This is a study of the supergene mineralogy of the San Xavier West mine located in the Pima mining district, Pima County, Arizona. The number and composition of secondary species collected are found to be closely related to the relative amounts of the various primary minerals and to the manner in which they were emplaced in the host rock. Supergene mineral species were selected that appeared to be in equilibrium with their environment, and certain assumptions are made concerning the stability fields of these minerals. The equilibrium conditions in most cases can be narrowed and often closely defined by combining the stability fields of several secondary minerals. On this basis, two acid environments and one alkaline environment are found to exist and are separable on the basis of mineralogy. The fields of chalcanthite and melanterite define a highly acid environment while those of goslarite and malachite define an environment of lower acidity. The association of calcite, rosasite, hemimorphite, and malachite indicate an alkaline environment. The acidity of the environments is principally determined by the amount of pyrite present, and pH may be lower than 3 if pyrite is abundant and reactive carbonate material lacking. Also, knowledge of stability relations allowed the history of enrichment and subsequent oxidation to be followed in a case where a transitional species has been removed from reaction by inclusions with gypsum.

The presence of frother, either singly or as a blend, is common to most mineral flotation processes. Frothers have two main functions: provide target hydrodynamic conditions and promote froth stability. In the thesis hydrodynamics is measured by bubble size and gas holdup and froth stability by water overflow rate. These functions are related to frother chemistry, or more particularly, molecular structure. However, the structure – function link is not well understood and forms the theme of the thesis. A methodology is developed to link one hydrodynamic function, bubble size reduction represented by CCC95 (critical coalescence concentration 95%), to frother structure represented by HLB (hydrophile-lypophile balance). The results indicate each frother family has a unique CCC95-HLB relationship. Empirical models are developed to predict CCC95 either from HLB or directly from frother structural parameters. Commercial frothers are shown to fit the relationships.Use of dual-frother systems to achieve independent control over the two frother functions is tested. Blends of Alcohols with Polypropylene Glycols (PPG) and Polyethylene Glycols (PEG) are selected in pre-mixed and base/additive styles. The base/additive system 1-Butanol/PEG (EO > = 4) gives the independence sought but not 1-Butanol/PPG which causes an increase in bubble size. To explain the bubble size effect, frother partitioning is determined. This is achieved by development of two analytical techniques: high pressure liquid chromatography (HPLC) and nuclear magnetic resonance spectrometry (NMR) associated with total organic carbon analysis (TOC). The 1-Butanol/PPG blends show strong partitioning to the froth which corresponds to the increase in bubble size. New frothers are synthesized to test the structure-function relationship by combining three groups: linear Alkyl-Ethoxy-Propoxylates (CnEOlPOm) and Alkyl-Propoxy-Ethoxylates (CnPOmEOl). It is found that varying the number of PO groups, the length of Alkyl group chain, and the position of PO and EO could affect the two functions. / Dans la plupart des processus de flottation, la mousse est communément présente soit seule, soit sous forme de mélange. Les mousses assument deux fonctions principales à savoir: créer les conditions hydrodynamiques et promouvoir la stabilité de la mousse. Dans cette thèse l'hydrodynamique est mesurée par la taille de la bulle et la rétention de gaz, tandis que la stabilité, par la vitesse de l'écoulement d'eau. Ces fonctions sont liées à la chimie des mousses et plus particulièrement à la structure moléculaire. Une méthodologie qui permet de relier la fonction hydrodynamique à la taille de la bulle représentée par le CCC95 (concentration critique de coalescence 95%) et à la structure de la mousse représentée par le HLB (budget hydrophile-lyophile), a été développée. Les résultats montrent que chaque famille de mousse possède une relation CCC95-HLB unique. Des modèles empirique été développés afin de prédire le CCC95 soit à partir du HLB, soit directement à partir des paramètres structurels de la mousse. Les mousses commerciales s'accommodent de ces relations. L'utilisation des systèmes à deux mousses visant un contrôle indépendant des deux mousses, a été testée. Les mélanges d'alcool avec les propylène-glycols (PPG) et les polyéthylène-glycols (PEG) ont été sélectionnés et préalablement mélanges et des additifs de base étaient utilisés. Le système Base/Additif 1-Butanol/PEG (EO>= 4) ont prouvé l'indépendance recherchée à la différence de 1-Butanol/PPG qui a cause une augmentation de la taille de la bulle. Pour expliquer l'effet de la taille, la partition de la mousse a été déterminée. Cela a été réalisé grâce au développement de deux techniques: la chromatographie gazeuse de haute pression (HLPC) et la résonance magnétique (RMN) associée à l'analyse du carbone total (TOC). Les mélanges 1-Butanol/PPG ont révélé une forte partition avec la mousse qui correspond à l'augmentation de la taille de la bulle. Des nouvelles mousses ont été synthétisées afin de tester la relation structure-fonction en combinant trios groupes: linéaire Alkyl-Ethoxy-propoxylates (CnEOPOm) et Alkyl-Propoxy-Ethoxylates (Cn(POnEO). Il a été observé qu'en faisant varier le nombre de PO groupes la longueur de la chaîne du groupe Alkyl et les positions PO et EO peuvent affecter les deux fonctions.

