Geochemistry
metaperidotites compositions. The Si02, FeOtot, CaO, MnO and Na2Û contents vary within the two extreme carbonate and ultramafic compsositions. The TiÛ2 content with relatively low values plots distinctly outside the mixing line between calcite marbles and ultramafitites. With the exception of the low
Fe and high Mn values, all major element concentrations fall within the compositional range existing in between the chemical components of caldtised serpentinite breccias from the Atlantic sea (Sawyer et al.,
1994a). Similarly, the trace element abundance (Figure 6-32) with extremely low Y and Zr contents and higher values for the compatible elements Ni and Cr indicate compositions intermediate between the carbonatic and ultramafic extremes. Two of the analysed ophicarbonates (SO103C and SU51C) are of special interest. They are diopside-rich fragments of ophicarbonatic veins within ultramafic rocks. Sample
SO103 is characterised by high values of Zr (49 ppm) and Y (41 ppm) comparable with Zr and Y abundance of the calcite-marbles. The same sample displays major-element contents that are in agreement with the related contents of the calcite component of the serpentinite breccias. By contrast, in sample SU51C, Zr and Y are present only in trace amounts similar to ultramafic rocks. It is evident that the ophicarbonatic material filling veins and pockets within the metaperidotites rock has mixed features between typical calcite marbles and ultramafic rocks of the Chiavenna Unit. This hypothesis would suggests a sedimentary or tectonic origin for these rocks where the mantle rocks were exhumed and fractured on the ocean floor and successively filled with carbonate material of sedimentary origin. This hypothesis is also supported by the recover of Cr rich calcsilicate nodules within the calcite marbles
(section 2.2.3). These Cr enriched boudinaged thin layers unusual for pure carbonates imply a direct contact between the calcite marbles and the ultramafic rocks and possible sedimentary or tectonic mechanisms determined the assimilation of ultramafic material within the carbonates.
6.7 Conclusions
The geochemical characterisation of the metamorphic ultramafic, mafic and carbonate rock-sequence leads to significant conclusions.
The metaperidotites of the Chiavenna unit can be considered former Iherzolites of distinct features, which document a composite mantle history:
Early partial melting processes, probably commenced in the garnet stability field and followed by
processes in the spinel stability field, depleted the Iherzolites in light rare earth elements.
Successively, part of these depleted mantle residua (type B enriched metalherzolites) underwent
cryptic mantle metasomatism, which selectively enriched the mantle rocks in light REE.
Subsequently, the peridotites were variously serpentinised by the exposure on the oceanic seafloor.
The Chiavenna amphibolites are subdivided in banded and massive amphibolites. Both group of rocks display identical chemical features. They are tholeiitic in composition and show a strong affinity for N-type 103 Geochemistry mid ocean ridge basalts (MORB). Despite the relatively modest Ca-enrichment, also the analysed rodingite samples are chemically indistinguishable from the amphibolites.
The banded amphibolites may represent crystallised liquids while the massive amphibolites are
former shallow depth dolerite bodies.
The boudinaged rodingites are mafic dikes similar to the amphibolites, which crosscut the
metaperidotites before the Alpine deformation phase and during serpentinisation. The occurrence of
boudins parallels to the foliation of the host metaperidotite supports this hypothesis.
None of the amphibolites show textural or compositional evidence of gabbros.
The calcite-marbles intimately associated with the amphibolites are interpreted to represent former carbonates of sedimentary origin.
The oceanic stage for the entire sequence is documented by the occurrence of:
Serpentinised metaperidotites.
Rodingite dikes within the metaperidotites.
Ophicarbonatic material filling fractures and pockets within the metaperidotites.
Cr-rich boudinaged nodules within the calcite-marbles that suggest a direct contact between
ultramafic and carbonate rocks.
Mafic breccias embedded within the calcite-marbles, which argue for tectonic or sedimentary
processes involving the two lithologies before the Alpine collision.
105 Dumortierite
7 Dumortierite
This chapter focus on to the boron-rich mineral dumortierite, which was observed and collected within the study area. Rarely abundant, dumortierite is reported to occur at numerous localities throughout the world. It is considered an important rock-forming mineral that acts as a boron sink under certain PT and bulk chemical compositions.
Boron plays a significant role in the evolution of geological systems due to its influence on several transport processes. In particular, because of the strong partitioning of boron into the fluid phase under many geological conditions, boron is considered as an indicator of subsurface fluid-rock interaction
(hydrothermal processes related to the formation of e.g. pegmatites, veins, dehydration and melting reactions) and as a tracer in surface geochemical processes. The occurrence of boron-bearing minerals such as tourmaline, dumortierite, grandidierite, kornerupine, axinite, etc. is therefore significant as they may provide additional information on the behavior of boron during many geological processes.
Dumortierite was detected in the investigated area, as small needles in pegmatite and aplitic dikes within gneiss blocks not found in situ. The blocks are found along the Schiesone River near Prata
Camportaccio, where the Chiavenna unit is underlain by the Gruf unit. The blocks are supposed to be part of this latter gneiss unit where analogous lithologies can be found. The dumortierite needles (Figure 7-1 a) are of intense blue color and a few tens of millimeters in size.
Figure 7-1: Dumortierite needles in pegmatites and aplitic dikes within gneiss blocks of the Gruf unit, b) Photomicrograph of dumortierite needles under plane polars.
In thin-section, the mineral is transparent with a vitreous shine and displays a strong blue to violet pleochroism (Figure 7-1 b). The needles are formed by thin, prismatic blasts with negative elongation. They are associated with potassium-feldspar, quartz, biotite, white mica and rare small garnet grains. 106 Dumortierite ^ym^^^^^^^^mÊÊÊÊMMÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊÊm
7.1 The Gruf complex
The dumortierite needles described in this study were found within aplitic and pegmatitic dikes that crosscut gneiss blocks supposed to be part of the adjacent migmatitic Gruf unit.
The Gruf unit or complex has a complex structural and metamorphic history and its tectonic position in the Eastern Pennine Alps is still uncertain. The migmatites bearing folded Gruf gneisses underlies the
Bregaglia intrusives while their contact with the adjacent ophiolitic Chiavenna unit is characterised by a vertical recrystallised mylonite zone (Schmutz, 1976). The Gruf complex consists mainly of migmatitic biotite-feldspathic gneisses and cordierite-bearing pelitic gneisses (Wenk et al., 1974; Bucher-Nurminen and Droop, 1983). Locally these pelitic gneisses also contain lenses of ultramafic and calc-silicate rocks
(Moticska, 1970). Furthermore, the Gruf unit contains rare sapphirine-bearing alominous magnesian granulites, which were discovered by Cornelius (1916). These rocks are of special interest because of the their unusual high Mg-AI-rich bulk composition and their metamorphic history. Many are the workers who have studied the Gruf sapphirine-granulites, this section summarises the results on the metamorphic evolution of the Gruf sapphirine-granulites and of the entire Gruf unit presented by Droop and Bucher-
Nurminen (1983 and 1984).
The authors suggest for the sapphirine-granulites an origin through in-situ partial melting. For these rocks they document a metamorphic pressure, temperature and time trajectory consisting of five different mineral-stages (Droop and Bucher-Nurminen, 1984; Fig. 16). The proposed metamorphic loop starts with a staurolite-stage followed by a pyrope-stage, a prismatic sapphirine-stage, a cordierite stage and finally the metamorphic cycle ends with a simplectitic stage. The authors consider these different mineral stages as equilibria of one single metamorphic cycle and extend this metamorphic P-T-time loop to the entire Gruf unit. Although, Hercynian whole rock ages have been obtained from one Gruf rock sample (Gulson, 1973) the authors assume that high-pressure conditions of 10 kbar and temperatures of 830 °C were attained by the Gruf complex during Alpine metamorphism, but to confirm this assumption detailed geochronological studies remain to be done. The metamorphic trajectory described by the authors considers the Gruf complex to undergo a nearly isothermal decompression achieved through an extremely rapid uplift of the unit. Furthermore, they indicate as the earliest possible decompression-age for the sapphirine-granulites
38 Ma years and estimated for the Gruf complex an uplift rate of 2.2 mm/year in the period between 38 and 30 Ma.
The tectono-metamorphic evolution presented by Droop and Bucher-Nurminen (1983, 1984) for the
Gruf unit and the results obtained by Schmutz (1976) and the metamorphic data presented in this study for the Chiavenna ophiolitic rocks are in agreement. Indeed, the comparison shows that the thermal metamorphic condition of contact metamorphism for the Chiavenna unit could have been occurred through the late Alpine tectonic emplacement of the hot adjacent Gruf unit (Chapter 5). 107 Dumortierite
7.2 Crystal structure
Dumortierite, which has the ideal structural formula SißB [AI6.75 Llo.25 O17.25 (OH)o.7s] (Moore and Araki,
1978), was discovered more than a century ago (Bertrand, 1880). Its typical blue or violet and sometimes
pink colour. Its composition inspired further crystallochemical investigations. The first general formula for dumortierite, space group (orthorhombic Pmcn) and unit cell dimensions were established in 1958 by
Claringbull and Hey. These authors considered dumortierite an anhydrous mineral. The first hydrous dumortierite formula was introduced later by Moore and Araki (1978).
The crystal structure of dumortierite (Figure 7-2) consists of three distinct types of chains built of AIÛ6 octahedra, aligned parallel to the c-axis. Two of the chains contain face-sharing octahedra, the third has edge-sharing octahedra. All chains are arranged in a pseudohexagonal fashion with a central chain linked to six surrounding external chains by isolated SiÛ4 tetrahedra. The external chains are interconnected by
planar BO3 groups.
Experiments on natural and synthetic dumortierite have been performed by Werding and Schreyer
(1983a, 1983b). Their results demonstrated that dumortierite can only be stabilised under hydrous conditions and requires at least trace amounts of water to be present during formation.
Figure 7-2: Crystal structure of dumortierite after Alexander et al. (1986).
7.3 PETROGENETIC system and stability field
Numerous borosilicates have been described in literature. Nevertheless, experimental studies on boron-bearing minerals are scarce compared to other rock-forming systems. Dumortierite may be represented in the borosilicates system ABSH (AI203-B203Si02-H20) (Figure 7-3), that contains a large number of solid phases stable at elevated temperatures and pressures. In this system, dumortierite 108 Dumortierite tâÊSÊÊ
represents one of the most important Al-borosilicates in nature. Beside Al and B, additional elements such as Ti or Mg may also form major components of dumortierite.
The experimental studies carried out by Werding and Schreyer (1990) substantially improved the
knowledge of compositional variation of dumortierite under different pressure and temperature conditions.
Their results led to the substitution of the relatively simple structural formula proposed by Moore and Araki
(1978) with a more complex general formula, [Al7-xDx[l4lAlySi3-yOi5-3x-y(OH)3x+y](B03)] expressing the compositional variations (x > 0.1 and y < 0.5) that occurs as a function of pressure and temperature.
Dumortierite is stable over a wide range of P-T conditions, from low to high metamorphic grade (Figure
7-4) and up to at least 45 kbar (Werding and Schreyer, 1996). Experiments carried out at pressures around 100 kbar and temperatures of approximately 900°C indicate that under these conditions dumortierite reacts to form topaz-OH and an unknown phase (Schreyer and Werding, 1997). At low pressure and high temperature, the mineral breaks down to boron-mullite, while at pressure above 10 kbar
(and T: 800-900°C) the solid breakdown products of dumortierite are boron-free phases such as kyanite and corundum and all boron resides in the fluid phase.
Si02 Figure 7-3: Some Quartz Coesrte crystalline phases of the system ABSH projected from H2O after Werding [+H20] and Schreyer, (1996)).
AUSiOc minerals
ai2o3- I AI3B06 JeAIBOg B203
AI5B09 Jeremejevrte
Titanium-bearing dumortierite (Beukes et al., 1987) was studied experimentally by Werding and
Schreyer (1996) who showed that its maximum Ti-content may reach 0.3 Ti per formula unit. No PT- dependence could be established.
Magnesium-dumortierite is a high-pressure mineral occurring, for instance, in coesite bearing ultra-high pressure rocks of the Western Alps (Chopin et al., 1995). It has been accepted as a new mineral species and represents a phase of the quinary system MgO-AI203-B203-Si02-H20. Its stability field has yet to be constrained by further experimental work. 109 Dumortierite
7-4: of dumortierite. i Figure PT-stability Ptoti • o 1 Preliminary results from Werding and kbar | 40- •[ kyanitet-corundum o o Schreyer (1996) for detailed explanation o sillimanite+corundum refer to Werding and Schreyer (1996). 35- a. andalustte+corundum * o o + kadinite+diaspore
30- solid symbols and crosses3 « growth of dumortierite Kyanite +Corundum 25- [< +Fluid
20- -L Dumortierite 15- 7
"^ : 10- ^ o Quartz ~~ ^ /' + B-MuUrt"' *" +Fluid 5- ++ -* aZD ^
~~ — — Andaluslte — after 1996 __ C Werding&Schreyer _ 1 i 1 i T 1 ^— 200 400 600 800 1000 T°C
7.3.1 Chemical analysis
Selected dumortierite material from the Chiavenna area was analysed for major element composition by electron microprobe (emp) work.
Boron concentrations were measured by inductively coupled plasma mass spectrometry (icp-ms)
Laser-ablation. In this technique, an absolute calibration was performed adjusting the Al-ions with the
AI2O3 content furnished from the emp measurements. The specific standard was NIST 610. Details on ICP-
MS laser-ablation technique are given in the Appendix.
Electron microprobe analysis (Table 7-1) yields total oxide concentrations (excluding B) of approximately 92 in weight % including the water content which was calculated. The dumortierite analysed is characterised by high AI2O3 and SiÛ2 contents, while the concentrations of additional constituents like
Ti, Fe, Mg, and P do not exceed 1 wt.%. These data are in agreement with data reported in the literature for natural dumortierite samples (Willner and Schreyer, 1991; Beukes, et al., 1987; Chopin, et al., 1995), although the Ti content of the Chiavenna dumortierite is invariably lower than the values observed in samples from other localities. This difference may be related to whole rock compositional differences, in this case, to a limited Ti availability in the precursor rock material.
The ablated material of six different 10 u,m craters was analysed by icp-ms in order to determine the boron concentration of the mineral. The measurements were carried out on the same dumortierite blasts, which were selected for emp analysis. The average boron concentration obtained is 22668 ppm (part per million), the recalculated B2O3 concentration (Table 7-2) is 7.30 wt%. This value indicates that the dumortierite is particularly enriched in boron. Indeed, the B-atom number per formula unit (a.p.f.u.) is 1.27, that is approximately 0.27 higher than the values reported for dumortierite from other localities, which are normally fixed at 1 B (p.f.u.). 110 Dumortierite
Table 7-1: Electron microprobe and ICP-MS laser-ablation analyses of dumortierite from Chiavenna: oxide wt% and boron in ppm for the ICP-MS laser ablation, the last column gives the recalculated oxides average composition of dumortierite
Electron microprobe data (wt.%)
Oxide I II III IV Averaqe
S1O2 29,73 29,60 30,02 30,11 29,87
T1O2 0,55 0,25 0,35 0,19 0,34
AI2O3 58,30 59,15 58,72 58,18 58.59
FeO 0,35 0,39 0,38 0,39 0,38
MnO 0,02 0,02 <0 01 <0.01 0,01 MqO 0,83 0,79 0,79 0,93 0,84
CaO 0,02 0,01 0,01 0,01 0,02
Na20 0,02 0,02 0,01 0,03 0.02
K2O 0,01 <0.01 <0.01 0,02 0.01
P2O3 0,29 0.17 0,07 0,13 0,16 H20 1,37 1,38 1,37 1.37 1.37
Sum 91.49 91.78 91.73 91.38 91,60
ICP-MS USER ABLATION DATA (PPM)
Element V Vi VII VIII IX X Averaqe
B 24228 25036 23759 22648 17182 23153 22668
7.3.1.1 Discussion
The chemical analyses performed on natural dumortierite needles in this study provide an average
boron-concentration of 1.27 atoms per formula unit. This value is substantially higher than the value
normally observed for other natural and synthetic dumortierite minerals (number of B-atoms = 1 p.f.u.).
Recent studies (Bloodaxe et al., 1999; Dyar, et al.,1999; Ertl et al., 1997; Schreyer et al., 2000) on natural
and synthetic tourmalines showed that this mineral also may contain more boron than the 3 atoms p.f.u.,
which are considered to be a limiting value in the tourmaline formula due to structural constraints. In
addition, sophisticated spectroscopy measurements carried out during these investigations, indicated that
boron might be accommodated not only on the trigonal BO3 structural site, but also on the tetrahedral
position. Hence, the occurrence of excess boron detected in the Chiavenna dumortierite may be explained
by the accommodation of the exceeding 0.27 B p.f.u. on the tetrahedral site. If this is the case, the general
structural formula of dumortirite given by Werding and Schreyer (1996) [Al7-xnx[l4lAlySi3.yOl5.3x.y(OH)3x+y](B03)],
where (x>0.1 and y<0.5) should modified to include excess boron on the tetrahedral site, resulting in a
new formula of the type: [Al7-xDx[[4lAlyBzSi3.(y+2)Ol5-(3x+y+z)(OH)3x+y+z](B03)],
Thus, the structural formula of the analysed dumortierite composition would be:
[AI6.4Do.6[[4lAlo2B02/SJ2530l6(0H)l 73](B03)]. 111 Dumortierite V'-MMËÊËÊk
The introduction of B on the tetrahedral site results in an excess negative charge in the structural formula.
Similarly, the presence of excess boron on the tetrahedral site within tourmaline causes proton deficiencies, Schreyer et al. (2000) proposed to rebalance the deficiencies using a BHSLi substitution.
7ab/e 7-2: Comparison of the recalculated average oxide composition of dumortierite needles from Chiavenna and natural dumortierite analyses from the literature. Last column gives number of ions p.f.u. based on 18 oxygens.
Dumortierite compositions # atoms p.f.u.
I II III IV V* I III IV V
Si02 29,87 39.20 31.9 31.13 31.7 Si 2.8 3.05 2.95 3
Ti02 0,34 1.04 2.76 3.05 4.28 Tii 0.02 0.2 0.22 0.3
AI2O3 58,59 47.40 58.51 57.14 47.3 Al« 0.2
B2O3 7.3 1.73* 5.27° 5.76° 5.1 Fe 0.03 0.03 0.03 0.01
FeO 0,38 0.16 0.49 0.44 0.09 AI« 6.4 6.59 6.39 5.28*
MnO 0,01 <0.01 n.a. n.a. n.a. D 0.6 0.18 0.41
MqO 0,84 0.03 n.a. n.a. 7.8 B 1.27 0.87 0.94 0.83
CaO 0,02 0.03 n.a. n.a. n.a. 0 16 17.46 16.78
Na20 0,02 1.14 n.a. n.a. n.a. OH 2 0.54 1.22 2.21"
K2O 0,01 1.4 n.a. n.a. n.a. Mq 0.12 1.1
P2O3 0,16 0.16 n.a n.a. 0.07 totalAl
H20 1,37° 6* 0.83 1.16 3.5 ** expressed as H Sum 98.91 99.70 99.76 99.45 99.9
I Chiaven na (this study)
II Bushma nland, South Africa (Willner&Sc hreyer, 1991) lll+IV, respectively JGH21 7-1 and DUM1
V Mg-dunîortierite, analys sVII+Vllafrom Western Alps ( Chopin et al., 1995)
In order to redefine the compositional range of dumortierite and its correct structural formula, further detailed crystal chemistry investigations must be performed.
