<p>Layered igneous intrusions such as the Stillwater Complex in Montana contain the most economic concentrations of platinum-group elements (PGE) in the world, yet the processes involved in the enrichment of these PGE remain unclear. Some researchers propose that the PGE were enriched into sulfide phases through purely magmatic processes, while others postulate that late-stage, high-temperature fluids caused remobilization of the more soluble elements upwards from the base of the crystal pile. Although much work has been carried out on the economic PGE-enriched ore zone (J-M reef), the silicate mineralogy and the bulk geochemistry of the Complex, the detailed petrographic trends have not been investigated. This dissertation comprises a detailed petrographic study into the assemblages associated with sulfide and other trace minerals throughout the stratigraphy.</p><p>Sampling was carried out from both surface outcrops and drill cores over four consecutive field seasons. Polished thin sections were produced which were then examined by petrographic microscope and electron microprobe. In addition, bulk rock analysis was carried out by x-ray fluorescence spectrometry (XRF).</p><p>In brief, the sulfide and trace mineral assemblage studies described below reveal a number of interesting observations. An upwards trend from pentlandite-rich to pyrrhotite-rich to chalcopyrite + pyrite-rich assemblages is observed below the reef, and the same trend occurs above the reef with the transition occurring just below the reef, in upper GN-I. Trace element analysis shows that Cu levels are higher above the reef than below it, and that although Zn and Cu contents are correlated below the reef, a restricted range of Zn contents occurs above the reef, while Cu is highly variable. As all `low-temperature' assemblages (those associated with extensive silicate alteration or the presence of greenschist facies minerals such as chlorite, clinozoisite and epidote) were discounted, the majority of sulfide assemblages present were either pristine(multiphase, often globular in shape, with no associated silicate alteration) or high-temperature (multiphase, with high-temperature minerals such as biotite, hornblende, carbonates, etc, and with little associated silicate alteration) in occurrence. Some differences were observed between the hanging-wall and footwall rocks, including the presence of native copper, sphalerite in a calcite-hornblende vein, and high-temperature carbonates in footwall and not hanging-wall rocks. The high-temperature carbonates observed comprise dolomite with exsolved patches of calcite. The textural relationships and Fe-Mn compositions of the Stillwater carbonates are similar to those of mantle carbonates. High-temperature desulfidation is also observed both above and below the reef, in the form of pyrite being converted to magnetite, and chalcopyrite to a Cu-Fe-oxide (delafossite). Both sets of assemblages are associated with little to no silicate alteration. When taken together, the upwards increase in Cu and S, the variable Cu contents above the reef, the native copper, high-temperature carbonates and high-temperature sphalerite-bearing veins below the reef, and the evidence for desulfidation are all most readily explained by the remobilization of selected phases by a high-temperature fluid. This dissertation provides evidence that the fluid present in the latter stages of Stillwater formation had a carbonic as well as a Cl-rich component, and would therefore have been efficient in PGE remobilization.</p> / Dissertation

The J-M Reef of the Stillwater Complex, Montana (a large layered mafic intrusion), is one of the highest grade platinum group element (PGE) deposits known in the world, producing primarily palladium and platinum in a 3.4:1 ratio. "Ballrooms" of the Stillwater Complex are anomalously wide areas within or stratigraphically below the J-M Reef that host platinum group element mineralization. Ballrooms have two typical morphologies (type 1 and type 2); the first is an abrupt thickening of the mineralization that extends below the Reef Package and the second is a gentle widening of the Reef Package and associated reef mineralization to a width of over 6 m. Ballrooms are highly variable in size. Minimum dimensions for ballroom designation are a thickness (perpendicular to strike) of 6 meters and a length of 5 meters (parallel to strike). Mineralization contacts are irregular but sharp and are characterized by a dramatic decrease in sulfide content (from one to two percent in ballrooms to only trace amounts, Whole rock major and trace element compositions of rocks from ballrooms exhibit a strong geochemical control by cumulus phases. There are no significant major or trace element differences in the rocks from the two ballroom types. Moreover, cumulate mineralogy in ballrooms shows no variation from cumulate mineralogy in the JM Reef. Magnesium, Fe, and Cr exhibit a strong correlation with one another and the other major elements but do not correlate with Cu, Ni, and S. This indicates that Cu, Ni, and S were controlled by processes other than those controlling the distribution of the major elements in cumulus phases. Cl-rich hydrous phases in the ballrooms (apatite and phlogopite) are evidence for the presence of Cl-rich fluids that interacted with melt in the mineralized zone, inferred to coincide with the growth of cumulus silicate phases. Pegmatitic textures also evidence the presence of fluid. The concentrated fluids played the major role in the formation of these anomalously rich ore morphologies. This fluid likely originated when intercumulate melt became fluid saturated during crystallization of the cumulate pile at the base of the magma chamber and migrated upward as Boudreau (1999) suggests. This fluid appears to have been concentrated in some areas to form locally enriched areas of PGE mineralization (ballrooms). Areas of extensive fluid-melt interaction could produce type 2 ballrooms, while type 1 ballrooms were formed where there was little or no melt present when the upwelling fluid became sulfide saturated. The fluid generation and migration may have been caused by an eruption of flood lava from the crystallizing magma chamber. It is possible that even a small eruption from the chamber could generate a large enough pressure decease to induce fluid saturation in the melt remaining in the cumulate pile. This process may have repeated each time lava erupted from the evolving chamber and created multiple sulfide horizons in the Stillwater Complex. Evidence of sulfide remobilization and low temperature secondary alteration is abundant in ballrooms. The secondary alteration phases include sericite, zoisite/clinozoisite, serpentine, magnetite, pyrite, talc, and chlorite. A regional metamorphic event at 1.7 Ga that changed the Pb isotopic composition of the sulfides is likely the cause of the alteration. This low temperature hydrothermal event locally remobilized sulfides, chalcophile elements, and PGEs in the J-M Reef and ballrooms and may have variably depleted or enriched parts of the mineralization. This remobilization of sulfides, chalcophile elements, and PGEs has had a significant influence on the local distribution (centimeters to a few meters) of PGE bearing sulfides.

The Stillwater Complex J-M reef, the only economic platinum deposit in the United States, consists of a 0.5 to 4 m-thick stratiform horizon of PGE-rich sulfides in an Archean layered mafic intrusion. The origin of this reef has been studied extensively using geochemical methods, yet remains highly debated. Dynamic magmatic processes have been virtually ignored in these geochemical studies. Magnetic methods provide a proven inexpensive approach to offer rapid, and reproducible results to deliver insight into these dynamic processes. I propose to investigate the variations of magnetic properties of layered rocks of the Stillwater Complex in the stratigraphic vicinity of the J-M reef. In this study, detailed magnetic methods were performed on a 115 ft core containing the J-M reef and adjacent rocks. A previously undiscovered cyclicity of magnetic susceptibility was found in the hanging wall and J-M reef section. Further tests were performed to determine the origin of the magnetic cycles. The footwall section lacked the magnetic properties seen in the J-M reef and hanging wall rocks. Both anisotropy of magnetic susceptibility, and high field magnetic data was collected at a high resolution interval along the core. It is possible that the results of this study can be used to constrain the origin of the ore body.