As part of a sustained effort to model and simulate the thermodynamic behavior of steelmaking slag, binary and ternary phase diagram systems have been modeled using the CALPHAD (CALculation of PHAse Diagrams) approach. The understanding, modeling and thermodynamic simulation of oxide systems including phosphorus oxide are of special importance to the steelmaking industry given requirements for steels with ever lower phosphorus contents while facing the processing of ores with increasing phosphorus content.In this project, two binary systems (MgO-P2O5 and MnO-P2O5) and one ternary system (CaO-MgO-P2O5) were modeled to extend the oxide database of the Factsage thermodynamic simulation software package. This present study is part of a broader effort conducted to develop a comprehensive thermodynamic database for the CaO-MgO-MnO-FeO-Fe2O3-Al2O3-Na2O-P2O5 / Fe Liq. system and allow the thermodynamic simulation of steelmaking processes such as dephosphorization. All available experimental data in the literature were critically assessed and a set of thermodynamic model parameters were developed to simulate reliably the experimental results within experimental error limits. The properties of liquid oxides and solid P2O5 compounds were described by thermodynamic models and standard Gibbs energy functions.The thermodynamic property of the liquid solution (or slag) were described using the Modified Quasi-Chemical Model (MQM) developed by Pelton et. al. (2000). This model allows to simulate short range ordering (SRO). Thermodynamic properties of solid solution phases are described using the Compound Energy Formalism (CEF) by Hillert (2000) which considers the crystal structure of each solid solution. / Dans le cadre des efforts soutenus pour modéliser le comportement thermodynamique des scories sidérurgiques, des systèmes binaires et ternaires furent modélisés en utilisant l'approche CALPHAD. Les systèmes incluant les oxydes de phosphore sont d'une importance particulière pour l'industrie sidérurgique qui doit satisfaire des demandes pour de l'acier avec des niveaux de phosphore plus bas tout en faisant face à l'exploitation de gisements aux teneures en phosphore chaque fois plus élevés.Dans ce projet, deux systèmes binaires (MgO-P2O5 et MnO-P2O5) et un système ternaire (CaO-MgO-P2O5) furent modélisés pour développer la base de données du logiciel de simulation thermodynamique Factsage. Cette étude se cadre dans des efforts de développement d'une base de données du système CaO-MgO-MnO-FeO-Fe2O3-Al2O3-Na2O-P2O5 / FeLiq. qui pourrais permettre la simulation de procédés sidérurgiques tels que la déphosphorization de l'acier.Toutes les données expérimentales dans la littérature furent évalués de manière critique et optimisées pour obtenir des paramètres de modèles thermodynamiques pouvant reproduire les données expérimentales à l'intérieur des marges d'erreurs expérimentales. Les propriétés thermodynamiques des oxides liquides et des composantes solides contenant du P2O5 furent décrites en utilisant des modèles thermodynamiques ainsi que des fonctions standards d'énergie de Gibbs.Les propriétés thermodynamiques de la solution liquide furent décrites en utilisant le Modèle Quasichimique Modifié (MQM) développé par Pelton et. al. (2000). Ce modèle tiens compte de l'ordre à courte portée des solutions liquides. Les propriétés thermodynamiques des solutions solides furent décrites par le Formalisme Énergétique des Composés (CEF) par Hillert (2000) qui tiens compte de la structure cristalline de chaque solution.