113 Conclusions
8 Conclusions
This chapter summarises the Alpine metamorphic evolution of the mafic, ultramafic and carbonate bearing
Chiavenna unit and its chemical characteristics. It focuses on the major evidence confirming the aims and hypotheses exposed in section 1.2. Finally, it considers the role and the significance of the Chiavenna unit in relationship with the geotectonic evolution of the surrounding nappes.
8.1 Alpine metamorphic evolution
8.1.1 Recorded metamorphic events
Field and petrological analyses reveal the rare presence of pre-Alpine features in the Chiavenna unit.
The only preserved pre-Alpine "trace" consists of a mantle layering within the ultramafic rocks where centrimetre-scale dunitic and pyroxenitic layers are crosscut by a weakly developed approximately N-S oriented magnetite-schistosity, discordant to the dominant E-W schistosity (S2). The mantle layering is pre-Alpine in age while both schistosities are related to the Alpine collision between Adria and Europe.
The study of Huber and Marquer (1998) on the structural evolution of the Chiavenna unit interpreted the
N-S schistosity (S1) as the structural relic related to a first deformation phase (D1) occurred during the nappe-staking, while the E-W schistosity (S2) took place successively during a syn-collisional extension phase (D2).
The present study evidences the occurrence of two distinct Tertiary metamorphic events characterised by distinct deformation styles, temperature conditions and space-distribution. The metamorphic events consist of a synkinematic greenschist facies metamorphism and a superimposed near static late Alpine thermal event. The restricted isograds distribution subparallel to the contact between the Chiavenna and the Gruf units argue for a local increase of temperature similarly to contact metamorphism driven by the emplacement of a adjacent hot body.
The main features characterising the regional dynamic event according from the distinct lithologies, are:
A well developed antigorite and/or chlorite schistosity (S2) where large olivine blasts together with
magnetite bands are elongated parallel to the schistosity.
The occurrence of rotated tremolite and olivine aggregates replacing former diopside, rotated within
the schistosity (S2).
The presence of a relic magnetite schistosity (S1) within the tremolite and olivine aggregates.
A developed actinolite-low Al-hornblende schistosity (S2) within the epidote and low Ca-plagioclase
bearing amphibolite rocks. This schistosity is also recognisable in amphibolites, which recrystallised at
higher temperature conditions.
Schistose metamorphic low-grade assemblages in calcite-marbles. 114 Conclusions
Outcrops located in the northern part of the Chiavenna unit best preserve evidence of this metamorphic event, which features become less visible towards the south.
The successive thermal metamorphic event variably overprinted the rock-assemblages and textures formed by the dynamic event with new near static textures and mineral-assemblages typical for progressively increasing temperature metamorphism. The major features related to this near static thermal metamorphism are:
The occurrence of metamorphic reactions of increasing grade along a prograde path where new
mineral phases are formed. The inferred temperature increase is of approximately 200°C. This
increase occurs between less than 500°C and approximately 700°C across an average distance of
2.25 kilometres.
The growth of new mineral phases, inducing the replacement of schistose rock-textures with coarse¬
grained granoblastic textures. The occurrence of coarse-grained granoblastic textures is particularly
evident in the ultramafic rocks where dehydration reactions such as the antigorite-out, talc-out and
Mg-amphibole-out reactions liberated significant amounts of water enhancing the recrystallisation of
the rocks.
The occurrence of exchange reactions involving solid solution phases such as hornblende and
plagioclase. During increasing temperature, the amphiboles of the mafic system modify their
compositions towards progressively higher edenite, AIIV, AIVI and Ti contents, while plagioclase
generally increases its anorthite component until clinopyroxene appears. The compositional variations
of these phases occur without inducing significant changes in the rock-microfabric.
These features are clearly related to a steep temperature gradient of approximately 90°C/km oriented towards the southern Gruf unit contact.
Independent pressure estimates obtained for calcite marbles of the Chiavenna unit and tonalitic dikes crosscutting the unit yield pressure values of approximately 4 kbar. This refers to the conditions achieved by the rocks during the high temperature metamorphism and represents the pressure conditions acting during the near static thermal event described above.
In this study it is proposed to interpret the dynamic event as a synkinematic greenschist facies metamorphism associated with the Tertiary E-W syn-collisional extension phase which developed similar structures in the neighbouring Tambo nappe (Baudin and Marquer, 1993; Marquer et al., 1994). For the thermal event, the narrow isograd zones, restricted to the southern part of the Chiavenna unit and the temperature increase of approximately 200°C across a distance of less than 2.5 kilometres argue for a local thermal overprinting such as a contact metamorphism induced by the emplacement of a neighbouring hot body. The isograds distribution in the field, subparallel to the present tectonic contact 115 Conclusions
between the Chiavenna and the Gruf unit, suggest the migmatitic Gruf unit as the possible heat source accountable for the high temperature gradient observed in the Chiavenna unit.
8.1.2 Time constraints of the alpine evolution
39Ar/40Ar amphibole dating of several amphibolites from different temperature zones and from different distances from the Gruf unit confirms the occurrence of multiple metamorphic events. The comparison between argon data of extensively recrystallised amphibole samples at high temperature with the results of relic-rich amphibole samples recrystallised at lower temperature showed that the intensity of recrystallisation controls the Ar-Ar ages providing less discordant age-spectra for samples re-equilibrated under high temperature conditions. The age spectra obtained for amphiboles recrystallised at lower temperature conditions display scattered values related to the presence of inherited older amphiboles chemically distinguishable from the younger amphibole compositions. The most coherent age-spectra support the occurrence of a metamorphic event changing the compositional features of the amphiboles between 30 and 33 Ma. Noteworthy, none of the amphiboles provide evidence for a metamorphic event at
25 Ma.
The age-spectra of amphibole samples recrystallised at lower temperatures contain relic amphibole compositions correlating with ages of 45-47 Ma and compositionally distinct amphiboles yielding ages between 30 and 33 Ma. The occurrence of chemically distinct amphiboles is confirmed by electron microprobe analyses performed on several amphibole separates, which support the coexistence of two amphibole generations: a first generation marked by high Ca/K ratios and low edenite components
(39Ar/40Ar ages of 45-47 Ma) and a second generation characterised by low Ca/K ratios and high edenite values (39Ar/40Ar ages of 30-33 Ma).
8.1.3 Interpretation and thermal model
Petrologic and isotopic investigations on amphiboles from the Chiavenna amphibolites demonstrate that the amphibole compositions are controlled by exchange reactions occurring during increasing temperature conditions induced by a near static thermal event occurred between 30 and 33 Ma. This study proposes to consider:
The approximately 45-47 Ma old low edenite and high Ca/K ratio amphibole compositions linked to a
regional medium temperature metamorphism compared to the high temperature amphiboles. It seems
possible that this age represents the cooling age of amphiboles recrystallised during the synkinematic
greenschist facies metamorphism, which accompanied the E-W extensional deformation phase.
The 30-33 Ma old recrystallisation of high edenite and low Ca/K ratio amphiboles related to the near
static thermal event. On a regional scale, the thermal changes affecting the Chiavenna unit can be
age-wise associated to the thermal anomalies, which brought to the emplacement of the Masino- 116 Conclusions
Bregaglia intrusions rather than to the thermal conditions responsible of the younger Novate granite
intrusion.
The age-constraints and the petrological features achieved in this study are in agreement with the hypothesis of regarding the emplacement of the migmatite-bearing Gruf unit as the heat source accountable for the isograd zones and the metamorphic transformations affecting the southern part of the
Chiavenna unit.
A simple mathematical modelling in which the emplacement of a hot body with similar physical and morphological features as the Gruf unit within an area of ultramafic composition was calculated. The main goal of these calculations were to verify whether the Gruf unit may have provided sufficient enthalpy to enhance the thermal overprinting observed in the Chiavenna unit rocks. The parameters used for the calculations were chosen to simulate the physical and morphological characteristics of both Gruf and
Chiavenna units, as observed in the field and as obtained from experimental data on similar rocks. The results indicate that the temperature boundary conditions coincide with the temperature differences between the regional and the thermo-metamorphic overprinting observed in the Chiavenna metaperidotites. The computed data are comparable to the real petrologic conditions observed in the field.
The location of the calculated isograds is in agreement with the position of the mapped isograds with the only exception of the enstatite-bearing assemblages, which were not reproduced by the computing of the thermal model. The results indicate that the heat furnished by a similar migmatitic body may be sufficient to enhance prograde metamorphic reactions typical for contact metamorphism, as the reactions observed in the ultramafic system considered in this study.
8.1.4 General considerations on the tertiary metamorphic evolution of the Chiavenna
unit
The Tertiary near static thermal event described for the Chiavenna unit presents metamorphic features which are typical for contact aureole but also for areas affected by low-pressure regional metamorphism
(regional high temperature-gradient metamorphism). Although the metamorphic features are identical, the two types of metamorphism are related to distinct roots. Low-pressure regional metamorphism is related to regional thermal anomalies, while contact aureoles classically surround magmatic intrusions. For the high temperature gradient observed in the Chiavenna unit, in this study it is proposed to consider the scenario of local contact metamorphism.
The Tertiary thermal conditions of the Central Alps were influenced by several thermal events: 1) the rising of the Lepontine dome in the more western Lepontine Alps (35-23 Ma), 2) the Masino-Bregaglia tonalite-granodiorite magmatic intrusions (29-32 Ma) and 3) the southern Novate granitic intrusion (25 Ma).
The age constraint for metamorphism in the Chiavenna unit suggests that the Novate granitic intrusion 117 Conclusions
was not responsible for the thermal overprinting of the Chiavenna unit. Thermobarometric studies (Engi et al., 1995) on the Tertiary metamorphic conditions in the eastern Lepontine Alps and their surroundings documented for areas including the Chiavenna unit temperatures between 550 and 600°C which are lower then the temperature conditions proposed in this study for the southern part of the unit. Furthermore, in the areas exposed towards the west of the MB intrusions, the regional metamorphic peak was coincident with the Novate granite genesis (Villa and von Blanckenburg, 1991) and therefore it is younger then the thermal event that overprinted the Chiavenna unit. Contact aureoles were described around the Masino-
Bregaglia intrusion-area (Trommsdorff and Connolly, 1996; Wenk et al., 1974). However, the distances separating the Chiavenna unit from the MB intrusions are too large to develop contact metamorphism. On the other hand, the emplacement of the Gruf unit appears to be the only possible mechanism to explain the observed features in the Chiavenna unit. Futures detailed studies on the metamorphic isograds distribution in the Gruf complex and on its age-evolution are necessary to confirm the hypothesis proposed
in this study.
The analysis of the metamorphic evolution presented in this study for the Chiavenna unit introduces
new elements, which support or contrast the results proposed in former studies. While Schmutz (1976) linked the observed increasing metamorphic grade to a progressive Alpine regional metamorphism, Huber and Marquer (1998) and Huber (1999) essentially described a retrograde metamorphic evolution for the same rocks. The results presented in this study support part of Schmutz (1976) constraints, but clearly refuses the interpretation of Huber and Marquer (1998) and Huber (1999). The prograde metamorphic evolution progressively affecting the rocks is well documented by a great number of mineralogical and petrological features. Distinct from Schmutz (1976), the results of this study evidence the occurrence of a late Alpine local metamorphic event which overprinted the Chiavenna rocks with progressively increasing temperature conditions.
8.2 BULK-ROCK CHEMICAL CHARACTERISATION
The geochemical characterisation of the metamorphic ultramafic, mafic and carbonate rock-sequence leads to the following significant conclusions.
The metaperidotites of the Chiavenna unit can be considered former Iherzolites with distinct features documenting a composite mantle history. Early partial melting processes probably commenced in the garnet stability field and depleted the Iherzolites in light rare earth element. Part of these depleted
Iherzolite residua (type B enriched metalherzolites) underwent then cryptic mantle metasomatism, which selectively enriched the mantle rocks with light REE. Subsequently, a widespread serpentinisation, documented by the presence of low Ca metaperidotites with negative europium anomalies, occurred on 118 Conclusions
the seafloor. On the basis of this complex mantle evolution and the documented fertile character of all
metaperidotites, a subcontinental derivation for these metaperidotites is proposed.
The Chiavenna amphibolites are subdivided in banded and massive amphibolites. Both group of rocks display identical chemical features. They are tholeiitic in composition and show a strong affinity to N-type
mid ocean ridge basalts (MORB) environment. The banded amphibolites are probably derived from
crystallised liquids while the massive amphibolites should represent shallow depth dolerite bodies. None of the amphibolites show textural or compositional evidences of gabbros. Boudins of metamorphosed
rodingite rocks are embedded within the ultramafics. Despite a relative modest Ca-enrichment acquired during rodingitisation, these rocks are chemically indistinguishable from the amphibolites. The rodingites
most likely represent tholeiitic dikes, which intruded the peridotites before or during serpentinisation.
The calcite-marbles intimately associated with the amphibolites and embedding meter-scale amphibolite breccias are interpreted to represent former carbonates of probably sedimentary origin. This
lithological sequence, overturned during the Alpine tectonic (Schmutz, 1976; Huber and Marquer, 1998) and lacking in several elements of classical "complete ophiolitic sequences" is interpreted to represent the
remnants of an oceanic basin. It has been showed (Lemoine et al., 1987; Piatt et al., 1989; Froitzheim et al., 1996) that the Tethys developed small oceanic basins floored by subcontinental mantle rocks instead of by "typical" oceanic lithosphère and often lacking in several layers of the "ideal" ophiolitic sequences.
Similarly, also present passive continental margins (Boillot et al., 1995) are floored by subcontinental
mantle rocks. Evidence for the emplacement of the Chiavenna metaperidotites on the ocean floor are:
A widespread hydrothermal alteration (serpentinisation) modifying the composition of the
metaperidotites.
The occurrence of ophicarbonatic material filling fractures and pockets within metaperidotites.
The presence of boudins of metamorphosed rodingite dikes.
The occurrence of Cr-rich nodules embedded within the calcite-marbles indicating possible erosion
processes which involved ultramafic and carbonates material, probably on the seafloor.
These aspects lead to consider the Chiavenna unit as an "incomplete" ophiolitic sequence whereby subcontinental mantle rocks instead of "typical" oceanic lithosphère were directly exposed on the ocean floor and were covered with tholeiitic N-type MOR basalts and carbonates. The missing of chronological constraints related to the magmatic history of the basalts hampers precise paleogeoraphic affiliation of this
"incomplete" ophiolitic sequence to the Jurassic Ligure-Piemontese oceanic basin or to the Cretaceous
Valais basin forming the Tethys in this area. The geochemical features of the Chiavenna unit are similar to
both types of basin. However, the geotectonic position of the Chiavenna unit between the Adula-Gruf
nappe and the Tambo nappe, situates it paleogeographically between the southern European margin and the Briançonnais terrain. Possibly, it may represent the southeastern remnants of the Cretaceous Valais basin. 119 Appendix
A Appendix
A.1 Analytical Methods
A.1.1 X-Ray Fluorescence (XRF)
Chemical bulk-rock composition was determined by X-Ray fluorescence (XRF) analysis, with a sequential spectrometer (Philips PW 1404) at the Swiss federal laboratories for material testing and research EMPA in Dübendorf.
The analyses were calibrated using natural USGS rock-powders. The rock samples were crushed and successively ground in an agate mill (ultramafic samples) or in a tungsten carbide mill (mafic and carbonate samples). For the major elements measurement lithium tetraborate glass beads were fused from ignited (1050°C) rock powders while trace elements were determined using 10 g of rock powder. The intensities were corrected for instrumental drift, background and matrix effects. The detection limits for the trace elements are listed directly in the samples-tables.
The Ti-abundance used in the primitive mantle normalised trace elements plots (Figure 6-9) has been performed by XRF analyses and is not precise at low concentration levels. However, the computed Ti- contents of the metaperidotites are consistent with the abundance of adjacent REE and are consequently considered reasonable.
A.1.2 Inductively Coupled Plasma Mass Spectrometry (ICP-MS)
Similar to major and trace element determination the rare earth element measurements were performed at the EMPA in Dübendorf.
80-200 mg of sample-powders were dissolved in a mixture of HNO3 and HF acids in different concentrations using Teflon beakers of 160 ml volume. This mixture of powder and acids was placed in a microwave oven (MLS1200, ML GmbH) where the elevated pressure and temperature allow the acid to attack and dissolve the rock-powder. Contemporarily with the same procedure blank and standard rock- powder samples were prepared. Afterwards all the obtained dissolved mixtures were diluted with deionised water at 50 ml (ultramafic rock) or at 100 ml (basic rock) and measured with an ELAN 6000
(Perkin Elmer Sciex).
The data correction related to drift-effects was made using as internal standards the 111Cd correction factor for Ga, Ge, Rb, Sr, Y, Zr, Nb, Ba and La to Gd and the 185Re correction factor for Hf, Th, U and Tb to Lu. To avoid existing incompatibilities between the computed data and published standard data do to artifact introduced with the correction procedure was made a linear interpolation between the two standards. The formula used for the linear interpolation is: 120 Appendix
c(x) =Ccd(x)*m/b+CRe(x)*(b-m)/b where: c=concentration of element x, m=mass of element x and b=74=(mass Re-mass Cd). Detailed information to the presented procedure may be found in Müntener (1997).
The standard powder samples used for the measurements in this study are MRG-1 (gabbro), NIM-P
(peridotite) and PCC-1 (pyroxenite). Tantalum values performed with the icp-ms method have been excluded from the REE discussion of ultramafic and mafic rocks because of analytical errors, while the niobium, rubidium, barium and thorium have been included in the spider plots even if their values may be affected by possible contamination problems linked to the tungsten carbide mill used for the sample preparation. In addition, also Zr and Hf may present erroneous low values related to the presence of not dissolved zircon crystals in the acid mixture.
A.1.3 ICP-MS Laser Ablation
Rare earth element analyses on single mineral phases were performed using a laser-ablation microsampler coupled to an inductively plasma mass spectrometer at the "Institut für Isotopengeolgie und
Mineralische Rohstoffe" of the Swiss Federal Institute in Zürich (ETH). The ICP-MS instrument is an Elan
6000 (Perking-Elmer, Norwalk, CT, USA), while the excimer laser is a Complex 110i, Argon Fluoride 193 nm, Lambda Physik, Göttingen, Germany model, working with a gas mixture containing 5 % F2 in Ar with small amounts of He an Ne. The in situ microanalyses performed with the LA-ICP-MS technique have a detection limit of 1-10 ng/g in a 40 urn ablation pit. The quantitative determination of the trace elements in the mineral grains with the LA depends on the determination of at least one major element of known concentration, which is used as internal standard element to correct for fluctuations in the rate of the sample ablation. In the present study the concentration of S1O2 was used as internal standard, while as external standards were used NIST612 and BCR-2. Detail explanation on the LA-ICP-MS technique can be found in Günther et al. (1997).
