No description available.

Corundum from the Bo Ploi deposit, Kanchanaburi, Thailand and its associated Tertiary alkali basalt were studied in terms of mineralogy, geochemistry and petrogenesis to ascertain the genetic link between these materials. Gem quality corundum, sapphire variety, is composed of almost pure AI2O3 with a small but significant amount of FeO and TiC > 2. Optical spectroscopic studies reveal that the blue colour is caused by Fe2+ - Ti4+ intervalence charge transfer. The syn-genetic inclusions are mineral inclusions, including, feldspar, zircon, spinel, pyrochlore, boehmite, rutile, monazite, and opaque minerals fluid inclusions and silicate melt inclusions. Microthermometric studies of fluid and silicate-melt inclusions trapped in these sapphires suggest that they grew from silicate-melt under H2O-, COrrich and hypersaline brine conditions at the temperature approximately 800- 1000 C. The silicate-melt inclusions are trachyandesitic compositions indicating that the Bo Ploi corundum is magmatic origin forming from the extreme fractionation of trachyandesitic magma with additional alumina contamination from crustal materials and brought to the surface by later eruption of the Bo Ploi basalt. The sapphires oxygen isotope chemistry, 8180 6.5-9.6%o, suggests a mingling of crustal and mantle oxygen isotope thus confirming the role of crustal contamination. The Bo Ploi Tertiary basalt is a within-plate magma of basanite composition, which differentiated through the fractional crystallisation process without a significant crustal contamination, 5 O 6.0-6.9%o, although numbers of crustal xenoliths including granites, nepheline syenite and quartzite are found in it. The basanites also accommodated mantle xenoliths, mainly spinel lherzolite, which provided P-T scheme based on thermodynamic calculations of coexisting mineral equilibria in the range of 11 to 24 kbar and temperature of 1000 to 1200 C correspond to the depths of 36 to 79 km in the upper mantle.

Measurements of initial susceptibility, remanent magnetisations and hysteresis loops have been carried out at room temperature on a range of characterised iron oxides and sulphides in order to attempt qualitative and quantitative identification of the individual minerals. The minerals studied were magnetite, titanomagnetite, haematite, pyrrhotite and greigite. It was found to be possible to qualitatively identify all the minerals (except titanomagnetites from magnetites) from each other using simple susceptibility and remanence ratios. Using discriminant analysis on both the remanence and hysteresis loop data, it was found to be possible to also distinguish the titanomagnetites from the magnetites purely on the basis of the room temperature measurements. The hysteresis loop data have been used in the development of least squares minimisation algorithms. Two minimisation methods have been devised. In the first method the end-members are the hysteresis loops of different minerals and different domain states (e.g. multi-domain magnetite and single-domain greigite). Each mineral used shows distinct hysteresis characteristics allowing a maximum of ten components to be identified. The algorithm, for the first time, is able to reliably quantify paramagnetic components in natural samples. It can also distinguish between superparamagnetic and multi-domain magnetite using only room temperature hysteresis loop data. Secondly sediment sources (e.g. topsoils and subsoils) have been used to unmix natural sediments. The newly developed qualitative and quantitative methods have been applied to a range of sites. Mineral unmixing of the Papa Westray soils and also soils from the archaeological site Kissonerga-Mosphilia, Cyprus, indicate that the dominant magnetic domain state in the samples is superparamagnetic. The use of three independent magnetic identification techniques (qualitative, quantitative and thermomagnetic) have confirmed the presence of both magnetite and pyrrhotite grains in deep crustal rocks from the German KTB pilot borehole core.

Newlands and Bobbejaan kimberlites, South Africa, contain suites of highly chromian, garnet-rich peridotites amongst their xenolith population and an investigation of these xenoliths has been targeted because there is an overlap of mineral compositions with the garnet-chromite-olivine paragenesis found as inclusions in diamonds. A high proportion of garnets and chromites in these rocks plot in the diamond facies fields on Cr₂O₃-CaO and Cr₂O₃-MgO wt. % plots respectively. However, it has also been found that many Cr-rich assemblages are clinopyroxene-bearing (iherzolitic) as well as harzburgitic. Many samples have garnets with inclusions of serpentine ± chromite (+ clinopyroxine in iherozilitic samples) whose arrangement is sometimes indicative of exsolution or annealed exsolution textures. Modal reconstructions reveal that the original crystals were garnets of higher chromium content. A particularly strong chromite-clinopyroxene association is seen included within garnet in most iherozolitic samples; a feature not reported in other garnet iherzolites. Most the garnets have strongly developed zonation patterns, which are a result of diffusion towards the matrix (external zonation) followed by zonation towards inclusions (internal zonation). Cr-A1 and Mg-Ca inter-diffuse in both types of zonation, Ti may also be strongly zoned, whereas Fe is not distinctly zoned in any sample. External zonation is modelled to have occurred on an order of magnitude greater timescale than internal zonation (~20 Ma compared to ~2 Ma using D_Mg = 10^-20m²/s). Final P-T estimates on clinopyroxine inclusion-garnet boundaries indicate 40-50 kb and 900-1100°C, whereas core garnet compositions yield pressures and temperatures consistently a little higher. Garnet-matrix clinopyroxene P-T estimates yield higher temperatures than the final internal ones. Overall the samples plot on a relatively cool continental steady state geotherm (equivalent to 38 mW/m² conductive geotherm). In order to achieve the level of Cr-saturation in these rocks, repeated melt extraction is required in the spinel stability field. Since Cr stabilises spinel relative to garnet at high pressures, the rocks must have been buried to depths well in excess of 90 km (30 kb) for A1-garnet peridotite facies before the observed exsolution occurred. External zonation appears to have happened whilst garnet was in equilibrium with inclusions so any exsolution (forming the inclusions) represents relatively early decompression to > 50 kb (i.e. greater than the final pressure estimate – see above). External zonation in garnet is the result of equilibrium with the matrix which is predominantly along predictable compositional trajectories. Few samples, however, display evidence for equilibration at the rims with Ca-Ti rich metasomatic fluid thought to have percolated though the matrix. Finally, internal zonation is the result of late P-T re-equilibration with inclusions subsequent to any metasomatic effects from > 50 to 40-50 kb. Initial bulk REE patterns were essentially incorporated into a pre-exsolution, high-Cr garnet and later modified by exsolution of pyroxene.