Late Cenozoic volcanism on the Great Plains in northeastern New Mexico was closely related to the tectonic and thermal deformation along the transitional zone between the Colorado Plateau and southern Basin and Range province, which produced a broadly synchronous volcanism along the Jemez lineament. During late Miocene to Holocene volcanism on the Great Plains, a wide variety of magmas were erupted in the Raton-Clayton volcanic field (RCVF) on the eastern flank of the Rio Grande rift, northeastern New Mexico and southeastern Colorado. Late Cenozoic volcanic rocks from the RCVF range in composition from nephelinite through basalt to dacite and rhyolite; many are alkalic. Mafic feldspathoidal magmas were derived from an enriched lithospheric mantle by very small degrees of melting. While some show evidence of apparent olivine fractionation, most were not significantly fractionated and/or contaminated by the crust and provide the best representation of the isotopic composition of the lithospheric mantle. Alkali and subalkali basalts are also quite magnesian (i.e., MgO $>$ 8%), indicating very limited fractionation. Compositional variations of the mafic volcanic suite from the RCVF, therefore, were primarily produced by differential melting of their lithospheric mantle source. Intermediate to silicic volcanic rocks were generated by fractionation from mafic magmas derived from the mantle. Isotopic compositions of earlier volcanic rocks reflect their hypothetical lithospheric mantle source, whereas those of the basaltic trachyandesites indicate a second, more depleted mantle source that was involved in the latest stage of the magmatism. While very small isotopic deviation from the lithospheric mantle of the silicic rocks indicates a relatively closed magmatic system, limited isotopic variation of mafic rocks probably resulted from their short residence in the crust. Regional petrogenetic comparison of early silicic rocks also shows that late Cenozoic silicic magmatism on the Great Plains evolved in a more closed system.

Voluminous mid-Miocene (8-13 Ma) rhyolitic ash-flows and lava flows (Idavada volcanics) are exposed continuously along the northern (Mount Bennett Hills or MBH) and southern margins of the west-central Snake River Plain (SRP). These rhyolites unconformably rest on Cretaceous granitic rocks of Idaho batholith and Eocene Challis volcanics and are locally overlain by (or intercalated with) minor Tertiary basalt flows and fluvial and lacustrine sediments. At both margins, the rhyolites dip gently and thicken toward the SRP. NW-to EW-trending normal faults drop the western SRP forming a graben-like structure where the inferred source vents of the rhyolites are buried by younger basalt flows (Pliocene to Recent) and sediments. The stratigraphy of Idavada volcanics in the MBH and evaluation of correlations with those exposed in the adjacent areas suggest that the rhyolites exposed in the northern margin of the west-central SRP are different from the southern margin counterparts and that the majority of west-central SRP rhyolites were erupted almost concurrently with the formations of the normal faults that formf"d the western SRP graben. Idavada volcanics range in composition from quartz latite to high-silica rhyolite and are characterized by anhydrous mineral assemblages (Plagioclase ±Sanidine ±Quartz + Augite + Pigeonite ± Hypersthene + Fe-Ti Oxides ± Fayalite + accessory Zircon and Apatite), suggesting that the rhyolites were formed under dry conditions. Pyroxene temperatures of Davidson and Lindsley (1985) range from 800 (high-silica rhyolites) to lOOO°C (quartz latites). Major and trace element, and isotopic (87Sr/86Sr = 0.709-0.713; 143Nd/144Nd = 0.51219-0.51230) compositions of the rhyolites suggest that the rhyolites may have formed by partial melting (caused by intrusions of hotspot-related basaltic magmas into the lower crust) of sialic lower crust with significant Juvenile component under dry conditions. Major and trace element variations observed in the west-central SRP rhyolites can be explained by differences in residence times which resulted in variable degrees of fractional crystallization. Variations of isotopic compositions and estimated magmatic temperatures of the rhyolites suggest that higher temperature rhyolitic magmas assimilated more Archean upper crustal material than lower temperature ones.