A.1.4 Mineral separation and Ar/Ar Dating
Crushed rock-material was sieved in different fractions depending on the amphibole size and extends of mineral intergrowth. A fraction of 125-177 |jm of sample was chosen, washed, ultrasonically cleaned and successively the minerals were separated with standard high-density liquids and magnetic separation
(Frantz). Finally it was necessary to clean the separate with handpicking under a binocular microscope.
An amount of approximately 40 mg amphibole for each separates was irradiated in the TRIGA reactor in Pavia and subsequently analysed with the Ar-Ar stepwise heating method in the laboratory of Dr. Igor
Villa at the "Institut für Isotopengeologie Universität Bern" in Bern. Detail on the Ar/Ar technique can be found in Villa et al. (2000) and Villa (1992). 121 Appendix
„ iiü*«M^SlBlilliiBa^ffiffl^fflBBHHBBlBBBB
A.1.5 Electron Microprobe (EMP)
The mineral compositions were analysed using a Cameca SX50 electron microprobe provided with five crystal spectrometers at the "Institut für Mineralogie und Pétrographie" of the Swiss Federal Institute in
Zürich (ETH). The samples were coated with 200 À of carbon.
The working-parameters of the machine were set at 15 kV accelerating voltage, 20 nA beam current and using a beam size of approximately 1u.m. The counting times were 10-40 seconds and a ZAF-type correction was applied to the data. Natural and synthetic mineral standards were used for calibration. The various measured mineral concentrations were recalculated using a mineral-norm calculation program norm version 4.0 written by Ulmer (1993).
The mineral abbreviation used in the text and in the appendix is taken from Kretz (1983). 122 Appendix
A.2 Sample Location
Table A-1: Sample list with their location in the field. The coordinates are calculated on the basis of the Swiss coordinates system. Abbreviation are: S=Schiesone, B=Bregaglia, G=Gruf, U=ultramafic rocks, M=mafic rocks, C=metacarbonates, P=pelitic rocks, Ug=ultramafic rock in the Gruf Unit.
Sample list with field locations
Ultramafic rocks Mafic rocks
Sample Location Sample Location Sample Location Sample Location
SU2 750'880/129'800 SM18 751720/129'680 BM11 755'840/131'880 BM8 753780/131'500
SU3 750'880/129780 SM23 751'850/129'640 BM12 755'840/131'880 BM9 753780/131'500
SU8 750'880/129760 SM24 751'850/129'640 BM13 756'030/131770 BM10 753780/131'500
SU51 750'880/129780 SM25 751'850/129'620 BM14 756'030/131770 SM30 LOOSE BLOCK
SU4 751'100/129'880 SM26 751'850/129'620 BM15 756730/131'560 SM32 LOOSE BLOCK
SU6 751'180/130,080 SM28 751'840/129'590 BM16 756730/13V560 BM30 752'540/132'060
SU7 751'180/130'080 SM29 751'900/129'500 BM19 756'310/131'390 BM33 752'620/132700
SU2 750'880/129'800 SM62 752700/129'460 BM20 755'450/132'120 BM34 752'620/132700
SU5 751'360/130'020 SM34 752770/129'450 BM21 753750/131760 SM58 752'660/129750
SU15 751'680/129780 SM63 752'040/129'410 BM22 753750/131760 SM71 753'020/129'130
SU16 751'680/129780 SS33 752'080/129'390 BM23 753'070/131'150 SM80 752'300/129'840
SU17 751700/129700 SS36 752'080/129'390 BM24 753'070/131'150 BM37 752'620/132700
SU19 751720/129'680 SM64 752'100/129'410 BM25 753'070/131'150 BM38 752'620/132700
SU20 751780/129'650 SM59 752'230/129'390 BM35 752'620/132700 BM39 752'620/132700
SU21 751'810/129'650 SM60 752'230/129'390 BM36 752'620/132700 BM40 752'620/132700
SU22 751'810/129'650 SM61 752730/129'390 Carbonate Rocks
SU41 Not in situ SM43 752'470/129780 SC1 Not in situ SC45 752'520/129770
SU27 751'800/129'620 SM44 752'470/129780 SC12. 751'630/129'820 SC46 752'520/129770
SU75 751'910/130'270 SM37 752'600/129'310 SO100 751740/129700 SC52 752'910/129'150
SU76 751'919/130'270 SM54 752720/129700 SO101 751700/129700 SC53 752'910/129'150
SU77 752'080/130'040 SM55 752720/129700 SO102 751700/129700 SC72 753'030/129'130
SU78 752'080/130'040 SM57 752'690/129740 SO104 751700/129700 SC73 753'020/129'130
Mg13 752'100/129'400 SM58 752'660/129750 SB105 Not in situ SC74 753'020/129'130
Mg13a 752'100/129'400 SM71 753'020/129'130 SC13 751'630/129'820 SC96 750'320/128'830
Mg13b 752' 100/129'400 SM80 752'300/129'840 SC14 751/630/129'820 SC97 750'320/128'830
Mg13c 752' 100/129'400 SM81 752'300/129'840 SC39 752'600/129790 BC18 756'310/131'390
BU2 755740/132'630 SM82 752'300/129'840 SC40 752'680/129740 SG89 LOOSE BLOCK
BU3 755780/132'570 SM83 752'420/129'910 SG56 752'900/129710 SG86 751720/128780
BU6 752'530/131'970 SM84 752'520/130'000 SG65 LOOSE BLOCK SG91 750'320/128'830
BU7 753780/131'500 SM90 752'060/129'540 SUG66 LOOSE BLOCK SG92 750'320/128'830
BU27 752720/13T640 SM94 750'320/128'830 SUG67 LOOSE BLOCK SG93 750'320/128'830
BU28 752'210/131'580 SM95 750'320/128'830 SG68 LOOSE BLOCK SB105 LOOSE BLOCK
BU25 752'680/131'250 SM98 750'320/128'830 SG69 LOOSE BLOCK Dumortierite
BU32 751 '650/132'040 SM99 750'320/128'830 SG70 LOOSE BLOCK BP26 LOOSE BLOCK
BU41 752'620/132'200 BM1 755700/132'600 SG85 751720/128780 BG17 752'680/131'250
BM4 753'850/133'140 SG87 LOOSE BLOCK BP5 756780/130'560
SG88 LOOSE BLOCK BP31 753'940/133'130 .is O tJc CD F F g ^o ^ CD co CM co CM CM o o T— CD m 00 CD LO LO en CM eo Ü m 5 o o C0 •G o LO O) oo m o co co CM o o o en co CM O)" T— •* CM CM co CD O o o CM m" V co o o o CO r— r^- in CD o" eo" cm" oo" V V V V V LO c- CO o" o" co o o" o" o o" en CO CD U. 03 F tJ CO CD CO CO in r- f- co CM 00 LO co co co LO l~- co co o o o o 00 o O m co O LO o co co O 3 co 00 CM o o o m - CM CM co 00 co O) LO CD m Z> co" oo" en co co Ö V V V O O V o o" cm" o" cm" o" o" o" o" o" V V V V V V co CM V UJ m co o CO o LO O o 0. CD CD 05 00 CO CO co CM CO o CM co LO LO LO 05 LO ca. UJ 5 co CO co o o C0 o O o co 00 CM o o o m -r- l-~ CM CM 1*- m o T en h- cm" co" co" 00 o" V V V 00 LO V O o" o" o" o" o" o" en" o" o V V V V V V V V m CT> CD r- -J CO co o co ;» co o CO CO Ö 111 Cll co LT) c 1-1 X LO CM en o o en co in LO LO en CD -1 CO o o co O O o CD CM CD CM O o m o CD o f- co o" 1^. CM CM r- CT) o in >< < cm" co" co V V V r» 00 00 V CO 1- o" C\T co" o" cm" o" o" o" o" o" o V V V V V V V V CM V CD o Ul m 1- co o o Q. .Ç E CD o »^ LU Cl i>- co Cll X O) 00 co co LO LO co co CM m o o CM o o 00 LO o O o 00 CM co o C u o 00 00 eo_ o o co co m r- CM CM CD co m * CM in IC Z3 co" co" o" V V V l>- co LO a. o" cm" co" o" o" o" o" o" o" V V V V CO V cd co o CM CO CO CO z m o CM i Oc C 01 LU -JS o. r-- CD 5° co en co co LO o o 00 CM LO LO 9> o T— o CO o o o oo O LO LO CO f- o co CM o CM CM o> o o o LO CM CD f- o co co CO co" r-" \7 V CM en I-- en CM LO c ca o" co" co" o" co" o" o" o *" o" o" V V V V co co CD Ti¬ co o CM CO < o O c: 111 c F a. o .o Q > o ti. tn en CM co CM CM o o oo 00 CO co o o o 00 CM LO o o CM CM LO CD co CO CO m o co m CM CM m
Table A-3: Major and trace element concentrations in representative peridotite samples from the Chiavenna Unit. Major element compositions are presented in oxides and are calculated in wt%, minor element concentrations are expressed in ppm.
Refractory metalherzolites (Type C)
Sample SU16 SU8 Mg13 SU75 SU76
Si02 42,15 40,06 41,79 42,06 41,21
Ti02 0,07 0,09 0,08 0,06 0,09
Al203 3,05 1,22 2,87 3,72 3,47
Fe203' 8,69 11,73 9,04 7,44 9,11
MnO 0,13 0,19 0,12 0,11 0,11
MgO 40,17 43,44 39,48 41,51 41,17
CaO 0,99 0,38 0,02 0,05 0,09
Na20 0 0 0 0 0
K20 0 0 0 0 0
P2O5 0 0,01 0,01 0,01 0,01
L.O.I." 3,54 2,29 6 4,63 4,26
Cr203 0,36 0,21 0,38 0,37 0,38
NiO 0,34 0,35 0,32 0,36 0,29
Total 99,49 99,97 100,11 100,32 100,19
Fm <10 0 <10 <10 <10
Ba"1 <10 10 <10 <10 <10
Rb"' <8 0 <8 <8 <8
Sr 8 2 <15 <15 <15
Pb"1 <5 <5 <5 <5 <5
Th1" <5 <5 <5 <5 <5
u1" <10 <10 <10 <10 <10
Nb 6 1 6 6 5
La"' <20 <20 <20 <20 <20
Ce"' <15 <15 <15 <15 <15
Nd"' <25 <25 <25 <25 <25
Y 4 1 6 5 5
Zr"' 7 2 <10 <10 <10
V 77 62 64 51 89
Cr 3376 1927 2807 2938 2959
Ni 1965 2340 1962 2079 1755
Co 95 146 68 73 65
Cu 26 18 16 33 69
Zn 59 73 44 46 46
Ga'" 4 1 <2 <2 <2
Sc 16 8 14 19 18
s"1 616 0 <50 <50 117
' Fe203: total iron
" L.O.I.: loss on ignition
'" Trace elements below detection limit o c CD «* eo CM CO en 00 01 1»- co CM O) eo en r- en in eo m o co 1^- o in o m m O i~- CO CD § m oo" CM CO o o m o o CM m CM * * c: S *" o" CT>" en V CM V co l'¬ CM CD co ° S o" |-~" cm" o" o" o" V V V V CM c- H», co en o CD c- ï» ai .co F co en -c Cll eo CD en CO q- eo CO eo LO •* .O m LU CI) ' co'-fc -1 — n a. O S d O O O O in o h d O) O O o CD O CD c co (S q d o =cs J2 ~C0 =CD o 3 c (S U = H LO< < U. S S U z sZ CL j o z H "u. m te to CL 1- "b z _l U Z > N > u Z o O N o co co L? -^ ° c CD S 00 Q) r- 1— eo 00 co eo co CD CM CM sf CM en O o o o in in co CD eo 00 £ -S? O eo o o o 00 o O t CO in m sf CM CM o eo CD m st co 1^- cm" 00_ eo V sf v v V eo f- CD 00 CD CD co CM s 5 r-" en o" en o" o" o" o" en" V V V V CO co sf en CD -S O •* co co co in si¬ S E co CM eo co in eo sf CM CD in eo CM 00 co CM eo co_ eo en o 00 in o sf o m CM CD sf O in oo •sf o O m CM a> T— m m LO 2 oo" m" o" r-" o" oo" O V m v v V CM eo o s* sf 00 sf E x E -u CD co CO co co 00 CO CD co co sf eo CO eo in eo in eo eo oo_ o CM en 00 CM in o sf o m co o CM o ° 2 CO_ CM cm" 01 O CM o o CM CM ö o 00 oo sl- co CD m o" n-" en" CM V •si¬ v V si¬ en in CD CO o" «t o" o" cm" o" o" t CM v V V CM CM v •sf en « c£ CM o -S c "O «el- 00 en in 00 CD C CO eo I-. eo sf eo co in o 00 co eo CD io f- sf r-- o in o •sf o m m CT) o CO CO UJ sf o sf o sf o CD - CO CD lo CM CM en i- o co CO CD oo" m" en p oo" V v v V co o o OO si- 1— sf P 1- 2 en" o" oo" co" o" o" o" o" CD V V v v CM eo CD v _l CO sf r- en m Q. cd o s>-s CD *- s Z CM 00 CD CD O O 0. s* m CD in co •sf CT) sf t O in in in en 00 81 00 h- m r-- oo in in o •sf o en Ol oo r- o E 00 00 CM sf en CM CM LO_ o co 01 CM CM CD S s* CD o O m co C 2 r-" cm" m" o" co p oo" co V v v V si¬ CO si- co CM o" h-" CT) co" o" o" o" o" sf CM V V v v CM CM en sf v C — < m sf en "o 'S;2 cd CO -*S i= c CM C CD en co in co in •sf 00 m CM r- CM LO in en CD r» CM O CM CM C0 O LO m o sf o en O o s* CM s* o_ o co CM CD O cn" 1- CM CM o eo in i- o oo" h-" CM o" f- V v v V CM si¬ I-. co •sf f- 8 | 2 en" o" co" cm" o" o" o" o" en CM V V v v eo sf co sf o CM § à- o q> *—» îd ro f«. C j-: CD Sf sf en cm en CM co CO CO •sf m eo sf sr o o m m m r- CM o H) I-. CO 00 m •sf CD en in o CD CD m C en cm" CM O o eo" 00 m CM CM CM r-. o in r-" in" cm" eo O CM V h- v v V sf CM 00 in CM r— sf 2 o" sf" eo" o" o" o" o" en V V v v co v co •sf CD O Vi o 13 --5 -S O 2 CD 00 en 00 CM S CM CO CM f- en SI¬ en CM en m O O 00 o o m m m r- en -S 8 m co 00_ LO I-. eo m m sf O 00 f- co eo CM 'S E co eo" eo o o O CD eo CO CM CM CM t o p: C: 2 oo" en" v v v V CO •s* CD sf -D en" o" oo" cm" o" o" o" o" Ö V V V v v 00 CM co CD E o co ri¬ en m ° £ CM "o g" o § œ CD S Table A-6: Major and minor element concentrations of representative amphibolites, tonalités and calcareous rocks from the Chiavenna Unit. Major element compositions are presented in oxides and are calculated in wt%, minor element concentrations are expressed in ppm. Amphibolite Coarse grained Tonalité dikes Metacarbonates AMPHIBOLITES Sample BM1 SM81 SM82 SM59 SS33 SC97 SC40 SC53 SC52 SC45 Si02 46,04 48,76 46,61 57,69 58,24 79,39 73,65 70,46 70,24 71,51 Ti02 1,41 0,99 0,18 0,47 0,67 0,24 0,28 0,18 0,17 0,20 Al203 16,05 16,37 16,64 19,86 18,62 5,67 4,47 4,63 4,55 8,83 Fe203' 9,69 8,91 8,41 4,47 5,34 2,24 2,28 1,52 1,45 2,99 MnO 0,15 0,16 0,14 0,06 0,07 0,12 0,10 0,09 0,08 0,05 MgO 6,46 8,8 9,23 3,48 3,79 1,07 1,28 1,09 1,08 2,20 CaO 16,31 11,68 11,85 6,53 3,93 7,63 15,36 15,74 15,81 11,01 Na20 1,61 2,31 2,22 4,57 4,49 0,41 0,16 0,33 0,31 0,79 K20 0,09 0,4 0,55 1,36 2,91 0,57 0,16 1,00 1,17 0,01 P205 0,13 0,09 0,06 0,15 0,26 0,05 0,08 0,06 0,06 1,50 L.O.I." 1,15 1,45 2,67 1,35 0,95 1,65 1,50 4,27 4,12 0,21 Cr203 0,04 0,05 0,06 0,02 0,02 0,02 0,02 0,01 0,01 0,01 NiO 0,01 0,01 0,02 0,00 0,00 0,00 0,00 0,00 0,00 0,00 Total 99,14 99,98 99,27 100,01 99,29 99,06 99,34 99,38 99,05 99,31 F 236 <10 <10 654 1892 <10 71 683 579 542 Ba 14 <10 19 198 502 113 <10 140 129 <10 Rb <8 15 27 77 225 47 14 92 103 <8 Sr 138 148 158 343 377 295 271 384 371 897 Pb <5 <5 <5 9 14 <5 12 20 21 70 Th <5 <5 <5 <5 <5 <5 7 16 21 22 U <10 <10 <10 <10 <10 <10 <10 <10 <10 <10 Nb <4 9 8 11 <4 14 32 32 34 28 La <20 <20 <20 <20 <20 <20 <20 <20 <20 <20 Ce <15 <15 <15 <15 <15 <15 <15 <15 <15 <15 Nd <25 <25 <25 <25 <25 <25 <25 <25 <25 <25 Y 38 31 26 28 15 23 42 39 40 68 Zr 104 74 55 344 149 98 67 100 89 88 V 291 187 181 85 99 30 12 <10 <10 79 Cr 323 310 319 16 34 64 79 55 48 50 Ni 122 115 133 47 44 39 50 34 30 64 Co 46 13 12 <4 <4 15 28 26 27 21 Cu 52 68 67 90 62 11 48 36 34 4 Zn 86 60 65 60 111 67 84 64 59 61 Ga 16 11 10 22 28 4 2 2 3 17 Sc 37 34 29 11 14 <2 <2 <2 <2 <2 S <50 <50 <50 539 561 770 182 <50 <50 <50 1 Fe203. total iron " L.O.I.. loss on ignition '" Trace elements below detection limit 128 Appendix Table A-7: Major and minor element concentrations of representative amphibolites, tonalités and calcareous rocks from the Chiavenna Unit. Major element compositions are presented in oxides and are calculated in wt%, minor element concentrations are expressed in ppm. Metacarbonates Ophicarbonates Sample SC72 BC18 SC42 Mera9 SC74 SO100 SO103 SO103UM SC12 Si02 60,88 67,98 59,33 16,40 33,25 18,94 22,28 21,12 33.58 Ti02 0,09 0,07 0,10 0,10 0,30 0.00 0,04 0,02 0.04 Al203 1,91 1,60 1,73 2,25 11,58 0,41 1,72 4,30 1.66 Fe203' 0,85 0,61 0,95 0,93 4.36 1,83 