Abstract As is true of the other microsampling analytical techniques addressed in this book, laser Raman microprobe (LRM) spectroscopy can be applied to a wide variety of geological materials. The goal of this chapter is to demonstrate the usefulness of LRM analysis to economic geology in general, and to fluid inclusions in particular. The chapter begins with a brief consideration of the analytical challenge presented here, namely the physical nature of fluid inclusions and the types of information desired from their analysis. As a means of convincing the economic geologist of the applicability of Raman analysis, an overview is then presented of the types of direct and indirect information that can be obtained via Raman spectroscopy. After a brief discussion of the physical and chemical principles behind Raman spectroscopy and the typical instrumentation used in modern laboratories, numerous short summaries are presented on the successful application of Raman spectroscopy to different types of ore deposits. Owing to the breadth of the literature on LRM spectroscopy and the page limitations of this chapter, no attempt is made to cover the field completely. For additional summaries of Raman (microprobe) applications in geology, the reader is referred to McMillan (1989), Burke (1994), and McMillan et al. (1996).

We review mantle redox models and present new data for xenolithic and megacrystic intrinsic oxygen fugacity (IOF) studies from varied geologic settings. The roles of fluid inclusions, carbon, autoreduction, auto-oxidation, Ti 3+ , crystal defects, disproportionation, metasomatism, and gravity are examined with regard to their possible influences on redox data. In several IOF studies, the geothermometric determination for a multimineral sample yields very close temperature concordance with totally different geothermometric techniques; these specific studies mandate some confidence in the IOF f O 2 values obtained. Also of high confidence are the IOF data that come from nearly flawless, gem-quality megacrysts (GQ) of various silicates. Both types of these high-confidence IOF data indicate that the redox state of the mantle is nearer the wüstite-iron (WI) buffer than the quartz-fayalite-magnetite (QFM) buffer. The f O 2 data of the ilmenite-containing xenolithic assemblages that have either been calculated (Eggler, 1983), IOF-measured (Arculus and others, 1984), or gas mixture-equilibrated (McMahon, 1984) have all shown reasonable overlap in f O 2 -T values, and all indicate that these samples come from a mantle region more like (QFM). The ilmenitic xenoliths subjected to IOF analysis in our laboratory have exhibited auto-oxidation and therefore are difficult to interpret unequivocally. The high-confidence IOF data do argue strongly for mantle redox inhomogeneity, but much more work is needed to establish whether there is a general systematic redox decrease with depth into the mantle, on which the observed redox inhomogeneity is superimposed.

Petrography and electron microprobe analysis were used to investigate the mineralogy, bulk composition, abundance, and distribution of Cu-Fe-Ni sulfide phases in a suite of about 25 alkalic-basalt-hosted xenoliths from Kilbourne Hole, New Mexico. The xenoliths include spinel lherzolites, spinel clinopyroxenites, clinopyroxene megacrysts, mafic granuloblastites, and a sillimanite-quartz granuloblastite. Sulfide abundance is low in all the xenoliths, but shows a maximum (∼0.5 vol. %) in pyroxenites and megacrysts. The sulfides occur as interstitial to silicate and oxide grains, fully enclosed and isolated in grains, and along fracture surfaces in grains. The major sulfide phase is monosulfide solid solution (∼pyrrhotite), accompanied by variable amounts of pentlandite, and cubanite. Both the mineralogy and bulk sulfide composition are distinguishable among the xenolith types: Ni-rich for the lherzolites, Ni-poor for the crustal xenoliths, and intermediate for the pyroxenites, megacrysts, and composite xenoliths. The sulfides in the Kilbourne Hole spinel lherzolites most likely result from partial retention, in the residual silicate rock, of an immiscible sulfide formed during mantle partial melting.

Structural, lithologic, and geochemical relationships are integrated to provide a basis for modeling upper mantle metasomatism, the evidence of which is observed in mafic and ultramafic rocks of peridotite massifs and xenoliths from basalts and kimberlites. Mafic rock types associated with peridotite occur as dikes, and most metasomatic phenomena observed in peridotites appear to be directly related to the magmatic rocks. The chronology of mafic dike emplacement, established by crosscutting relationships, is the same in peridotite massifs and xenoliths in basalts, and progresses toward lower pressure assemblages. Metasomatic alteration of peridotite wallrock in the mantle occurs locally adjacent to the dikes, which are crystallized initial magmas or magmatic differentiates separated from their parent dikes. The alteration is caused both by diffusion and infiltration processes. Additional metasomatism (“cryptic”) can be caused by a CO 2 -rich, light rare earth element (LREE)-enriched gas phase evolved from the mafic dikes and distributed through hydro-fractures in peridotite. Crosscutting relationships of mafic dikes indicate that local metasomatic events have occurred serially in the melting history of peridotite, so that superposition of different types of metasomatic effects is to be expected. The sequential melting and metasomatic events are interpreted as consequences of diapiric rise of the host peridotite. The type and number of melting-metasomatic events evident in xenoliths is thought to be limited by the level of diapiric rise at which eruption occurs.

Available major, trace element, and isotopic data from peridotite xenoliths show that the chemical signature of “cryptic” metasomatism (characterized by enrichment of light rare-earth elements, LREE, relative to heavy rare-earth elements, HREE, in bulk rock and clinopyroxenes) also is seen in rocks affected by “patent” metasomatism (characterized by the presence of secondary hydrous minerals). Detailed major element, REE, and isotopic analyses of minerals in two similar composite xenoliths from Dish Hill, California, demonstrate that patent and cryptic metasomatism, as currently defined, can be caused by the intrusion of a mafic melt. Clinopyroxenes may become enriched in LREE at the same time that hydrous minerals are deposited and minerals of the peridotite wallrock are enriched in Fe, Ti, and Al by reaction with volatile-rich fluid from a dike (Fe-Ti metasomatism). Microscopic solid and fluid inclusions have been identified as sources of incompatible elements in peridotite xenoliths. Distributions of such inclusions in the composite xenoliths show that they can be added by mantle fluids. Limited data suggest that bulk rock enrichment of LREE/HREE in clinopyroxene-poor, low-CaO metasomatized anhydrous peridotite xenoliths can be controlled by olivine rather than by pyroxene. This control may be due to LREE-enriched inclusions in olivine and other minerals that do not accommodate REE in structural sites. If the chemical signature of cryptic metasomatism is caused by microscopic additions, the distinction between “patent” and “cryptic” categories becomes arbitrary. The evidence suggests that mantle metasomatism results from reaction between fluid differentiates of mafic melts and peridotite wallrocks. Variations of initial melt compositions, degrees of melt differentiation, and ratios of peridotite to melt all influence the LREE and isotopic compositions of metasomatized mantle rock.