Thermodynamics of mixing for solids and liquids in the anhydrous apatite system Ca$\sb3$(PO$\sb4)\sb2$ - CaF$\sb2$ - Ca(OH)$\sb2$ - CaCl$\sb2$ are quantified using mathematical modeling of new and existing experimental data. These data apply to apatite experiments or to understanding natural apatites. Fluorapatites (FAp) are hexagonal, although hydroxyapatite and chlorapatite (ClAp) are monoclinic. Although the monoclinic and hexagonal endmembers are believed to be immiscible, no evidence of immiscibility exists. Miscibility is achieved by movement of the anions away from the sites occupied in the end member and by expansion of the apatite lattice. Anion-anion interaction and lattice deformation may lead to possible nonideal mixing of the endmembers. Thermodynamic treatment of experimental data gathered above 475$\sp\circ$C shows that apatite solid solutions may be considered ideal. Standard state data for chlorapatite are derived. Thermodynamic modeling of the molten mixtures uses a stoichiometric approach with Ca$\sb{1.5}$PO$\sb4$ and Ca$\sb{0.5}$X (X = F, OH, Cl) liquid phase components. New data are presented for joins Ca(OH)$\sb2$ - CaCl$\sb2$ and CaF$\sb2$ - Ca(OH)$\sb2$. Molten mixtures on the join CaF$\sb2$ - CaCl$\sb2$ are ideal. Ca(OH)$\sb2$ - CaCl$\sb2$ and CaF$\sb2$ - Ca(OH)$\sb2$ molten mixtures show positive enthalpies of mixing typical of common cation-mixed anion systems. The phosphate-salt joins show small positive entropies of mixing due to formation of polymeric species in the phosphate melt not explicitly considered in the model. Heats of fusion for solids and standard state data for the liquid phase components, CaClF, CaOHCl and Ca$\sb2$PO$\sb4$Cl are derived. The apatite thermodynamic data are applied to existing data for the Bishop Tuff (BT) and new data for the Fish Canyon Tuff (FCT). Results show that both magmas were saturated with a fluid phase prior to eruption, in agreement with earlier studies. Apatite inclusions in FCT phenocrysts preserve the pressure- temperature path of the magma, requiring a higher pressure for plagioclase crystallization than for hornblendes. Similar treatment of the BT shows that the pressure paradox observed by earlier workers is also found in the apatite chemistry. Applications are limited currently by analytical problems, but the ratio of X$\sb{\rm FAp}$/X$\sb{\rm ClAp}$ is probably accurate.

Metasedimentary and metabasaltic rocks of the Pelona schist of Sierra Pelona, southern California, preserve a relatively high P/T inverted metamorphic gradient (Graham and Powell, 1984) and provide an opportunity to study possible compositional changes in an oceanic slab progressively metamorphosed under P-T conditions similar to those in shallow parts of some subduction zones. Progressive metamorphism resulted in continuous major element compositional change in plagioclase, white mica, amphibole, and epidote. Variations in whole-rock compositions do not correlate with increasing metamorphic grade and largely appear to reflect protolith heterogeneity. Loss of H$\sb2$O-rich fluid during chlorite breakdown reactions largely accounts for the decrease in weight percent LOI and H content with increasing metamorphic grade. Comparison with unmetamorphosed equivalents and lower temperature, high-pressure metamorphic rocks (including metabasaltic samples from the Shuksan schist and the Franciscan Complex) suggest that As and Sb may be lost from metasedimentary rocks and that B may be lost from metabasaltic rocks at temperatures less than 450$\sp\circ$C. Trace element analyses obtained by ion microprobe for minerals in the Pelona, Catalina, and Shuksan schists document the mineral residencies of trace elements and the redistribution of trace elements among minerals as modal abundances vary. Boron, Ba, Li, Rb, and Cs are strongly concentrated in micas. In samples with coexisting white mica and biotite, higher concentrations of B, Ba, and Sr are observed in white mica, relative to Li, Rb, and Cs, which preferentially substitute into biotite. Cesium and Rb partition in a relatively constant ratio between white mica and biotite. Strontium is strongly concentrated in epidote, but is increasingly incorporated into white mica and plagioclase as the modal abundance of epidote decreases and as plagioclase compositions become more calcic. Similarly, Li appears to be repartitioned into amphibole as chlorite decreases in modal abundance. These observations support a model of gradual release of fluid-mobile trace elements during progressive metamorphism (as from subducted slabs). Prograde continuous reactions involving mica may particularly control the mobility of alkali and alkaline earth elements enriched in arc magmas and provide a mechanism for fractionating incompatible element ratios (e.g. B/Cs) during subduction zone metamorphism.