4,9 1,64 4.79 MnO 0,04 0,06 0,03 0,10 0.19 0,28 0,13 0,07 0.12 MgO 0,83 0,72 1,07 1,59 2.78 23,4 21,46 9,88 23.47 CaO 21,87 15,39 22,64 45,29 30.23 29,36 20,14 44,60 22.69 Na20 0,11 0,07 0,08 0,16 1.24 0.00 0.00 0,03 0.00 K20 0,25 0,43 0,23 0,22 0.11 0.00 0.00 0,00 0.00 P205 0,05 0,04 0,06 0,07 0.11 0,01 0,03 0,14 0.04 L.O.I." 12,48 12,33 13,20 31,48 14.40 25,4 29,08 17,23 12.30 Cr203 0,01 0,01 0,01 0,01 0.02 0,01 0,13 0,04 0.12 NiO 0,00 0,00 0,00 0,00 0.00 0,01 0,11 0,02 0.14 Total 99,37 99,31 99,43 98,60 98.57 99,65 100,02 99,09 98.95 F 2994 2540 3041 4027 1718 2243 1439 1927 1069 Ba 13 10 37 <10 14 <10 <10 <10 <10 Rb 25 28 16 8 10 <8 <8 16 <8 Sr 268 148 367 160 238 232 110 451 97 Pb 14 <5 12 <5 8 <5 <5 36 <5 Th 11 <5 12 <5 <5 <5 <5 26 <5 U <10 <10 <10 <10 <10 <10 <10 27 <10 Nb 27 7 27 11 <4 7 <4 33 <4 La <20 <20 <20 <20 <20 <20 <20 <20 <20 Ce <15 <15 <15 <15 <15 <15 <15 <15 <15 Nd <25 <25 <25 <25 <25 <25 <25 <25 <25 Y 35 18 37 22 23 12 <3 34 5 Zr 48 32 58 21 70 11 <10 40 <10 V <10 <10 <10 <10 <10 <10 16 <10 16 Cr 17 20 19 6 59 <6 1442 201 1431 Ni 18 17 23 18 76 109 759 241 972 Co 37 24 40 37 16 <4 27 52 84 Cu 15 <3 <3 <3 34 <3 4 4 17 Zn 39 27 58 34 98 21 27 29 38 Ga <2 <2 <2 <2 11 <2 <2 2 <2 Sc <2 <2 <2 <2 <2 <2 <2 <2 <2 S <50 <50 <50 <50 <50 <50 326 <50 360 1 Fe203: total iron " L.O.I.: loss on ignition 111 Trace elements below detection limit 129 Appendix A.3.2 Laser ablation icp-ms analyses Table A-8: Representative rare earth element abundance measured in single mineral phases of selected ultramafic rocks from the Chiavenna unit. The single element concentrations are given ppm. Ree-abundance (ppm) in amphiboles of enriched metalherzolites (type b) BU28 Mg-amphibole Tremolite La 0,01 0,01 0,01 0,01 0,01 0,01 0,18 0,24 0,30 0,01 0,01 0,23 0,26 Ce 0,01 0,01 0,01 0,01 0,01 0,01 0,88 1,34 1,63 0,58 1,40 1,46 1,48 Pr 0,01 0,01 0,01 0,01 0,01 0,01 0,20 0,19 0,29 0,12 0,25 0,27 0,30 Nd 0,01 0,01 0,01 0,01 0,01 0,01 1,49 1,64 1,54 0,96 1,88 1,35 1,59 Sm 0,01 0,01 0,01 0,01 0,01 0,01 0,84 0,68 0,70 0,54 1,21 0,59 0,80 Eu 0,01 0,01 0,01 0,01 0,01 0,01 0,38 0,34 0,37 0,19 0,27 0,35 0,33 Gd 0,01 0,01 0,01 0,01 0,01 0,01 0,95 0,73 0,95 0,53 0,01 0,66 0,75 Tb 0,01 0,01 0,01 0,01 0,01 0,01 0,32 0,21 0,27 0,18 0,33 0,24 0,27 Dy 0,01 0,01 0,01 0,01 0,01 0,01 2,02 1,63 2,15 1,40 2,37 1,81 2,15 Ho 0,01 0,01 0,01 0,01 0,01 0,01 0,46 0,45 0,43 0,30 0,37 0,43 0,49 Er 0,21 0,01 0,01 0,01 0,01 0,01 1,21 1,27 1,56 0,79 1,64 1,18 1,57 Tm 0,01 0,01 0,01 0,01 0,01 0,01 0,17 0,17 0,20 0,10 0,01 0,18 0,18 Yb 0,33 0,01 0,34 0,25 0,01 0,38 0,83 1,16 1,18 0,60 1,39 1,05 0,98 Lu 0,09 0,01 0,06 0,05 0,01 0,08 0,08 0,12 0,13 0,05 0,01 0,11 0,10 (continued) Ree-abundance (ppm) in amphiboles of enriched metalherzolites (type b) BU27 Tremolite Mg-amphibole La 0,18 0,25 0,21 0,13 0,11 0,18 0,22 0,27 0,09 0,11 0,11 0,01 0,01 Ce 0,83 1,08 1,12 0,69 0,72 0,79 1,24 1,60 0,39 0,64 0,48 0,01 0,07 Pr 0,15 0,21 0,21 0,16 0,21 0,20 0,28 0,44 0,11 0,16 0,15 0,01 0,01 Nd 1,04 1,54 1,57 1,35 1,85 1,31 1,81 3,58 1,14 1,48 0,86 0,01 0,01 Sm 0,46 0,88 0,82 0,85 0,94 1,19 0,77 1,90 0,96 1,15 0,65 0,01 0,01 Eu 0,20 0,28 0,28 0,22 0,24 0,25 0,24 0,55 0,27 0,26 0,21 0,01 0,01 Gd 0,78 1,64 1,51 1,68 1,76 1,77 1,46 3,73 2,05 2,33 1,02 0,01 0,17 Tb 0,19 0,34 0,28 0,30 0,38 0,36 0,32 0,76 0,43 0,45 0,21 0,01 0,03 Dy 1,73 2,80 2,66 3,05 3,35 2,87 2,08 6,60 3,70 3,35 1,64 0,01 0,15 Ho 0,42 0,68 0,67 0,68 0,76 0,63 0,63 1,58 0,77 0,79 0,39 0,05 0,07 Er 1,33 2,02 1,98 1,91 2,28 2,05 1,74 4,75 2,03 2,05 1,07 0,01 0,21 Tm 0,16 0,22 0,21 0,23 0,27 0,26 0,23 0,57 0,28 0,28 0,17 0,05 0,05 Yb 0,92 1,09 1,19 1,05 1,21 1,16 0,95 2,40 1,24 0,82 1,01 0,62 0,52 Lu 0,10 0,10 0,13 0,07 0,09 0,11 0,12 0,19 0,12 0,08 0,10 0,11 0,09 130 Appendix Table A-9: Representative rare earth element abundance measured in single mineral phases of selected ultramafic rocks from the Chiavenna unit. The single element concentrations are given ppm. Ree-abundance (ppm) in amphiboles of fertile metalherzolites (type a) SU6 Tremolite La 0,02 0,01 0,01 0,01 0,02 0,01 0,73 0,01 0,01 0,01 0,01 0,01 0,01 Ce 0,11 0,05 0,03 0,01 0,09 0,12 0,09 0,11 0,10 0,05 0,10 0,09 0,09 Pr 0,07 0,02 0,02 0,01 0,06 0,06 0,06 0,06 0,07 0,01 0,07 0,04 0,06 Nd 0,83 0,25 0,31 0,01 0,72 0,71 0,61 0,71 0,70 0,29 0,70 0,37 0,61 Sm 0,55 0,21 0,28 0,01 0,49 0,54 0,37 0,45 0,69 0,40 0,66 0,26 0,41 Eu 0,29 0,09 0,15 0,01 0,25 0,25 0,22 0,21 0,28 0,16 0,26 0,12 0,19 Gd 1,16 0,46 0,70 0,01 1,13 1,03 0,99 0,79 1,27 0,76 1,15 0,53 0,73 Tb 0,20 0,10 0,14 0,01 0,25 0,22 0,22 0,21 0,27 0,13 0,18 0,10 0,16 Dy 1,51 0,72 1,28 0,01 1,90 1,64 1,54 1,40 1,69 1,13 1,44 0,80 1,15 Ho 0,38 0,15 0,27 0,01 0,43 0,40 0,33 0,35 0,44 0,23 0,37 0,18 0,25 Er 0,95 0,62 0,75 0,01 1,34 1,17 1,10 1,07 1,50 0,69 1,17 0,61 0,83 Tm 0,15 0,09 0,11 0,01 0,20 0,20 0,16 0,16 0,24 0,11 0,13 0,10 0,11 Yb 1,05 0,67 0,57 0,01 1,34 1,19 1,20 0,95 1,53 0,82 0,95 0,87 0,83 Lu 0,17 0,08 0,10 0,01 0,16 0,15 0,16 0,13 0,18 0,10 0,16 0,13 0,12 (continued) Ree-abundance (ppm) in amphiboles of refractory metalherzolites (type c) SU76 Mg-amphiboles La 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Ce 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,02 0,01 0,01 0,01 0,01 0,01 Pr 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Nd 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Sm 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Eu 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Gd 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 0,01 Tb 0,03 0,03 0,03 0,03 0,01 0,03 0,02 0,04 0,01 0,01 0,01 0,01 0,03 Dy 0,29 0,29 0,23 0,37 0,28 0,31 0,29 0,39 0,40 0,33 0,39 0,25 0,26 Ho 0,10 0,08 0,08 0,12 0,10 0,10 0,11 0,12 0,14 0,08 0,13 0,06 0,08 Er 0,37 0,43 0,34 0,50 0,33 0,37 0,38 0,55 0,59 0,35 0,53 0,30 0,39 Tm 0,06 0,07 0,06 0,08 0,08 0,09 0,07 0,09 0,07 0,07 0,09 0,06 0,07 Yb 0,57 0,64 0,54 0,59 0,56 0,63 0,64 0,70 0,73 0,58 0,68 0,46 0,54 Lu 0,08 0,09 0,08 0,12 0,10 0,09 0,07 0,12 0,11 0,08 0,11 0,07 0,09 Appendix A.3.3 ICP-MS ANALYSES Table A-10: Bulk rock ICP-MS analyses of representative mafic rocks from the Chiavenna Unit. Bulk rock icp-ms analyses of Amphibolites concentrations in ug/g BM4 SM71 SM59 SM37 SM34 BM8 SM24 SM54 SM32 SM18 SM55 SM30 Rb 4,45 3,85 80,7 7,09 6,48 4,03 2,06 2,21 0,61 1,77 2,43 0,57 Sr 313 175 354 181 178 288 57,8 168 176 142 171 17,9 Y 43,6 24,7 17,1 42,8 35,3 33,4 25,1 29,7 34,8 22,3 29,8 7,22 Zr 2,93 18,8 5,54 51,4 24,7 30,1 28,2 23,7 38,5 11,5 23,7 8,23 Nb 7,59 0,872 4,57 2,22 1,94 1,53 0,395 1,65 1,87 0,572 1,83 0,248 Ba 95,9 9,35 218 29,7 11,1 51,2 4,68 5,96 1,89 2,09 3,97 0,29 Hf 0,223 0,954 0,332 2,22 1,19 1,24 0,992 1,13 1,42 0,515 1,13 0,242 Ta 1,27 0,272 1,99 0,51 0,47 0,305 0,155 0,445 0,687 0,208 0,475 0,138 Th 0,69 0,0455 2,22 0,127 0,137 0,0812 0,0869 0,0981 0,107 0,0588 0,119 0,0131 U 0,212 0,018 1,45 0,103 0,257 0,223 0,333 0,134 0,0414 0,0776 0,219 -0,004 La 11,453 5,0895 9,3811 5,5758 5,1258 8,6432 5,0853 2,1142 5,0805 3,9589 5,4363 0,0232 Ce 21,401 7,0674 23,236 14,382 11,229 10,929 7,9183 10,367 10,624 7,0411 10,376 1,8553 Pr 3,3958 1,1268 3,2129 2,4866 1,8447 1,7308 1,2729 1,6326 1,7382 1,1358 1,7197 0,2031 Nd 16,842 5,8713 13,959 12,938 9,8117 8,8396 6,8196 8,3108 8,2079 4,9859 8,2876 -0,072 Sm 5,3042 2,3553 3,6121 4,4937 3,6447 3,0463 2,6558 2,9942 3,0705 2,0221 3,1716 0,4501 Eu 1,7626 0,8164 0,779 1,5763 1,2558 1,3811 1,0144 1,0473 1,2647 0,8921 1,2263 0,2333 Eu 1,7503 0,8308 0,7743 1,5726 1,2671 1,3971 1,0376 1,0544 1,1889 0,8068 1,1321 0,143 Gd 6,3087 3,1134 3,1503 5,8758 4,7295 3,9271 3,4353 3,8876 4,3876 2,8326 4,1247 0,747 Gd 6,3366 3,022 3,2495 5,8614 4,5982 3,8211 3,3296 3,8491 4,5367 2,9142 4,2166 0,861 Tb 1,0685 0,5323 0,4571 1,0384 0,8286 0,7146 0,5789 0,6814 0,8314 0,5115 0,7284 0,1465 Dy 7,5768 4,0958 3,0695 7,3905 6,0889 5,5216 4,3079 5,04 5,9305 3,8047 5,1753 1,2605 Ho 1,6037 0,8954 0,5905 1,5853 1,3082 1,2111 0,8997 1,0853 1,2816 0,8351 1,1042 0,2686 Er 4,3474 2,4733 1,5588 4,3895 3,6733 3,4661 2,4471 2,995 3,5249 2,3634 3,0679 0,6518 Tm 0,6381 0,3716 0,2298 0,6462 0,5436 0,5271 0,3575 0,4449 0,5123 0,3523 0,4529 0,0971 Yb 3,815 2,2967 1,3726 3,9508 3,3467 3,2087 2,1766 2,6728 3,1058 2,1701 2,7301 0,5516 Lu 0,569 0,3494 0,2013 0,5857 0,5035 0,4798 0,3268 0,4079 0,4719 0,3329 0,4137 0,0723 132 Appendix TfîM'ffl (continued) Bulk rock icp-ms analyses of Amphibolites element concentrations in Lig/g BM1 SM63 SM28 SM43 SM80 BM25 BM34 BM15 SM94 BM11 BM13 SM81 BM14 Rb 0,997 3,36 2,25 4,22 39,4 5,76 0,438 5,04 30 8,63 7,17 17,3 3,72 Sr 144 180 172 287 170 254 40,9 113 172 186 155 164 197 Y 32,5 28,1 25,1 25,9 23,2 36,3 24,5 36,2 36,3 27,8 37,3 23,8 32,3 Zr 6,76 21,5 12,3 19 12,1 41,4 34,7 12 20 7,57 12,3 20,7 10,1 Nb 1,81 1,98 1,18 11,1 0,982 2,39 2,88 3,17 3,57 1,76 3,15 1,13 17,2 Ba 1,93 8,26 3,72 20,9 30 13,1 6,37 19,7 14,2 36 9,41 12,3 11,5 Hf 0,379 0,988 0,622 0,869 0,362 0,763 0,729 0,359 0,632 0,25 0,362 0,515 0,353 Ta 0,457 0,481 0,409 1,98 0,238 0,351 0,369 0,386 1,13 0,298 0,403 0,328 7,62 Th 0,115 0,127 0,107 7,39 0,0475 0,132 0,156 0,163 0,154 0,0987 0,167 0,219 0,179 U 0,0573 0,292 0,0567 2,11 0,101 0,147 0,0563 0,0634 0,463 0,0756 0,0656 0,58 0,0601 La 3,8768 4,5905 11,037 27,29 6,7416 7,1168 9,6332 9,3353 8,2684 2,9721 5,5874 7,6595 6,3789 Ce 8,4467 10,938 8,6814 55,471 6,1212 12,495 10,121 12,795 13,195 9,8512 13,095 7,8692 10,895 Pr 1,4879 1,7618 1,4279 5,9963 1,0351 1,9647 1,4766 2,1147 2,1647 1,6124 2,1347 1,2784 1,7666 Nd 7,1968 8,0555 6,6108 21,913 6,1863 12,353 7,7212 12,053 12,153 9,292 12,309 7,8686 9,8221 Sm 3,0526 3,0026 2,5784 4,3447 2,2989 3,6316 2,5447 4,1674 4,2121 3,2037 4,1874 2,5895 3,4058 Eu 1,1958 1,1858 1,0658 0,9814 0,8625 1,4626 1,1379 1,4026 1,4226 1,1237 1,4526 1,0079 1,1379 Eu 1,0966 1,0966 0,9837 0,8787 0,8621 1,4547 1,1158 1,4047 1,4103 1,1324 1,4647 0,9963 1,1458 Gd 4,2284 3,8784 3,4403 3,8761 3,1489 4,6863 3,4047 5,3521 5,3882 4,1711 5,5061 3,5168 4,4684 Gd 4,3157 4,0157 3,5871 4,54 3,3439 5,0297 3,6692 5,745 5,7512 4,3663 5,775 3,7816 4,7721 Tb 0,7781 0,6937 0,6109 0,6285 0,57 0,7989 0,5998 0,9332 0,9535 0,7218 0,9777 0,6331 0,8139 Dy 5,5805 4,9174 4,37 4,0226 3,7989 5,47 4,0316 6,2326 6,2595 4,7663 6,4758 4,0321 5,4168 Ho 1,2271 1,0471 0,95 0,8546 0,8236 1,2032 0,8792 1,3361 1,3261 1,0145 1,3889 0,8552 1,1732 Er 3,3917 2,8989 2,6072 2,3555 2,3214 3,3936 2,5097 3,6891 3,7163 2,9059 3,8591 2,3742 3,3308 Tm 0,4885 0,4273 0,386 0,3469 0,3414 0,5001 0,3734 0,5418 0,5505 0,4235 0,572 0,3484 0,4988 Yb 2,8861 2,6461 2,35 2,1659 2,0157 2,9795 2,2416 3,2634 3,2095 2,4617 3,3034 2,0776 2,8695 Lu 0,4055 0,3927 0,35 0,3275 0,3195 0,4762 0,3701 0,5135 0,5164 0,4018 0,5367 0,3365 0,4761 133 Appendix Table A-11: Bulk rock ICP-MS analyses of representative ultramafic rocks from the Chiavenna Unit. BULK-ROCK ICP-MS ANALYSES OF ULTRAMAFIC ROCKS. ELEMENT CONCENTRATIONS ARE GIVEN IN Lig/g FERTILE METALHERZOLITES (TYPE A) Enriched metalherzolites (Type B) SU4 SU6 BU6 SU15 BU28 SU77 BU35 BU27 SU51 BU41 Rb 0,259 0,251 0,108 1,2 0,46 13 0,0571 0,113 0,231 0,122 Sr 4,21 4,23 4,91 7,16 5,28 1,65 3,71 5,33 38,7 6,29 Y 2,29 3,74 2,68 2,1 2,36 1,82 1,06 1,9 1,52 1,88 Zr 0,658 1,14 0,621 0,739 0,405 0,463 0,928 0,716 0,379 0,877 Nb 0,0863 3,9 0,0483 0,153 0,195 0,0897 0,292 0,102 0,134 0,109 Ba 0,0979 0,364 0,195 0,917 0,228 3,41 0,0998 0,174 1,03 0,522 Hf 0,0319 0,0606 0,0306 0,0378 0,021 0,0268 0,0181 0,0297 0,0168 0,0253 Ta 0,0605 2,04 0,0189 0,0738 0,0487 0,0487 0,0273 0,0187 0,0343 0,0923 Th 0,00045 0,00563 0,00044 0,00208 0,0155 0,00086 0,00396 0,0187 0,00181 0,00407 U 0,00236 0,00234 0,00157 0,104 0,029 0,0976 0,0106 0,0211 0,00664 0,0264 La 0,0055 0,0112 0,0251 0,0216 0,1319 0,0634 0,0596 0,0844 0,1193 0,0975 Ce 0,0281 0,0484 0,102 0,0372 0,3519 0,1821 0,1799 0,1941 0,2467 0,1667 Pr 0,0139 0,0203 0,0289 0,0137 0,0537 0,0281 0,0321 0,0264 0,0345 0,0337 Nd 0,156 0,2404 0,2509 0,1395 0,3122 0,1745 0,1906 0,1733 0,2039 0,2027 Sm 0,1231 0,1877 0,1538 0,1047 0,1343 0,0934 0,0707 0,088 0,0911 0,0915 Eu 0,047 0,0811 0,0657 0,0455 0,0617 0,0319 0,0337 0,0255 0,0311 0,049 Gd 0,2359 0,3599 0,2805 0,194 0,235 0,1645 0,1145 0,1698 0,1519 0,1704 Tb 0,0472 0,0723 0,0551 0,0406 0,0448 0,0332 0,0217 0,0336 0,0281 0,0336 Dy 0,3475 0,5325 0,3965 0,2982 0,3293 0,2447 0,1527 0,2522 0,1974 0,2469 Ho 0,0794 0,1248 0,091 0,0689 0,0774 0,0589 0,0368 0,0603 0,0469 0,058 Er 0,24 0,3671 0,27 0,2056 0,2315 0,1836 0,1139 0,1803 0,1458 0,1797 Tm 0,0373 0,0565 0,041 0,0343 0,0376 0,031 0,0199 0,0278 0,0247 0,0284 Yb 0,2407 0,3498 0,2467 0,24 0,2408 0,2129 0,1319 0,1711 0,1799 0,1783 Lu 0,0405 0,0589 0,0421 0,0418 0,0442 0,0399 0,0275 0,034 0,0321 0,031 134 (continued) Element concentrations in Lig/g Refractory metalherzolites (Type C) Sample SU75 SU76 SU8 SU16 Mg13 Rb 0,552 0,359 0,0754 2,48 0,363 Sr 0,153 0,157 0,76 2,79 0,364 Y 1,7 1,07 0,661 0,887 1,51 Zr 0,254 0,196 0,35 0,538 0,285 Nb 0,148 0,0642 0,0584 0,0646 3,61 Ba 0,296 0,169 0,655 1,27 0,176 Hf 0,0142 0,0113 0,0128 0,0273 0,0159 Ta 0,0774 0,0322 0,0305 0,0145 1,87 Th 0,00207 0,00335 0,00036 0,00143 0,00466 U 0,00579 0,015 0,00162 0,0563 0,0165 La 0,014 0,011 0,0059 0,0098 0,0154 Ce 0,0366 0,022 0,018 0,0118 0,0454 Pr 0,0074 0,0044 0,005 0,0042 0,0082 Nd 0,0708 0,0391 0,0454 0,0433 0,0612 Sm 0,0458 0,0279 0,0299 0,0332 0,0423 Eu 0,0112 0,0045 0,01 0,0118 0,0109 Gd 0,1076 0,0622 0,0553 0,0675 0,0935 Tb 0,0254 0,015 0,0103 0,016 0,0236 Dy 0,2061 0,1287 0,0829 0,1202 0,2096 Ho 0,0573 0,0363 0,0216 0,0304 0,0544 Er 0,1804 0,1251 0,0735 0,1002 0,1781 Tm 0,0316 0,0224 0,0138 0,0186 0,0302 Yb 0,2085 0,1459 0,1054 0,1443 0,1923 Lu 0,0385 0,0287 0,0228 0,0299 0,0334 135 Appendix A.3.4 Argon analyses Table A-12: Ar/Ar isotope-analyses. The argon isotopes are given in picoliters per gram (pl/g). The first and second columns contain the number of heating steps and the related temperature values, respectively. The *Ar indicates the difference between total40Ar minus the atmospheric40Ar. Sample BM4 Step T(*C) 40artot. 