No description available.

No description available.

When observed optically, alkali feldspars often appear dusty or turbid. The cause of this turbidity is at or beyond the limits of optical resoltuion but by using electron microscopy it has been shown to be intimately related to the presence of micropores. Micropores are found in all pink or white alkali feldspars and are absent only from fresh black feldspars from rocks (such as fayalite or pyroxene bearing rapakivi granites, granulites and a lunar breccia) which petrographically show evidence of a very dry history. In general, micropores, averaging 0.5μm in length, make up around 1.5% of the alkali feldspars although porosities of up to 4.5% have been found. The shape of micropores is commonly controlled by Adularia habit overgrowths. Back-scattered electron imaging and transmission electron microscopy have shown that micropores are absent form strain controlled, coherent perthites and are found only in deuterically coarsened, incoherent, patch perthites. Exceptions to this rule are rare trains of micropores which cross strain controlled perthites and which are the remains of microtracks which healed in crystallographic continuity at temperatures above the alkali feldspars solvus. The character of the pristine areas varies with bulk composition. Ore-rich areas are normally 'tweed' orthoclase, with or without straight lamellar perthites. Intermediate compositions are lamellar or braid perthites of low albite and microcline. Turbid areas contain deuteric perthites and microcline in all cases. The structural integrity of the crystal is also lost in the turbid areas which are made up of numerous subgrains less than 1μm across with micropores in the gaps between the subgrains. Cleavage fragments from the Klokken layered syenite, South Greenland, were heated to 700<SUP>o</SUP>C at 0.1 GPa in 99% H<SUB>2<SUP>18</SUB></SUP>O for 75hr and then polished and imaged on an ion micropore for <SUP>18</SUP>O. It was found that the <SUP>18</SUP>O had penetrated into the microporous parts of the crystals and not into the pristine parts. At least some of the micropores and subgrain boundaries are therefore interconnected and have made the feldspar permeable as well as porous.

No description available.

No description available.

This thesis is concerned with the mineralogy and chemistry of metasomatic reaction zone formed around high-grade blueschist facies blocks (15 kb, 450-500°C) of meta-igneous and metasedimentary lithologies within a serpentinite melange matrix on the island of Syros, Greece. The zones are composed of a reaction rind, cm to 10's of cm thick, containing abundant chlorite, sodic and calcic amphibole and talc, and a zone of altered block. The zones are the result of diffusion-controlled bimetasomatism and, importantly, the products of a discrete, short-lived infiltration event producing amphiboles, which occurred during bimetasomatic zone growth. The generalised bimetasomatic zone sequence is: block | altered block | chlorite | talc ± chlorite ± carbonate ¦ serpentinite matrix with carbonate only occurring in the 'talc ± chloride ± carbonate' zone where the 'block' is carbonate-bearing schist. The position of the original block-matrix contact is generally within the chlorite zone, and is in some cases marked by a relict zone of chromite and Cr-rich pyroxene. The composition and width of the 'altered block' zone was found to vary with block composition. In meta-felsites, with unaltered cores of jadeite + quartz + accessories, the 'altered block' zone is around 20-30 cm wide, and is characterised by a progressive outward decrease in quartz content with a thin zone of omphacite rock occurring at the outermost extent of this zone. In dioritic metagabbros, containing glaucophane-epidote-garnet-pyroxene ± minor quartz, bulk compositional change towards the block edge is less marked but altered block zone-width is greater. The volume proportion of garnet and pyroxene also increases towards the block edge as a result of desilication by the reaction: glaucophane + epidote + omphacite₁ = garnet + omphacite₂ + SiO₂ + H₂O in which the omphacite becomes more jadeitic. In glaucophane-epidote-garnet-pyroxene ± minor quartz ferrogabbros, this desilication reaction also occurs, but no consistent bulk compositional changes towards the block edge are measurable. The zone of 'altered block' in ferrogabbros is estimated to be up to 2-3m wide. The 'altered block' zone at edge of carbonate-bearing schists is composed of 'pseudo-eclogite', an omphacite ± epidote ± garnet rock, the product of a metasomatic decarbonation reaction with antigorite as the CO₂ sink.