An investigation of the petrography, mineralogy and chemistry of primitive basalts found in a E-W transect of southern Washington and northern Oregon provides constraints concerning the thermal structure and composition of the mantle wedge beneath the Cascadia subduction zone. The Cascades represents a warm, slowly subducting end-member of subduction zone types. Chemical and petrographical characteristics of basalts from a transect across southern Washington representing at least six variants of primitive melts can be recognized. Estimates of magma temperature and depth based on mineral and whole-rock chemistry provide estimations on the Cascadia subduction zone's thermal structure. Mineral chemistry and whole-rock trace element chemistry are further used to interpret the mantle chemistry and the processes involved in producing subduction zone basalts. The results can be explained by multiple sources found within the subduction zone mantle, melting at various depths and temperatures.

Fibroferrite [Fe(OH)SO4•5H2O] is a product of the AMD (acid mine drainage) process and forms by chemical weathering and oxidation of iron sulfides in hydrothermal and magmatic ore deposits. The present work consists of two distinct parts. The first part is a crystallographic and optical study of 30 samples of fibroferrite from natural occurences and mine waste. The second part covers the description of all the experimental attempts to synthesize fibroferrite. Powder-diffraction data were collected and the Rietveld refinement procedure was conducted for each of the 30 natural samples in order to refine the values of the lattice constants of each sample. The a and c values of the lattice constants of fibroferrite measured in this study were found to conform to the values reported in the most recent literature. Optical observations were conducted, using a spindle stage, on each of the 30 natural samples. The purpose of the optical analysis was the selection of crystals adequate for a single crystal X-ray diffraction study. Samples were selected for a study on a rotating anode source and further studies at a synchrotron facility. None of the selected crystals was suitable for a single crystal X-ray diffraction study and the unit cell could not be found. Energy dispersive X-ray spectroscopy chemical analyses using a scanning electron microscope were performed on the 10 natural samples supplied by the Royal Ontario Museum and the Canadian Museum of Nature. The energy dispersive X-ray spectroscopy chemical analyses did not allow a real understanding of the level of substitutions of other cations for ferric iron within the atomic structure of fibroferrite. Several experiments were performed in order to synthesize fibroferrite. Commercially available reagents were used. All the experimental attempts conducted failed in synthesizing fibroferrite. A too restricted Eh and pH range of the laboratory conditions at which the experiments were conducted and the formation of metastable phases could be evoked as reasons of this failure. / Thesis (Master, Geological Sciences & Geological Engineering) -- Queen's University, 2010-05-06 15:08:18.092

Determining whether or not a gold gravity circuit should be installed in a gold plant requires a prediction of how much gold will be recovered. This has always been a difficult task because recovery takes place from the grinding circulating load, in which gold's behavior must be described. / A population-balance model (PBM) to predict gold gravity recovery was developed at McGill University in 1994 (Laplante et al, 1995). The objective of this research was to make this PBM user friendly. This was achieved in two different ways. First, the behavior of gravity recoverable gold (GRG) in secondary ball mills and hydrocyclones was described by two parameters, tau and R-25mum, and these parameters were linked to the circulating load of ore and the fineness of the grinding circuit product, for easy estimation. Second, the database of simulations produced by the PBM was represented by two multilinear regressions (one for coarse GRG, the other for fine GRG) linking the predicted GRG recovery to the natural logarithm of tau, R-25mum , the size distribution of the GRG and the recovery effort (Re ), defined as the proportion, in %, of the GRG in the circulating load recovered by gravity. Re was found to be the most significant parameter, tau the least. The GRG size distribution, represented either by two (coarse GRG) or three (fine GRG) points on the cumulative passing curve, has a significant impact on recovery. A total of twenty different GRG size distributions were used to generate the simulation database. / The multilinear regressions were tested on four case studies, and found to predict GRG recovery well within the precision with which the GRG content can be measured, a relative 5%. Whenever size-by-size recovery data are available, the PBM itself would be used; if not, the simpler regressions would be preferred.