39AR 38AR 37AR 36AR AGE± 15 1 710 906 37.6 12.2 512.33 3.43 (-)22.3+16 2 905 2848 309 24.3 3995.9 8.68 22.5±1.8 3 951 1567 23.4 19.4 4173.7 4.2 32.6±2.6 4 980 1721 267 22.8 5971.71 4.66 35.5±1.9 5 1003 1097 146 14.4 2771.69 3 39 25.0±5.6 6 1024 822 74.8 7.59 1421.15 2.63 24.3±8.8 7 1035 876 82.2 9.55 1719.38 2.97 18.5±4.9 8 1057 960 92.6 7.97 2033.99 1.83 73.5+4.2 9 1077 2096 179 19.0 4018.2 6.7 28.9±3.7 10 1090 2369 104 11.9 2484.5 7.32 45.0±7 11 1129 9330 37.0 3.43 905.74 2.0 12.9±16 12 1162 1152 101 10.1 2432.1 2.91 56.1 ±5.8 13 1217 1402 142 15.0 3350.0 2.69 71.6+2.5 14 1267 1530 124 12.4 2810.5 3.36 71.6±3.3 15 1389 3391 461 47.3 9955.81 7.06 52.9±0.83 34.73 mg J=1.84*10'3 Ar* 8480 (continued) Sample SM28 Step T(°C) 40ARTOT. 39AR 38AR 3rAR 36AR AGE±15 1 711 17902.7 10.55 2.62 347.22 6.37 (-)75.73+66 2 914 985.4 13.34 1.48 770.41 3.18 94.34±62 3 956 665.0 14.26 0.43 592.13 2.37 9.05±38 4 985 1520.1 203.84 16.6 10207.5 5.02 48.09±1.3 5 1022 2690.1 536.34 41.05 24690.45 10.30 34.85±0.62 6 1058 571.2 30.18 2.39 1259.02 1.70 65.46±11.0 7 1094 713.2 48.22 4.58 2177.16 2.94 1.87±14 8 1155 1140.1 121.54 9.5 5923.68 4.58 23.57±3.7 9 1209 901.0 48.09 4.05 2385.84 3.13 39.53±13 10 1309 132.1 84.71 6.53 4181.54 4.78 32.20±5.2 11 1387 1608.8 96.57 8.67 4533.36 6.3 11.34±6 136 Appendix (continued) Sample SM44 old Step TfC) 40ARTOT. 39AR 38AR 37AR ^AR AGE+18 4 960 1390 5180 145 14300 7.92 28.8±0.1 5 972 1300 4870 126 13400 5.23 30.3±0.06 6 1010 8320 3020 75 8470 4.38 30.0±0.11 7 1090 3330 996 28.2 5180 3.17 33.1±0.46 8 1153 1810 325 11.1 8590 5.68 28.9+1.7 9 1289 1240 174 60.5 6410 4.44 28.4±2.5 10 1443 2610 406 17.6 8770 7.10 34.4±1.1 43.43 mg J=1.84*10'3 Ar* 37900 (continued) Sample SM44 new Step T(-C) 40ARTOT. 39AR 38AR 37AR 36AR AGE±18 1 710 1640 11.3 2,42 173 5,17 56.2±25 2 917 2,360 182 6,26 877 4,52 31.1±1.7 3 927 2360 288 8,71 888 3,59 24.4+1.1 4 943 1630 202 5,41 593 1,48 31.6+1.6 5 947 1290 171 5,34 479 0.84 32.4±1.6 6 959 1460 200 4,87 569 2,02 23.5±1.6 7 970 1460 181 4,14 510 2,14 24.6±2.0 8 1007 1920 233 3,68 683 2,52 27.1±1.1 9 1078 1370 119 3,76 836 2,57 29.4±2.4 10 1146 1070 31.6 2,05 1430 2,67 65.7±12 11 1426 2200 57.7 3,50 3830 6,68 47.4±3.3 -i 9« Q mr i- «a*-in3 Ar* O/H n 28.8 mg J=1.84*103 Ar* 9470 (continued) Sample SM60 Step T("C) 40ARTOT. 39AR 38AR 37AR 36AR AGE±18 1 714 5120 39 2.4 538 6.81 777±8.5 2 910 9360 172 14.6 1790 5.40 479±2.8 3 943 2590 292 30.2 2040 2.62 78.7±1.3 4 963 2300 453 46.9 2970 2.88 43.5±1.2 5 990 5680 1580 170 9870 5.56 35.8±0.2 6 1001 4330 1250 125 7850 4.33 34.6±0.29 7 1018 16508 330 32.8 2030 3.25 31.0±1.3 8 1045 1410 192 19.8 1680 2.63 46.8±3.2 9 1068 2510 547 58.1 4380 3.76 37.3±0.98 10 1094 3380 774 80.6 6530 3.56 43+0.45 11 1158 6010 1090 116 16000 6.49 57.3+0.32 12 1259 10000 1070 108 31900 11.40 99.7±0.36 13 1393 4250 593 59.4 9420 8.38 49.6+0.73 41.9 mg J=1.84*10"3 Ar*46100 137 Appendix (continued) Sample SM71 Step T(°C) 40ARTOT. 39AR 38AR 37AR ^AR AGE±18 1 710 96396 48.00 70.3 256.3 324.6 118.72±79 2 908 29128 106.2 33.6 885.2 94.75 130.01±12 3 935 4058 52.5 16.8 729.4 13.05 57.75±24 4 963 3011 102.1 32.6 1692 9.61 34.53±6.9 5 988 2652 152.7 48.0 2538 8.22 32.29±4.9 6 1000 6030 1069 310.2 17673 14.73 33.50±0.45 7 1020 640 1049 306.0 16369 15.73 35.41±0.33 8 1057 2058 110.9 33.0 17145 6.20 37.88±5.9 9 1080 2260 76.9 23.6 1244 6.68 58.34+8 10 1093 3110 148.6 46.2 24140 9.35 42.27±4.4 11 1121 3501 246.1 73.6 39431 10.20 37.91±1.6 12 1151 3798 222.5 67.2 35131 11.49 35.71±2.6 13 1217 3835 258.8 76.7 40970 11.18 38.53±2.5 14 1264 4111 229.1 69.9 3591 12.40 37.19±2.5 15 1338 4428 252.5 75.7 3915 12.55 47.60±2.3 16 1434 6545 78.5 25.0 1186 20.81 74.09±11 32.55 mg J=1.84*10"3 Ar* 14600 (continued) Sample SM81 Step TfC) 40ARTOT. 39AR 38AR 37AR 36AR AGE±18 1 702 18086 112.4 32.34 536.2 58.55 85.49+8.3 2 947 3224 176.3 8.29 1978 10.60 16.20+3.3 3 1009 9907 2982 116 45256 18.59 31.24±0.1 4 1036 2577 561.6 22.51 7816 5.04 35.48±0.6 5 1051 949 69.35 3.08 10196 2.76 36.01±9.6 6 1060 1175 78.48 3.68 11938 2.86 63.14±8.3 7 1093 1305 155.2 7.83 2321 3.24 39.93±3.6 8 1127 1534 246.9 10.24 3718 3.7 34.77±3.0 9 1160 1614 227.7 8.49 3463 4.32 31.31±1.5 10 1268 1733 242.7 9.96 3635 5.37 20.72±2.4 11 1395 2365 469.2 17.95 6559 6.51 26.44±1.0 12 1450 1463 21.69 0.613 27.10 4.69 53.06±41.0 138 Appendix A.3.5 ELECTRON MICROPROBE ANALYSES ULTRAMAMFIC ROCKS Table A-13: Representative amphibole compositions of selected ultramafic rocks. Typical compositions of ultramafic Amphibole (wt%) Tremolite Mg-amphiboles Sample SU77* SU6 BU2 SU76 BU28 BU28 BU27 Si02 54.95 56.03 57.07 56.41 58.37 58.24 57.81 58.01 57.74 Ti02 0.17 0.22 0.04 0.04 0.06 0.00 0.00 0.00 0.00 Al203 3.03 1.37 0.15 1.28 0.07 0.07 0.06 0.09 0.02 Cr203 0.14 0.20 0.11 0.06 0.02 0.03 0.01 0.06 0.00 Fe203 1.85 2.45 2.28 1.55 1.57 1.56 1.97 1.85 2.17 FeO 0.86 0.19 0.00 5.59 5.66 5.62 7.10 6.65 7.81 MnO 0.07 0.09 0.13 0.24 0.21 0.20 0.36 0.28 0.57 NiO 0.07 0.08 0.15 0.11 0.08 0.05 0.07 0.08 0.09 MgO 23.11 23.80 24.10 31.31 31.08 30.89 29.16 29.44 28.55 CaO 12.66 12.61 13.07 0.34 0.35 0.34 0.61 0.65 0.63 Na20 0.50 0.16 0.04 0.00 0.03 0.02 0.00 0.01 0.01 K20 0.05 0.04 0.01 0.01 0.00 0.00 0.00 0.00 0.00 H20 2.18 2.18 2.18 2.20 2.22 2.21 2.20 2.20 2.19 Total 99.63 99.42 99.33 99.14 99.72 99.22 99.37 99.33 99.77 (P.f.u.) 23 oxygens and 15 + A cations Ramp norm with 23 oxygens and fixed Fe3/Fetot Si 7.56 7.70 7.84 7.68 7.88 7.89 7.90' 7.90 7.89 Ti 0.02 0.02 0.01 0.00 0.01 0.011 0.0002 0.0002 0.00 Al 0.49 0.22 0.03 0.20 0.01 0.00 0.01 0.014 0.002 Cr 0.01 0.02 0.01 0.01 0.16 0.16 0.001 0.007 0.0003 Fe3 0.19 0.25 0.24 0.16 6.25 6.24 0.20 0.19 0.22 Fe2 0.10 0.02 0.00 0.64 0.64 0.64 0.81 0.76 0.89 Mn 0.01 0.01 0.02 0.03 0.02 0.02 0.04 0.03 0.07 Ni 0.01 0.01 0.02 0.01 0.05 0.05 0.008 0.009 0.01 Mg 4.74 4.88 4.93 6.35 0.01 0.005 5.94 5.98 5.82 Ca 1.87 1.86 1.92 0.05 0.00 0.00 0.09 0.09 0.09 Na 0.13 0.04 0.01 0.001 0.002 0.003 0.0000 0.002 0.002 K 0.01 0.01 0.00 0.00 0.01 0.005 0.001 0.001 0.001 H 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 2.00 XMg(Fetot) 0.94 0.95 0.95 0.89 0.89 0.89 0.87 0.86 0.84 Fe3+/Fe(tot 0.66 0.92 1.00 0.20 0.20 0.20 0.02 0.20 0.20 X o t— ^ co 0 0 co sf eo r- T- eo sf CD c- r- 0 0 r- f- •sf in co eo 00 eo m CO r-. CM CM f- eo CM en CM CM *f 0 c CO sf eo i— (0 m 0 0 in q> S CM o •sf CM en o p O O 0 m in O co O O 0 0 en p p cri c in en 3 CM CO CM sf CM O O 0 O O p p b Q 2 b b b b b b g b * m co co en co b b b b b b sf b b 00 * U. 15 b «* a 0 0 e LU CD N- CD T~ sf in CD CO 1-. 00 co CM CM sf 0 0 m 0 T- p T— co en eo co T- CM T— T- T3 sf f- eo r- eo CM eo 1-- CM c- T- v- •i= or co en i~- 00 0 0 m s C eo 00 •sf CM CM CM t- o en eo eo p eo O O O O O 0 O 0 en p csi p CM CT) co CM co _i b en CM O 0 0 O p O p O 0 b co b b b b b co b b b Ol p b * CD X to CM b CM b b b b b •sf b b b 00 * CO O c CO eo -c en X>. Q. CO 0 co ^~ l'¬ 0 CT) CD in 0 CM in 0 CD 0 o CO m 0 en in CO co CM CD sf o CM m en o r— -sf co 00 CM CD en 0 CM •sf 0 m 0 co fo o O O co p p eo 00 r- 0 0 0 0 0 0 CM l'¬ 0 O 0 0 en O 3 CM p p p CM 00 -s- LO sf en CO iri 0 0 0 0 p 0 p O 0 v- CD eo b b b b b b sf b b b en p b .C co b b b b b b b in b b b 06 * « S Vj r- in co 00 CD co C eo r- CO en 0 m sf 0 0 0 m O T- •sf r— in CM co i>- CM co r- t— O eu CM CO o CM en O (0 0 0 0 0 0 ^f m ai •sf CM 01 O eo O I-; CM o p o c 0 0 0 0 0 0 0 0 en 3 p i-^ p p CM 00 en eo eo eo -CD CM CO 0 0 0 0 p p 0 p 0 0 b m •sf d b b b b co b b b en b «fc eo b b b b b b b iri b b b CO * « 0ra LU 0 r" CM 0 CM CM 0 CD 00 CM T- c- 00 CM 0 r- 0 co or o m T— CO CM "D co LO T- f- T— T— CM 0 o CM 00 o eo eo o eo CM CM en CM CM in co eo 0 0 0 O 0 •sf c: p •sf o CM O O o C en 0 sf eo 0 eo O r- O O O 0 en 0 o 3 CM p CD CM cri CO CM co CM .o CM •sf 0 p p p O 0 O O O 0 b _^ t- m •sf b b b b b eo b b b """ en L0 b C sf b b b b b b b iri b b b 00 « * -* S ci 55 z p p O o p (0 0 0 0 O 0 O 0 0 0 3 cri p oi CT) CM CO 00 -1 CO b 0 0 0 O p 0 0 0 eo b b b b b •sf b b b en c b b _j O g b b b b b b b b b b 3 ra u. 0 "O O co CD co CM 0 co CM sf CD 0 co eo CM -•—# CO sf T- en in co eo CO CM o m co i-. o o m 0 0 co m 0 en O z o (D O 3 sf O p p sf p p p co 0 0 0 0 0 O 0 0 0 00 o en b eri cri en 0 0 0 CM p O 00 0 0 0 m co b b b b b •sf b b b Ol b b CD K b b b b b b b b 0b b 00 Oeo Q_ 2 O 0 _1 LU 0 < J a> « .CD eo o Q. O 6 n Q- S O d q O O O O CO CM 2 >. O O Ol O O c o> 10 ra >- O a> c n co CM 0 ^ Cl) 4) 0 n E È (0< tô P < O u. u. 2 z 2 U Z * X 1- iô H < ü u. U. 2 Z 2 Ü Z * X u. u. .is 00 co in sf CM "O r- CO CO sf o m c- CD r- CO CM 00 o T— 00 o co CM eo cn m CM CO in LO o sf O CD CM sf CM •sf in CO CO CO CO Ol CO CD p p (0 o o O o 3 h^ I-; CT) in sf CO b b CO p p o o o o p b CO b b b b b b CT) b in CM O) b b b b b d d d b ro .co sf _l CD CO eo 00 in •sf CO 1- 00 CO r- sf LU co T— eo sf •sf r- sf r— CO Ol CM CO T— r- •sf cn CM o 5 •sf o CM TD m o o o v- Z 3 CO p O CM sf O o o o C o o sf CO CM o in o o o o CO CT) CO CO CM I-; o b 0. m b b CO b b b b b CO p o °! p p o o o b b in o co CM en (0 b b b - b b d d d d d 2 c •sf o •sf •sf o o ra CM CM 00 00 CM •sf CO sf CD co CO o CO Ol sf o o o o sf CM CO eo CO O o Ol 1"- sf ^ -=f OO en CO o j: o o 00 CO CD o o m Ol o o r— Li. r- CM I-; eo CM CM CM r- CM CD m °! p p p O o o o o o o co b co 1-^ en o o CO CO 00 o o b o LL b d in CM b b b b b en p p p b Li. d d b b d d d b b b 0 0 Li. 0 in 00 r- CM in _ oo en Ol o CM cn o m sf sf T— in CO T— in CM 00 CO eo in CD t- o o CM CO o o I-. CD o o o o o sf o co T— eo f- CM o o o p sf o T- o T- CD o o cn o o o CD p eo cri (0 o 00 o o o P b b b iri b b b b b °! p O p O o m b eo en c d b b b d b b b b Ob Li b b ZLU LU E X i_ in CM ,_ CO CM i^ CO Ol o eo 00 r— eo o CM CO o o CO or CD m sf CD r- •sf 00 CO t— o c m V- CO v— m in sf CM eo T— o o Ol CM CM eo CM o >- en o co O en O p m o O sf O o o o o en 0. CD iri p p p en X o iri Q. p o o O o o O oq p o o O -> in b d b b b b CO b b b b b X en O b d b b b b d b b b K eo tc et> 0 O O) T3 CO in r- CM CO CM 00 00 Ol 10 co CT) m o o o m Q. eo CO •sf sf r- CO 00 CM •sf CO sf o 1 o o o ^ o o o o lo O Ol r- in m o o o p p o co o CM O m o o in o o o 01 o en CO p p cd b °> O o s- o 00 c b b b b iri b b b b b o CM o O p o o o d b b LU m CO t— T— « LU T— T~ d d b b O b b b b b Ht * Mafic rocks Table A-15: Selected representative amphibole compositions from amphibolites Amphibole compositions (wt%) Sample BM4 SM28 SM81 Si02 44.49 44.75 50.50 47.35 46.38 47.60 48.21 48.38 48.52 Ti02 0.55 0.21 0.21 1.01 1.09 0.99 0.85 0.79 0.62 AI2Oj 13.00 12.60 7.57 8.99 10.05 9.06 8.85 8.17 8.32 Cr203 0.06 0.03 0.03 0.04 0.01 0.05 0.09 0.11 0.00 Fe302 3.83 6.01 4.10 3.13 4.73 4.06 1.85 2.35 3.17 FeO 11.07 11.01 7.57 10.21 9.31 9.39 9.70 9.98 8.50 MnO 0.30 0.50 0.32 0.24 0.26 0.24 0.22 0.24 0.26 NiO 0.07 0.00 0.01 0.00 0.05 0.00 0.08 0.02 0.03 MgO 11.37 10.58 15.22 13.69 13.19 13.70 14.31 13.97 14.61 CaO 11.88 11.34 11.76 11.94 11.65 11.86 12.43 11.94 11.95 Na20 2.03 2.07 1.40 1.88 2.04 1.69 1.14 1.44 1.43 K20 0.30 0.34 0.16 0.15 0.17 0.14 0.53 0.33 0.20 CI 0.01 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 H20 2.07 2.07 2.12 2.08 2.08 2.09 2.09 2.07 2.08 F,CI=0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Total 101.02 101.50 100.08 100.71 101.00 100.89 100.34 99.78 99.70 (P.f.u.) 23 oxygens and 13 cations + K+Na, NAMP norm Si 6.4496 6.4850 6.4850 6.8211 6.6760 6.8305 6.9312 7.0011 6.9840 Ti 0.0598 0.0224 0.0224 0.1095 0.1183 0.1066 0.0914 0.0855 0.0666 Al 2.2215 2.1519 2.1519 1.5261 1.7058 1.5322 1.5002 1.3927 1.4124 Cr 0.0072 0.0033 0.0033 0.0050 0.0008 0.0061 0.0100 0.0124 0.0000 Fe3 0.4183 0.6549 0.6549 0.3396 0.5120 0.4385 0.1998 0.2555 0.3438 Fe2 1.3422 1.3350 1.3350 1.2299 1.1210 1.1272 1.1662 1.2083 1.0230 Mn 0.0368 0.0613 0.0613 0.0296 0.0312 0.0293 0.0269 0.0294 0.0316 Ni 0.0077 0.0000 0.0000 0.0000 0.0055 0.0000 0.0087 0.0025 0.0031 Mg 2.4568 2.2862 2.2862 2.9391 2.8294 2.9296 3.0656 3.0126 3.1354 Ca 1.8448 1.7611 1.7611 1.8435 1.7961 1.8239 1.9147 1.8508 1.8436 Na 0.5699 0.5818 0.5818 0.5241 0.5689 0.4707 0.3175 0.4031 0.3984 K 0.0564 0.0624 0.0624 0.0272 0.0316 0.0256 0.0979 0.0614 0.0367 CI 0.0022 0.0002 0.0002 0.0000 0.0005 0.0000 0.0000 0.0000 0.0007 H 1.9978 1.9998 1.9998 2.0000 1.9995 2.0000 2.0000 2.0000 1.9993 xMg (Fetot) 0.583 0.535 0.707 0.652 0.634 0.652 0.692 0.673 0.696 Fe3+/Fetot 0.238 0.329 0.328 0.216 0.314 0.280 0.146 0.175 0.252 AI(IV) 1.550 1.515 0.863 1.179 1.324 1.170 1.069 0.999 1.016 AI(VI) 0.671 0.637 0.398 0.347 0.382 0.363 0.431 0.394 0.396 Na(M4) 0.146 0.235 0.216 0.142 0.204 0.174 0.085 0.149 0.146 Na(A) 0.424 0.347 0.169 0.382 0.365 0.297 0.232 0.254 0.252 •is T3 r- eo in in CD CD •sf >- in Sr-- 00 en -sf m C CD o cm m o •sf m cm en co r- eo sf en cm co ai co CD •sf t- •<- I-. CM p co o o s- >sf sf co en CD CM CM CO t- CM O -sf O o o o cri ö - co eo CD CO r- 10 in CO -sf oo o m co in 00 en •sf t- CD en co 01 i-. T- r- co •* o o f- o CD CM oo o o m m 00 co -sf cm en r- CM CM CD CM 00 n. en CM co sf p O o m o CO CM CM -sf -s- CM 00 CM -s- •s- r— eo ,_: co O O en co sf o en ^ cri b eo 00 b b O O CM O O p i-^ •sf b b O O O O O CO CM co co •- en O o CD co c- 1-- en O •sf •sf m o f- m CD CM oo eo en sf m •sf m co in CM •sf CO en eo 01 m in f- CM sf O O o o o CM •sf CM 00 p cd T- o CD sf o o en p b CT) b cd cd b b T^ d d CM T^ CM t- o o o d d b b b en en cm en h- co T- 0 en m CD 00 00 f- r- co 00 m 01 o sf -sf f- en ô £ CO CO O eo f- sf en co o CM r- co 0 m 0 CD 1- r- CO 00 o en en m co p •sf co 0 •sf sf d d eo j; m in o cm csi b b CM b b b in f- co 1- 1- CD i~- m 0 m co CT) 00 l'¬ en m en T- T" I-. 01 o •sf m in ~ r- 01 CO eo CD CM en eo en oi p en o * O m •sf 1-. 0 O p p d * oi II >- en o o o en r- -sf p w cm ^ d d b b CM b b b cm -': si- k •sf CO «) eo in M S O m m en en i~- co co 52 t- oo in in r- en CO CO CM CM 00 eo co 0 s co 00 CM T- CM CM sf •sf sf 00 sf t- co cm o g S; CO I-. r- t- co O 00 en 00 -sf sf 0 0 d d cd JI CM CM b csi b b b d d d d ft' CM O O) CM en 1- in T- O) co in in sf o O Tf ^ in K co cm eo eo CM eo o co 00 !-• I- CM in co •s- -sf T- cm m co m r- en LO CM CM en co o T- LO CM 00 co 1-- in co co co M ^~ (M en O o o CD CM I-. 00 S -sf 1- CM 00 co c- 01 m en o cn o O O f o •sf co t- CO T- O O o en in T- o en sf 0 sf ö ° ö csi <£ o en 41 b b OCMOOCMt-OOO b b T- b d d CM O CM 00 CO CT) f m co 01 co CD t- 0 co en I-- sf 1- O sf sf in o o co r-. CO O -sf co 01 in cm co r- CM 00 sf cp CM O cq o CM -sf T- CO O CO CD O eo sf CO CM 1- 00 o p p in t- o sf p p 3 - 2 o co b b en b b p p d d T- d b b cm b b CM ^ odd ^ LU o o CD -1 4-1 4^ 3 o 0) 0) m U. U. I .g __^ Ç- x I "c 0. °- ° 11 5; s > > s. < °°~ o o O) î O M - O S 6 o n q - q ,"r 0) 4) = ^co_ < (I)I- 1- u. o o -+ CD 15 co m r- 00 en •sf eo r- t- en 00 en o m sf o o m o co o o CD CM co in co in CD sf T— r~ eo en en co co I-. co r- m CM sf CD m co r- m o m o CM u_ in sf O CM co o o O O 00 O sf f- o CM eo 00 o oo o CT) to CO od 1-^ CD p cd sf en co CM CM T- b T_ b b T~ d ed sf O p p p o o o p p p b c co eo r— en CM (D CO CM b b b b b b b b b -*—«o c: II CM t— eo •sf eo CM in CM eo eo o o o o o m o in in CD eo co co sf 00 co 00 I-- co co CM •sf CO sf in CO LU eo o O) CM o o 00 •sf en CM CM in in co t— •sf CO o o o I-- o O sf eo o LO Mf CO S) Z o p d ^~ b b eo (0 o o o p p O o o 4 c: 111 b CM b CM cri b b b b o p p p b b b 0) ai co in CD X CM b o b b b b b b b b b * C 4J4 cd g o *4 o g) «4 CD or '55 ra ro a. CO «4 >• ra CO "o Q. o o c -Q O CD CM CM CO co sf en sf CM I-- T— co r-. 5 o UJ z o en o 00 m m sf o m -5 o in m o I-- co o CO m co o r- co co sf CM r-— m CM ^ eo in ^— LO CO co co co r— co CM sf CM co. T— p p ^ in o o o O o in sf co £ p CM d co O O o U b "*-* T_ p o o o o p p p o" t CO m b b T- b b '- b b o ö b co CM b b d d b b b b b b b E .g CO CO •sf en sf o CM co CM CM o sf r— co sf T— CM T- in CM eo 00 CD CD co CM co en oo CM O O o o o O o Tf •sf , c sf co ^- co T- •sf çn •sf S O o •sf •sf o eo eo "sj-t., LU O p o o o o o .o cri iri ci cn ra o o o O o CO b b b b c d 1-^ r-^ b P p p p P ö b 0) m CM O) sz CM < o CM b d b b b b b b a éô -J *4 o CJ eo t o o If CD eo < c CM c I-- eo co sf O) co T— o en in o o o O o o o in < o 5 sf CM o r- o o o o en CT) CD I-. o o CM o o o o I-. CO co ^f m 0- s CM en co sf in 2 ^ O o p p p o o m o ro o m o O o o o o o co co CT) Lri en- o o o o o "S S? m b d b b b b d i-^ !-: d o p p p p p p p d b m CM en CM CD m CM b b d d b b b b b b E SZ co co i- co r~- in z r— in CM O o o o o o CM 00 o sf f- m eo CD O o o m O o sf co •sf •sf in T— •sf o o o in m o o o o O) CD CT) o Ol r- 2 O o p p o o O o p d O O o o o o o m o o o s cri od cd X O O LO O •sf o o o o p p co m b m d CD b b b d b d o p p p p co CM CM O co b CM d d d b b b b b d b CD o LU s— 0- 1- CO o "D £ o C S O ro Ü 0. UJ CM •sf eo CM eo r- co CM en CM _l en 00 O o o o o CM 00 sf T- CM o CD o CD o sf co CM O CD sf CM o eo o I-- o o o •sf S o en o L? 2 T— o T- o p p p O o p d f- o t- o T— o o o o en o o o sf LU 00 00 sf o CD o Tf o o o o o p m b d S^ b b b d b d o p p p p b b O Z eo CM CM " csi b CM d d d b b b b b -32 E CD LU co > 1- Z LU CO o 10 CL) c o 4* LU Q. +4 O -S a. g d d O O O ra '35 o -Q 0. E d O o CM 4-1 ro co CM d CM O Ol O o t- c Ol ra ra Q. LU n d i? a> c ra ra O o o 0) '5 0- co CÖ H < o u. u. 2 iE 2 Ü Z X X 1- «: iô H < O u. U. 2 Z 2 U Z * X N UJ "O CD in 00 CM co sf CM i— in CD O o m o o C0 I-. r-. c r- O CM •sf co O CM o in o en in LO sf co o CO o o CM en LU 00 o eo CO 00 co m 1- CM •sf O o o p 00 o CM o sf - o o o o r— ^> co CD Z p p p •sf en- CM r- 4^4 O CM c t-^ CM co O ci o p p p O LU CO b d b b CM d b b b b en p p p p b b b CD CD § X m CD b b b b b b b d d b co CO o Ë 4^ o Ö) o *4^ cc '55 ra co ro a 10 > .c ra lo c 0. o o c c O "5 T- CM v- co m o T— co T— 00 CD o o o I-. in CM a Ul z en CM CM co co ra co CD O o CD o o o o en t— co o CM o CD CM sf co T— o o o 00 en 00 _J CD CO a> O 00 CO eo c- CD g .co o en o sf p o p o b o sf o o o o o in •sf CM o csi c •*- b sf o O c: o o o o o -£ in b b d b T- d b b b b o p p p p p b b b O CM b b b b b b b d d b co CD en CD oo sf in r- CD CM CM CM 00 co CD o o o m r— sf co O) •sf CD sf co Ô eo eo CM co I-. CD CD I-. o o I-. CD o m o o co CD m OO co •sf CO • » c: CM o o en CM o o o O r— CD •s*. p o o p O o .o o iri i-^ cd iri u o o o o O O LU cd b b b d p p p p p p CD C co sf b CM b d b b en CM b b 4-c 1- i^ b sf b b b b cd b d b b 55 O Ë < o O_J O o > 0. 'ëû '5 CM If < 2 tr O r— co CM r- LO I-. o en CO C- eo en o o o o in UJ (0 eo en in CD •sf CD CD co co •sf CM in co r- ^x CM o ro sf o eo sf o m o en co m o CO T— o o CM p CM < •sf o in CM o o o CO o 00 m Tf CD p co CM b sf b b b d CO CD b b T— b + co o p p O o o co. O o' Q- m en i-^ CD CM CO b •sf b b b b CM d b b z eo eo z CD CM en I-. M— r- r- cn CM CM CD m CD o o o CT) sf CM 1 O CM r— o 00 o o o o co en oo oo co O) •sf CM 00 CM o o o co r- CM rci O il eo O O p CO p m p p O o o O o d r- T- o CM CM O o o o CO s O cri ci CD ro c CM o p p p O o o co d b cd b b b b b b b b o .c p p P p C CO co CM csi b b b O r- eo u CM b b b b b b b b b O CD LU eo 0- Z sf e £ CM O a< a O c sf O co I-. •sf CM in CO sf en eo -1 en CM CM ro o •sf f- m 00 •sf co CM 1- CO m ro r- 00 eo o M— CM CM O o m sf o m o o 00 L_ CM O in 4-> T- T— T— O 2 o co p r-: T— r- O o o p o en CM 00 o sf o o o o oï CD cri iri p en- ro en en 00 CO _ÇD LU d b eo b b b b b b b b o o p O o p O o o o o Z co CM co en CM csi b b p "o oo b b b b b b d d b b O b o -3? S o CD UJ o. CO > Is- I- r- 0) Z ra 44 LU (0 CO CD 3 o ï o c ra CD LU CM co a m a. a o cc CM CM O o ra Q. O O O O CM 4-c co CM o S D E q II ro o tM O O c O) (S < ."3 LU ra O q CD CD c ra es CM O o o CD n C Q. (0 (Ö p < CL U- U. 2 2 U z * (0 Ü U 1- «: (Ö H < 0. u. u. 2 2 O Z * co Ü (S 0. < X co CD fi r-i-ocofincMcninocMcMO t- m -sf in co rn tn 00 eo 00 sf eo co en t- c 3 x sf 00 0 O 0 COO-sfCMTfl-.CMT-COCDCOOO f d) T3 n Cl en r- in in eo 0 •. eo 0 O OOO'sfCM-c-CO-c-OCOCDCMOO co sf eo CM eo en ° O O CM c- 0 0 O enoooocMoppppoo .- c Q m 0 0 0 0 O O CM en bob UJ Q. C •r^ööööööciciciciöci CO C~ CO D 5 co r« co *C x:ü CM C Q_ co m co CM COOr-O COCMOCOinCMCDO O 0 eo 0 co O 00 sf I-. O m •sf O m O COT-OCMCOf-CMO CD < o < co 0 0 0 0 0 O sf CDOCOO 1OOOOOCMCOO co r- = O •sf 00 co •sf OOOt-O ci- CO 0 CM 0 0 O O O CM 0 0 en OOOp-t-;l--00 b b sf eo b iri b bbbbcdbbb x uj Mr m T) c- CO H or^-fm-sfv-enoLneoococM en c- en LU CD 00 in eo r- 0 eo co O CM CM CMr-T-eneoooooor^-Loo-r-T- co m T- eo Z f- CT) 0 •sf in co CD sf CM en CM m CM O O CMCMCO-sf-c-'sfcOOCMCOOOO 0 0 ^ ra V a. f- O eo CO iL c > < O T— *"- CM 0 O0 0 eo O O O Ol oo^-co-s-^oopcopoo bob 0 eo cob-c^boddddcModd — CM CO O < 3 (0 T- CM C0 r>ci C; CD CO r- CD - •sf sfCMCOOOI-.00 O CO g U. co CO 00 00 00 co sf CM O cococoo 1-. en O en ni CO O ocoh-OCT)T-r— in CM 1 C O 00 sf 0 O O enooooeoco iOt-oo s p § co Ü O *- O 0 c 0 O O c eu ppOO -c- O CM O CO en ppppr-;ppt- p ? ? m co n d d d d d bcN'bb à 7 n- AlmandineGrossular co < Andradite cm CD CI 00 CM-s-r-ooir-r- i— co 0 S S CM en 01 - 00 00 sf O cnt-ino en CM I-- 0 O eo •sf ro CM sfininooT-cn £! 0 i-. i 1 ° en sf O 00 000 Ol 0 O 0 O O OOOT-OT-1-.-sf O O P ~i °° 00 T— •sf co "*"- *"" c c oioaioi-;pp e eo 0 CM O 0 0 CO O O 01 pppo p P P -ri cM'b^bddd bcsibb 0 0 o co 1-- en r- en cm en CD CM CM en 0 CMincno 0 0 T- 01 O eo •sf O m co co en - f- 0 t- lolot-o 0 I-- cri "0.567 E eo O 0 cri O O) O O 00 oioin-c-eooo ior-oo iO *0.375 H— sf en c CO c 1- r- 0 Piemontite 0 O 0 O 0 O O O poppppp popp "*0.002 co rM en -+—j CM cm c\i b b cm b b -^ O 1- d b b b -§ e) co u ZoisiteEpidote .0 co Q_ 111 Ol (O S (D c- O ID r- CO CO CM O CT m eo l'¬ CD *- * eo eo 00 O co O sf en r- 0 f cm 0 0 ** eo O en cmoooo "0.592 eo tr CM eo O 0 co T~ O O co co 0 f-OfT-CT>00 iT-OOOO ,0 *0.343 CD eo l'¬ *** O CD sf en c c *- enoeooinoo 0 0 0 O 0 O O O opoo *"0.002 .^ co CM en Q CM csi b csi b b b d d cm d d ^ E O0 -^ Cil CD eo 00T-cOT-inr-.cn eoocoin 0 t- CD CO v- 00 CM en in O sf m CM sf O O in CM co 0 sf m f- 0 f ocMcnin 0 -c in 00 cri m cri co 0 en * 0 10 S •sf O 00 00 sf eo sf O co oineoo-c-cMin ,T-eninoo co in LO ,o 1— co 0 co CD LU CM CM CM in c O CM CM c CM coocMoeor-o ppcoo 0 c- F _l •sf 0 O O 0 en d d r O codcModdd cd-c^bb cm bob E 03 £ T T) co U. CD .CD incMinoo H—( CD CM CM CD CD r-fOf01OCM 0 CO CO CM O t- CT) CO 0 CO CM CO CM CO r-eocMcoen-sfoo ooenenoo 0 in in cd 00 -c- en Cl s* < m eo 1— co O in eoT— eo in 1C CO 0 m m sf O cocDTfor— OLO lOcoor-. iO i-- cm in -sf CD co in CM l'¬ p p C") c c CM O sf 0 JI 0 CM co 0 CM O en TfOOOCMCOp ppCÛO d d t^ O) 2 eo b cm d d d d cd-^bb csi odd en -g m 0CO 00 en 1 Cll ^ E (Fetot) Cll c- _i r? 0. LU ra -c- O O Fe3+/Fetot -0 E co O 0 O q 4-» Z d d q q C 0 o> O O ^ co ra xMg cn < n d 0 ._ ._ — u cd cd — .2* Na(M4) AI(IV)AI(VI)Na(A) f~ £ (0 2 (0 1- < 0 u. u. 2 Z 2 U Z * (0 X H (0H References Alexander, V.D., Griffin, D.T. and Martin, T.J. (1986) Crystal chemistry of some Fe- and Ti-poor dumortierites. Am Mineral., 71, 786-794. Anonymous. (1972) Ophiolites: Geotimes, Dec. 1972.24-25. Apted, M. and Liou, J. (1983) Phase relation among greenschist epidote-amphibolite and amphibolite in a basaltic system. Am. J. Sei. 283-A, 328-354. Baudin, T. and Marquer, D. (1993) Metamorphism and deformation in the Tambo nappe (Swiss Central Alps); evolution of the phengite substitution during Alpine deformation. Schweiz. Mineral. Petrogr. Mitt, 73, 285-299. Baudin, T., and Marquer, D. (1993) Metamorphism and deformation in the Tambo nappe (Swiss Central Alps); evolution of the phengite substitution during Alpine deformation. Schweiz. Mineral. Petrogr. Mitt, 73, 285-299. Bedini, R.M. and Bodinier, J. (1999) Distribution of incompatible trace elements between the constituents of spinel peridotite xenoliths: ICP-MS data from East Africa Rift. Geochem. Cosmochem. Acta 63 (22); 3883-3900. Bedini, R.M., Bodinier, J.-L., Dautria, J.-M. and Morten, L. (1997) Evolution of LILE-enriched small melt fractions in the lithospheric mantle: a case study from the East African Rift. Earth Planet. Sei. Lett. 153; 67-83. Belluso, E., Ruffini, R., Schaller, M. and Villa, I.M.. (2000) Electron-microprobe and Ar isotope characterization of chemically heterogeneous amphiboles from the Palala Shear Zone, Limpopo Belt, South Africa. Eur. J. Mineral. 12,45-62. Bernoulli, D., Giger, M., Müller, D.W. and Ziegler, U.R.F. (1993) Sr-isotope.stratigraphy of the Gonfolite lombarda Group ("South-Alpine Molasse", northern Italy) and radiometric constraints for its age distribution. Eclogae Geol. Helv. 86/3,751-767. Bertrand, E. (1880) Sur un minéral bleu de Chaponost, près Lyon. Sur un autre minéral bleu du Chili. Bull. Soc. Min. France, 3,171-172. Beukes, G.J., Slabbert, M.J., de Bruiyn, H., Botha, B.J.V., Schoch, A.E. and van der Westhuizen, W.A. (1987) Ti-dumortierite from the Keimoes area, Namaqua mobile belt, South Africa. N.Jb.Miner.Abh., 157(3), 303-318. Bloodaxe, E., Hughes, J., Dyar, M., Grew, E. and Guidotti, C. (1999) Linking structure and chemistry in the Schorl-Dravite series. Am. Mineral., 84,922-928. Blusztajn, J. and Shimizu, N. (1994) The trace-element variations in clinopyroxenes from spinel peridotite xenoliths from southwest Poland. Chem. Geol., 111,227-243. Bodinier, J.L. (1988) Geochemistry and petrogenesis of the Lanzo peridotite body, Western Alps. Tectonophysics, 149, 67-88. Bodinier, J.L., Dupuy, C. and Dostal, J. (1988) Geochemistry and pertogenesis of Eastern Pyrenean peridotites. Geochem. Cosmochem. Acta, 52,2892-2907. Boillot, G., Agrinier, P., Beslier, M.O., Cornen, G., Froitzheim, N., Gardien, V., Girardeau, J., Gil-lbarguchi, J.-L, Kornprobst, J., Moullade, M., Schärer, U. and Vanney, J.-R. (1995) A lithospheric syn-rift shear zone at the ocean-continent transition: preliminary results of the GALINAUTE II cruise (Nautile dives on the Galicia Bank, Spain). CR. Acad Sei Paris, 321, série II a, 1171-1178. Bucher-Nurminen, K. and Droop, G.A. (1983) The metamorphic evolution of garnet-cordierite-sillimanite- gneissesofthe Gruf Complex, eastern Pennine Alps. Contrib. Mineral. Petrol., 84,215-227. Cannât, M., Karson, J.A. and Miller, D.J.E. (1995) Proc. ODP, Init. Repts. 153, Site 920 (Mid-Atlantic Ridge): College Station. TX (Ocean Drilling Program). Chazot, G., Menzies, M. and Harte, B. (1996) Determination of partition coefficients between apatite, clinopyroxene, amphibole, and melt in natural spinel Iherzolites from Yemen: Implications for wet melting of lithospheric mantle. Geochim. Cosmochim. Acta, 60(3), 423-437. Chopin, C, Ferraris, G., Ivaldi, G., Schertl, H.P., Schreyer, W., Compagnoni, R., Davidson, C. and Davis, A.M. (1995) Magnesiumdumortierite, a new mineral from very-high-pressure rocks (western Alps) II. Crystal chemistry and petrological significance. Eur. Mineral, 7, 525-535. 148 References Claringbull, G.H. and Hey, M.H. (1958) New data from dumortierite. Mineral. Mag. 31,901-907. Coleman, R. (1966) New Zealand serpentinites and associated metasomatic rocks. New Zealand Geol. Survey Bull., 76,1-102. Connolly, J. (1990) Calculation of multivariable phase diagrams: an algorithm based on generalized thermodynamics. Am. J. Sei., 290,666-718. Cornelius, H.P. (1916) Ein alpines Vorkommen von Sapphirine. Cent. Mineral. 9-265. Davidson, C, Rosenberg, C. and Schmid, S.M. (1996) Synmagmatic folding of the base of the Bergell pluton, Central Alps. Tectonophysics, 265,213-238. Deer, W.A., Howie, R.A. and Zussman, J. (1997) Rock-forming minerals, Volume 2B, Double-chain silicates. The geological Society, London, p. 764. Deutsch, A. and Steiger, R.H. (1985) Hornblende K-Ar ages and climax of the Tertiary metamorphism in the Lepontine Alps (south-central Switzerland): an old problem reassessed. Earth. Planet. Sei. Lett., 72,175-189. Dick, H.J.B. (1989) Abyssal peridotites, very slow spreading ridges and ocean ridge magmatism. Geological Soc. Sp. Pub. 71-105 p. Dick, H.J.B., Fisher, R.L. and Bryan, W.B. (1984) Mineralogie variability of the uppermost mantle along mid-ocean ridges. Earth Planet. Science L., 69,88-106. Droop, G. and Bucher-Nurminen, K. (1984) Reaction textures and metamorphic evolution in sapphirin- bearing granulites of the Gruf Complex in Italian Central Alps. J. Petrol., 25,766-803. Dürr, S.B., Ring, U. and Frisch, W. (1993) Geochemestry and geodynamic significance of Northern Penninic ophiolites from the Central Alps. Schweiz. Mineral. Petrogr. Mitt., 73,407-419. Eggins, S., Rudnick, R. and McDonough, W. (1998) The composition of peridotites and their minerals: a laser-ablation ICP-MS study. Earth Planet. Sei. Lett., 154, 53-71. Engel, C, and Fisher, R. (1975) Granitic to ultramafic rock complexes of the Indian ocean ridge system, western India Ocean. Geol. Soc. Am. Bull., 86,1553-1578. Engi, M., Clifford, ST. and Schmatz, D.R. (1995) Tertiary metamorphic conditions in the eastern Lepontine Alps. Schweiz. Mineral. Petrogr. Mitt., 75,347-369. Ertl, A., Pertlik, F. and Bernhardt, H. (1997) Investigation on onelite with excess boron from the Koralpe, Styria, Austria. Oster. Akad. Wiss. Math. Naturw. Kl. Anzeiger Abt., I (134), 3-10. Evans, B. W. (1982) Amphiboles in metamorphosed ultramafic rocks. Reviews in mineralogy, 9B, 98-113. Evans, B.W. and Ghiorso, M.S. (1995) Thermodynamics and petrology of cummingtonite. Amer. Mineral. 80, 649-663. Evans, B.W., Ghiorso, M.S., Yang, H. and Medenbach, O. (in press.) Thermodynamics of amphiboles: Anthophyllite and ortho-clino phase loop Evans, B.W. and Medenbach, O. (1997) The optical properties of cummingtonite and their dependence on Fe-Mg order-disorder. Eur. J. Mineral., 9,993-1003. Evans, B.W. and Frost, B.R. (1975) Chrome-spinel in progressive metamorphism - a preliminary analysis. Geoch. Cos. Acta, 39,959-972. Fawcett, J.J and Yoder, H.S. (1966) Phase relationships of chlorites in the system MgO-Ab03-Si02-H20. Amer. Mineral. 51,353-380. Ferry, J. (1984) Phase composition as a measure of reaction progress and an experimental model for the high-temperature metamorphism of mafic igneous rocks. Amer. Mineral., 69,677-691. Florineth, D. and Froitzheim, N. (1994) Transition from continental to oceanic basement in the Tasna nappe (Engadine window, Graubünden, Switzerland): evidence for Early Cretaceous opening of the Valais ocean. Schweiz. Mineral. Petrogr. Mitt., 74,437-448. Frey, F.A., John-Suen, C. and Stockman, H.W. (1985) The Ronda high temperature peridotite: Geochemistry and petrogenesis. Geochimica Cosmo. Acta, 49,2469-2491. Frey, F.A., Shimizu, N., Leinbach, A., Obata, M. and Takazawa, E. (1991) Compositional variations within the lower layered zone of Horoman peridotite, Hokkaido Japan: constraints on models for melt- solid segregation. J. Petrol.(Special Lherzolites lusse), 211-227. Froitzheim, N., Schmid, S.M. and Frey, M. (1996) Mesozoic paleogeography and timing of eclogite-facies metamorphism in the Alps: A working hypothesis. Eclogea geol. Helv., 89/1,81-110. 149 References - -- i I I .III ««» Frost, Ca., and Frost, B. (1989) Sr and Nd isotopic effects on C02-streaming in dehydrated amphibolites, Morton Pass, Wyoming. Geol. Soc. Amer. Abstract Programs, 21, A85-A86. Gillis, K., Mével, C, and Allan, J.E. (1993) Proc. ODP, Init. Reports, 147, Site 895, Hess Deep: College Station. TX (Ocean Drilling Program). Goldsmith, J. and Newton, R. (1977) Scapolite-plagioclase stability relations at high pressures and temperatures in the system NaAISi308-CaAI2Si208-CaS04. Amer. Mineral., 62,1063-1081. Gordon, T.a., and Greenwood, H. (1971) The stability of Grossular in H2O-CO2 mixtures. Amer. Mineral., 56,1674-1688. Greenwood, H. (1963) Synthesis and stability of anthophyllite. J. Petrol., 4,317-351. Greenwood, H. (1967) Wollastonite: Stability in H2O-CO2 mixtures and occurrence in contact- metamorphic aureole near Samo, British Columbia. Amer. Mineral., 52,1669-1680. Gulson, B.L. (1973) Age relations in the Bergell region of the South-East Swiss Alps: With some geochemical comparisons. Eclogae. Geol. Helv., 66,293-313. Günther, D., Frischknecht, R., Müschenborn, H.-J. and Heinrich, CA. (1997) Direct liquid ablation: a new calibration strategy for laser ablation-ICP-MS microanalysis of solids and liquids. Fresenius J. Anal. Chem. 359, 390-393. Günther, D., Frischknecht, R., Heinrich, CA. and Kahlert, H.-J. (1997) Capabilities of an Argon fluoride 193 nm excimer laser for laser ablation inductively coupled plasma mass spectrometry microanalysis of geological materialst. J. Anal. At. Spectr. 12, 939-944. Hammarstrom, J.M.a., and Zen, E. (1986) Aluminium in hornblende: An empirical igneous geobarometer. American Mineral., 71,1297-1313. Hansmann, W. (1996) Age determinations on the Tertiary Masino-Bregaglia (Bergell) intrusives (Italy, Switzerland): a review. Schweiz. Mineral. Petrogr. Mitt., 76,3,421-453. Heaman and Parrish (1988) In Geochronology and thermochronology by the 39Ar/40Ar method. Ed. McDougall et al. Oxford University, Oxford, p. 272 Hébert, R., Adamson, A. and Komor, S. (1990) Metamorphic petrology of ODP Leg 109, Hole 670A serpentinized peridotites: serpentinization processes at a slow spreading ridge enviroment. Proceedings of the Ocean Drilling Program, Scientific results, 106/109,103-115. Hekinian, R., Bideau, D., Francheteau, J., Cheminée, J., Armijo, R., Londsdale, P. and Blum, N. (1993) Petrology of the East Pacific Rise crust and upper mantle exposed in Hess Deep (Eastern Equatorial Pacific). Journal of Geophysical Research, 98 B5,8069-8094. Hermann, J. and Müntener, O. (1996) Exhumation-related structures in the Malenco-Margna-System: implications for paleogeography and consequences for rifting and Alpine tectonics. Schweiz. Mineral. Petrogr. Mitt., 76,501-519. Hermann, J. (1997) The Braccia gabbro (Malenco, Alps): Permian intrusion at the crust to mantle interface and Jurassic exhumation during rifting. Institut für Mineralogie und Pétrographie. 193. ETH- Zürich, Switzerland. Hofmann, A.W. (1988) Chemical differentiation of the Earth: the relationship between the mantle, continental crust and oceanic crust. Earth Planet. Science Lett., 90,297-314. Holland, T.a., and Powell, R. (1998) An internally consistent thermodynamic data set for phases of petrological interest. J. Metamorphic Geology, 16, 309-343. Huber, R.K. (1999) Tectonometamorphic evolution of the eastern Pennine Alps during Tertiary continental collision: Structural and petrological relationships between Suretta, Tambo, Chiavenna and Gruf units (Switzerland/Italy). Institut de Géologie. 125. Université de Neuchâtel, Switzerland. Huber, R.K. and Marquer, D. (1996) Tertiary deformation and kinematics of the southern part of the Tambo and Suretta nappes (Val Bregaglia, Eastern Swiss Alps). Schweiz. Mineral. Petrogr. Mitt., 76, 383-397. Huber, R.K. and Marquer, D. (1998) The tectonometamorphic history of the peridotitic Chiavenna unit from Mesozoic to Tertiary tectonics: a restoration controlled by melt polarity indicators (Eastern Swiss Alps). Tectonophysics, 296, 205-223. lonov, D.A., Savoyant, L. and Dupuy, C. (1992) Application of the ICP-MS technique to trace element analysis of peridotites and their minerals. Geostandards Newsletter, 16/ 2,311-315. 150 References Irvine, T. and Baragar, W. (1971) A guide to the chemical classification of the common calc-alcaline rocks. Can. J. Earth Sei., 8,523-548. Jackson, S. (1996) computer program Lamtrace 1.52. Jagoutz, E., Palme, H., Baddenhausen, H., Blum, K., Cendales, M., Dreibus, G„ Spettel, B., Lorenz, V. and Wanke, H. (1979) The abundances of major, minor and trace elements in the earth's mantle as derived from primitive ultramafic nodules. Proc. Lunar Planet. Sei. Conf 10th, 2031-2050. Jenkins. (1980) in Metamorphose Script. ETH-Zürich, Switzerland Johnson, K.TM., Dick, H.J.B.a., and Shimizu, N. (1990) Melting in the oceanic upper mantle: An ion microprobe study of diopsides in abyssal peridotites. J. Geophy. Research, 95, NO. B3., 2661- 2678. Kimball, K.L., Spear, F.S. and Dick, H.J.B. (1985) High temperature alteration of abyssal ultramafics from the Islas Orcadas fracture zone, South Atlantic. Contrib. Mineral. Petrol, 91, 307-320. Koppel, V. and Grünenfelder, M. (1975) Concordant U-Pb ages of monazite and xenotime from the Central Alps and the timing of the high temperature Alpine metamorphism, a preliminary report. Schweiz. Mineral. Petrogr. Mitt, 55,129-132. Kretz (1983) Amer. Mineral. 68,277-279 Kuno, H. (1968) Differentiation of basalt magma. In H.P. Hess, A. Basalts: the Poldervaart treatise on rocks of basaltic composition, II. 623-688. Wiley & Sons, New York. Laird, J. and Albee, A. (1981) Pressure-temperature and time indicators in mafic schist: their application to reconstructing the polymetamorphic history of Vermont. Am. J. Sei, 281,127-175. Leake, B. (1978) Nomenclature of amphiboles. Am. Mineral, 63,1023-1052. Lemoine, M, Tricart, P. and Boillot, G. (1987) Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): In search of genetic model. Geology, 15(622-625). Luth, W, Jahns, R. and Tuttle, O. (1964) The granitic system at pressure of 4 kilobars. J. Geophys. Res, 69, 759-773. Marquer, D. (1991) Structures et cinématique des déformations alpines dans le granite de Truzzo (Nappe de Tambo: Alpes centrales suisses). Eclogea Geol. Helv, 84(1), 107-123. Marquer, D, Baudin, T, Peucat, J, and Persoz, F. (1994) Rb-Sr mica ages in the Alpine shear zones of the Truzzo granite: timing of the Tertiary alpine P-T-deformations in the Tambo nappe (Central Alps, Switzerland). Eclogea Geol. Helv, 85(3), 1-61. Marquer, D, Challandes, N. and Baudin, T. (1996) Shear zone patterns and strain partitioning at the scale of Penninic nappe: the Suretta nappe (Eastern Swiss Alps). J. Struct. Geol, 18(6), 753-764. McDonough, W.F. (1994) Chemical and isotopic systematics of continental lithospheric mantle. In H.O.A.L. Meyer, O.H, Ed. Fifth International Kimberlite Conference, 1, p. pp. 478-485. CPRM, Rio de Janeiro, Brasil. McDonough, W.F. and Frey, F.A. (1989) Rare earth elements in the upper mantle. In B.R.M. Lipin, G.A. Rare earth elements, Reviewers in Mineralogy Ud. 21,99-145. Mineral. Soc. America. McDonough, W.F.a, and Sun, S.S. (1995) The composition of the Earth. Chemical Geology, 120, 223- 253. Menzies, M.A, Long, A, Ingram, G, Tatnell, M. and Janecky, D. (1993) MORB peridotite-sea water interaction: experimental constraints on the behaviour of trace elements, 87Sr/86Sr and 143Nd/144Nd ratios. Magmatic Processes and Plate Tectonics, Geological Soci. Special Pub, 76,309-322. Milnes, A. and Pfiffner, A. (1980) Tectonic evolution of the Central Alps in the cross sectionSt. Gallen- Como. Eclogea Geol. Helv, 73,619-633. Moore, P.B. and Araki, T. (1978) Dumortierite. Si3B[AI675Si025O1725(OH)075]: a detailed structure analysis. N. Jahrb. Mineral. Abh, 132,231-241. Moticska, P. (1970) Pétrographie und Structuranalyse des westlichen Bergeller Massifs und seines Rhamens. Schweiz. Mineral. Petrogr. Mitt. 50, 355-443. Müntener, O. (1997) The Malenco peridotites (Alps): Petrology and geochemestry of subcontinental mantle and Jurassic exhumation during rifting. PhD. Thesis ETH-Zürich p. 205, Zürich, Switzerland. Neuenschwander, S. (1996) Ein Beitrag zur Geologie der Cima Lunga-Einheit, Alpe Gagnone, Tessin. Institut Mineral. Petrogr. unpub. Dipl. UNI/ETH-Zürich, Switzerland. 151 References Newton, R. and Goldsmith, J. (1975) Stability of scapolite meionite (3CaAl2Si202 .CaCÛ3) at high pressure and storage of CO2 in the deep crust. Contrib. Mineral. Petrol, 49,49-62. Norman, M. (1998) Melting and metasomatism in the continental lithosphère: laser ablation ICP-MS analysis of minerals in spinel Iherzolites from eastern Australia. Contrib. Mineral. Petrol, 130, 240-255. O'Reilly, S, Griffin, W. and Ryan, C. (1991) Residence of trace elements in metasomatized spinel Iherzolite xenoliths: a proton-microprobe study. Contrib. Mineral. Petrol, 109,98-113. Oterdoom, W. (1980) Scapolite in metamorphic calc-silicate rocks: crystallographic and phase-relations. PhD. Thesis ETH-Zürich. p. 146, Zürich, Switzerland. Ottonello, G, Ernst, W.G. and Joron, J.L. (1984) Rare Earth and 3d transition element geochemistry of peridotitic rocks: I. peridotites from the Western Alps. J. Petrol, 25,343-372. Pearce, J.A. and Cann, J.R. (1973) Tectonic setting of basic volcanic rocks determinated using trace element analyses. Earth Planet Sei. Lett, 19,290-300. Peccerillo, A. and Taylor, S. (1976) Geochemistry of calc-alkaline volcanic rocks from the Kastamonu Area, Northern Turkey. Contr Mineral Petrol, 30,63-81. Pfiffner, M. (1999) Genese der Hochdruckmetamorphrn ozeanische Abfolge der Cima Lunga-Einheit (Zentralalpen). PhD. Thesis ETH-Zürich p. 248, Zürich, Switzerland. Platt, J, Behrmann, J, Cunningham, P, Dewey, J, Helman, M, parish, M, Shepley, M, Wallis, S. and Weston, P. (1989) Kinematics of the Alpine arc and the motion history of Adria. Nature, 337,158- 161. Pognante, U, Lombardo, B, and Venturelli, G. (1982) Petrology and Geochemestry of the Fe-Ti gabbros and Plagiogranites from the Western Alps ophiolites. Schweiz. Mineral. Petrogr. Mitt, 62, 457- 472. Purdy, J.W.a, and Jäger, E. (1976) K-Ar ages on rock forming minerals from the Central Alps. Mem. Ist. Geol. Min. Univ. Padova, 30,1-30. Puschnig, A. (1998) Th Forno unit (Rhetic Alps): Evolution of an ocean floor sequence from rifting to Alpine orogeny. Institut Mineral, und Petrogr. ETH, Zürich Switzerland. Rampone, E. (in prep.) Rampone, E, Hofman, A.W. and Raczek, I. (1998) Isotopic contrasts within the Internal Liguride ophiolite (N. Italy) the lack of a genetic mantle-crust link. Earth and Planetary Science Letters, 163, 175- 189. Rampone, E, Hofmann, A.W, Piccardo, G.B, Vannucci, R, Bottazzi, P. and Ottolini, L. (1996) Trace element and isotope geochemistry of depleted peridotites from an N-MORB type ophiolite (Internal Liguride, N. Italy). Contrib. Mineral. Petrol, 123,61-76. Rampone, E, Hofmann, A.W, Piccardo, G.B, Vanucci, R, Bottazzi, P. and Ottolini, L. (1995) Petrology, Mineral and Isotope Geochemistry of the External Liguride Peridotites (Northern Apennines, Italy). J. Petrol, 36/1,81-105. Rampone, E. and Morten, L. (2001) Records of crustal metasomatism in the garnet peridotites of the Ulten Zone (Upper Austroalpine, Eastern Alps) J. Petrol, 42/1,000-000. Reusser, E. (1987) Phasenbeziehungen im Tonalit der Bergeller Intrusion. PhD. Thesis ETH-Zürich p. 220, Zürich, Switzerland. Rice, M. (1983) metamorphism of rodingites: Part I. Phase relations in a propotion of the system CaO- MgO-AI203-Si02-C02-H20. Am. J. Sei, 283(A), 121-150. Ringwood, A.E. (1975) Composition and petrology of the Earth's Mantle. 618 p, New York. Rollinson, H. (1993) Using geochemical data: evaluation, presentation, interpretation. 352 p. Russ-Nabelek, C. (1989) Isochemical contact metamorphism of mafic schist, Laramie Anorthosite Complex Woyoming: Amphibole compositions and reactions. Ame. Mineral, 74,530-548. Sawyer, D.S., Withmarsh, R.B, and Klaus, A.E. (1994b) Proc. ODP, Init Reports, 149 Iberia Abyssal Plain: College Station, TX (Ovean Drilling Program). Scambelluri, M, Müntener, O, Hermann, J, Piccardo, G.B. and Trommsdorff, V. (1995) Subduction of water into the mantle: History of an Alpine peridotite. Geology 23,459-462. Schmid, S.M, Berger, A, Davidson, C, Gieré, R, Hermann, J, Nievergelt, P., Puschnig, A. and Rosenberg, C (1996) The Bergell pluton (Southern Switzerland, Northern Italy): overview 152 References accompanying a geological- tectonic map of the intrusion and surrounding country rocks. Schweiz. Mineral. Petrol. Mitt, 76, 329-355. Schmid, S.M., Pfiffner, O.A. and Schreurs, G. (1997) Rifting and collision in the Penninic zone of eastern Switzerland. In O.A.a. Pfiffner. Deep structure of the Swiss Alps: results of NRP 20. Schmid, S.M., Pfiffner, O.A., Froitzheim, N, Schönborn, G. and E, K. (1996b) Geophysical-geological transect and tectonic evolution of the Swiss-Italian Alps. Tectonics, 15,1036-1064. Schmidt, M. (1992) Amphibole composition in tonalité as a function of pressure: an experimental calibration of the Al-in-homblende barometer. Contrib. Mineral. Petrol, 110,304-310. Schmutz, H. (1976) Der Mafitit-Ultramafitit Komplex zwischen Chiavenna und Bondasca. Beitr. Geol. Karte Schweiz, 149,1-73. Schreyer, W, Wodara, U, Marler, B, von Aken, P, Seifert, F. and Robert, J. (2000) Synthetic tourmaline (olenite) with excess boron replacing silicon in the tetrahedral site: I. Synthesis conditions, chemical and spectroscopic evidence. Eur. J. Mineral, 12,529-541. Seyfried, J.R. and Dibble, W.E. (1980) Seawater-peridotite interaction at 300°c and 500 bars: implications for the origin of oceanic serpentinites. Geochim. Cosmochim. Acta. 44,309-321. Shervais, J.W. (1982) Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth Planet Sei. Lett, 101-118. Spear, F.S. (1981) An experimental study of hornblende stability and compositional variability in amphibolite. American Journal of Science, 281,697-734. Stampfli, G. (1993) Le Briançonnais, terrain exotique dans les Alpes? Eclogea Geol. Helv, 86,1-45. Sun, S.-s. and McDonough, W.F. (1989) Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. Geol. Society Sp. Publ, 42, 313-345. Thompson, J.J, Laird, J.a, and Thompson, A. (1982) Reactions in amphibolite, greenschist and blueschist. J. Petrol, 23,1-27. Tiepolo, M, Tribuzio, R. and Vannucci, R. (1997) Mg- and Fe-gabbroids from northern Apennine ophiolites: parental liquids and igneous differentiation process. Ofioliti, 22 (1), 57-69. Tracy, R. and Frost, B. (1991) Phase equilibria and thermobarometry of calcareous, ultramafic and mafic rocks, and iron formations. In R.i. Mineralogy, Contact metamorphism, 26.207-280. Trommsdorff, V. and Connolly, J. (1996) The ultramafic contact aureole about the Bregaglia (Bergell) tonalité: isograds and thermal model. Schweiz. Mineral. Petrogr. Mitt, 76,537-547. Trommsdorff, V. and Evans, B.W. (1972) Progressive metamorphism of antigorite schist in the Bergell tonalité aureola (Italy). Amer. J. Sei, 272,423-437. Trommsdorff, V. and Evans, B.W. (1974) Alpine metamorphism of peridotitic rocks. Schweiz. Mineral. Petrogr. Mitt. 333-353. Trommsdorff, V. and Nievergelt, P. (1983) The Bregaglia (Bergell): lorio intrusive and its field relations. Mem. Soc. Geol. Ital, 26,55-68. Trommsdorff, V, Piccardo, G.B. and Montrasio, A. (1993) From magmatism through metamorphism to the sea floor emplacement of subcontinental Adria lithosphère during pre-Alpine rifting (Malenco, Italy). Schweiz. Mineral. Petrogr. Mitt, 73,191-203. Trommsdorff, V, Lopez Sanchez-Vizcaino, V, Goméz-Pugnaire, M.T. and Müntener, O. (1998) High pressure breakdown of antigorite to spinifex-textured olivine and orthopyroxene, SE-Spain. Contrib. Mineral. Petrol. 132,139-148. Trümpy, R. (1980) An outline of the geology of Switzerland. In S.G. Komm,. Geology of Switzerland, a Guide-book. 104. Ulmer, P. and Trommsdorff, V. (1999) Phase relations of hydrous mantle subducting to 300 km. In Fei, Y, Betka, CM. and Mysen, B.O. Mantle petrology field observations and high pressure experimentation. Geochem. Soc. Special publication. 259-281, Washington. Vannucci, R, Shimizu, N, Piccardo, G, Ottolini, L. and Bottazzi, P. (1993) Distribution of trace elements during breakdown of mantle garnet: en example from Zabargad. Contrib. Mineral. Petrol, 113, 437-449. Venturelli, G, Thorpe, R.S, and Potts, P.J. (1981) Rare earth and trace element characteristics of ophiolitic metabasalts from the Alpine-Apennine belt. Earth and Planetary Science Letters, 53, 109-123. 153 References Villa, I.M. (1992) Datability of Quaternary volcanic rocks: an 40Ar/39Ar perspective on age conflicts in lavas from the Alban Hills, Italy. Eur. J. Mineral, 4,369-383. Villa, I.M. (1998) Isotopic closure. Terra Nova, 10,42-47. Villa, I.M. and von Blanckenburg, F. (1991) A hornblende 39Ar-40Ar age traverse of the Bregaglia tonalité southeastern Central Alps. Schweiz. Mineral. Petrogr. Mitt, 71. Villa, I.M, Grobéty, B, Kelley, S.P, Trigila, R. and Wieler, R. (1996) Assessing Ar transport paths and mechanisms in the McClure Mountains hornblende. Contrib. Mineral. Petrol, 126,67-80. Villa, I.M, Hermann, J, Müntener, 0. and Trommsdorff, V. (2000) 39Ar-40Ar dating of multiply zoned amphibole generations (Malenco, Italian Alps). In press, von Blanckenburg, F. (1992) Combined high-precision chronometry and geochemical tracing using accessory minerals: applied to the Central-Alpine Bergell intrusion (central Europe). Chem. Geol, 100,19-40. Walker, C. (1991) North Atlantic Oceanic crust and Icland. In P. Floyd, Oceanic basalts. 311-351, van Noustrand, R. New York. Wenk, H. (1973) The structure of the Bergell Alps. Part II. Eclogea Geol. Helv, 66(2), 255-291. Wenk, H.R, Wenk, E.A. and Wallace, J.H. (1974) Metamorphic assemblages in pelitic rocks of the Bergell Alps. Schweiz. Mineral. Petrogr. Mitt, 54, 507-554. Werding, G. and Schreyer, W. (1996) Experimental studies on boronsilicates and selected borates. In Am. Mineral. Soc. Boron. Mineralogy and geochemistry, 33,117-163. Werding, G. and Schreyer, W. (1983a) High pressure synthesis of dumortierite. Terra cognita (EUG Abstract) 3:165. Werding, G. and Schreyer, W. (1983b) Synthesis, crystal chemistry and preliminary stability of dumortierite in the system Si02-Al203-B203-H20. Fortschr. Min 61 Beih, 1,219-220. Wiedenbeck, M.A. and Baur, H. (1986) Metamorphic mineral assemblages in pelitic rocks of the Bergell Alps. Schweiz. Mineral. Petrogr. Mitt. 54,507-554. RlNGRAZIAMENTI Ringrazio tutti coloro che non sono citati in questa pagina. A questi amici e colleghi va un grazie particolare per aver sostenuto il mio lavoro e soprattutto la mia voglia di continuare in ogni circostanza in questa esperienza non solo scientifica. Ringrazio il mio „Doktorvater" Volkmar Trommsdorff che mi ha offerto l'opportunité di awiare questo studio concedendomi liberté decisionale e fiducia. Le conoscienze di cui ho fatto tesoro in questi anni di permanenza all'ETH non sarebbero state possibili senza il suo intervento. Ringrazio Peter Ulmer per i suggerimenti e discussioni avute ma soprattutto per l'appoggio che nello sviluppô del mio lavoro mi ha dimostrato. Un grazie particolare va a Lauro Morten per aver suggerito la possibilité di awiare un dottorato all'estero e che con la sua presenza anche se a distanza ha seguito i miei studi. Di queste persone ho apprezzatto la competenza scientifica e la sensibilité che al di lé del ruolo rivestito e delle inclinazioni personali hanno dimostrato. Ringrazio Jamie Connolly per tutte quelle volte che con pazienza si è preso il tempo di spiegarmi ed aiutarmi nei meandri dei calcoli termodinamici che costituiscono una parte fondamentale di questo lavoro e soprattutto per quel suo modo rassicurante di dire: „No problem!". Ringrazio Igor Villa per aver messo a disposizione il suo laboratorio nonché le sue conoscienze in materia di dating of metamorphic amphiboles. Inoltre sono riconoscente a Günther Detlef e Thomas Pettke che con grande disponibilité hanno reso possibile le misure con il LA-ICP-MS. Un grazie particolare va a Jörg Hermann ed Othmar Müntener per tutto. Essi hanno arricchito il mio lavoro con discussioni critiche, suggerimenti ed entusiasmo. La stesura di questo manoscritto in lingua inglese non è stata cosa facile per questo sono particolarmente grata a tutti gli amici che leggendolo l'hanno migliorato: grazie a Katja, Trudi, Andy, Guy, Paolo ed Alessio. Le persone che vorrei ancora ringraziare sono tante ma mi limito a ricordare che ho apprezzatto l'aiuto e la disponibilité di tutti che ho incontrato all'IMP e non. Dedico questo lavoro a mia madré Anna e mio padre Salvatore. Curriculum vitae CaterinaTalerico Address Säntis-Strasse 21 CH-8008 Zürich 11.09.1967 born in Luzern (CH) to Anna Veltri and Salvatore Talerico Italian and Swiss citizenship 1974-1980 Primary school in Ebikon (LU) 1980-1983 Secondary school in Ebikon (LU) 1983-1987 High school (Liceo scientifico) in Luino (VA-ltaly) and in San Giovanni in Fiore (CS-ltaly) 1987-1994 Laurea in Scienze Geologiche at the University of Calabria Italy 1994-1995 Assistant position at the University of Calabria Italy 1995-1996 Exchange studentship (l/CH) at ETH Zürich 1996-2000 PhD Student with assistant position at ETH Zürich. Doctoral thesis: "Petrological and chemical investigation of a metamorphosed oceanic crust- mantle section (Chiavenna, Bergell Alps)". Advisors: Prof. Dr. V. Trommsdorff, Dr. P. Ulmer and Prof. Dr. L. Morten.