Search

This research developed and applied high-precision analytical methods (electron microprobe) to determine the major- and minor-element abundances (Si, Ti, A], Cr, Fe, Mn, Mg, Ca, Na, K) in three suites of Apollo 14 volcanic green glasses. Since the precision of these data was a factor of 5-10 better than previously published results, the data provided the strongest constraints yet achieved on the geochemical processes that operated at depths less than 400 kilometers inside the Moon. The results of this work were submitted to the 29th Lunar and Planetary Science Conference, formed the basis for a Master's Thesis, and were published in the top-tier geochemical journal. Additional high-precision data on the trace abundances of sodium (Na), sulfur (S), and potassium (K) in the Apollo 17 volcanic orange glasses (74220) were also acquired during this grant-period. These data show diffusion profiles within individual 200-micron diameter, glass spherules that were apparently generated during the fire-fountain eruption of this magma in the presence of a transient vapor phase. Although these results have not yet been published, publication is planned in 2001 after additional numerical modeling has been completed. This work has also catalyzed on-going collaborative work with Professor Timothy Grove (MIT) to analyze and understand the origin of the Apollo 15 volcanic green glasses. That work will be completed for publication in 2001.

A 6 mm-diameter dark spherule, 15434,28, from the regolith on the Apennine Front at the Apollo 15 landing site has a homogeneous glass interior with a 200 microns-thick rind of devitrified or crystallized melt. The rind contains abundant small fragments of Apollo 15 olivine-normative mare basalt and rare volcanic Apollo 15 green glass. The glass interior of the spherule has the chemical composition, including a high FeO content and high CaO/Al2O3, of a mare basalt. Whereas the major element and Sc, Ni, and Co abundances are similar to those of low-Ti mare basalts, the incompatible elements and Sr abundances are similar to those of high-Ti mare basaits. The relative abundance patterns of the incompatible trace elements are distinct from any other lunar mare basalts or KREEP; among these distinctions are a much steeper slope of the heavy rare earth elements. The 15434,28 glass has abundances of the volatile element Zn consistent with both impact glasses and crystalline mare basalts, but much lower than in glasses of mare volcanic origin. The glass contains siderophile elements such as Ir in abundances only slightly higher than accepted lunar indigenous levels, and some, such as Au, are just below such upper limits. The age of the glass, determined by the Ar-40/Ar-39 laser incremental heating technique, is 1647 +/- 11 Ma (2 sigma); it is expressed as an age spectrum of seventeen steps over 96% of the Ar-38 released, unusual for an impact glass. Trapped argon is negligible. The undamaged nature of the sphere demonstrates that it must have spent most of its life buried in regolith; Ar-38 cosmic ray exposure data suggest that it was buried at less than 2m but more than a few centimeters if a single depth is appropriate. That the spherule solidified to a glass is surprising; for such a mare composition, cooling at about 50 C/s is required to avoid crystallization, and barely attainable in such a large spherule. The low volatile abundances, slightly high siderophile abundances, and the young age are perhaps all most consistent with an impact origin, but nonetheless not absolutely definitive.

Parameterizations have been developed for impact glass compositions which makes its possible to acquire geochemical information about pre-impact target materials. The impact glass populations among unrelated highland regoliths including Apollo 16 regolith 64001 and the lunar meteorites ALHA81005 and MAC88105 have been compared. It is concluded that the Apollo 16 impact glasses are compositionally more diverse than those in the lunar meteorites. This diversity reflects the complex geology of the Apollo 16 region, in which extensive exposures of mare basalts occur within 300 km. Apollo 16 regolith 64001 contains a distinctive subset of impact glasses with major element compositions similar to lunar highland meteorite. These impact glasses (less than 100 micrograms) indicate that the lunar highlands are geochemically more diverse than the compositional range currently represented by large (more than 1 gram) samples. Lunar impact glasses provide geochemical constraints on the full range of regolith compositions that occur on the moon.

The pristine glasses of Delano are the most primitive lunar basaltic magma compositions discovered to date. They are grouped into two (and possibly three) arrays: a low-alumina array and a high alumina array. These glasses are very olivine normative and are multiply saturated at pressures of approximately 20 kbar, implying a depth of origin of 400 to 500 km in the Moon. Thus, these glasses appear to be the best candidates for primitive partial melts of the upper lunar mantle. One of the most perplexing characteristics of the pristine glasses is a positive correlation between Ni and SiO2 within each array. This is contrary to the terrestrial experience, where Ni is observed to positively correlate with MgO and negatively correlate with SiO2. These systematics are believed to be due to the depletion of Ni by olivine fractionation. The difference between the lunar and terrestrial Ni vs. SiO2 trends may be partially ascribed to the Ti-rich component. In the case of the pristine glasses, SiO2 increases not because of olivine fractionation, but because they contain less of the high-Ti component. An attempt was made to model this variation in Ni and SiO2 with a simple assimilation-fractional crystallization (AFC) model. Silica and Ni both decreased dramatically as the AFC process proceeded. Only 15 to 20 percent AFC was necessary to produce the observed variation, and the SiO2 vs. Ni variation was modeled quite well. The D(Ni) for olivine/liquid in this model was taken to be 10 and the olivine was assumed to be Fe sub 80. However, the results of this model for Ti and Mg were less than satisfactory. It seemed difficult to achieve the high TiO2 contents of some glasses (16 to 17 wt. percent) by this method. Continual addition of ilmenite by AFC could indeed raise the titania concentrations to the necessary levels, but only by enriching the magma in FeO and greatly depleting the magma in MgO. An attempt was made to circumvent this problem by using armalcolite, (Fe, Mg)Ti2O5, in the AFC model, and the results are presented.

This workshop on lunar mare volcanism was the first since 1975 to deal with the major scientific advances that have occurred in this general subject, and the first ever to deal specifically with volcanic glasses. Lunar volcanic glasses are increasingly being recognized as the best geochemical and petrologic probes into the lunar mantle. Lunar volcanic glasses, of which 25 compositional varieties are presently known, appear to represent primary magmas that were produced by partial melting of differentiated mantle source regions at depths of perhaps 400 to 500 km. These high-magnesian picritic magmas were erupted onto the lunar surface in fire fountains associated with the release of indigenous lunar volatiles. The cosmic significance of this volatile component, in an otherwise depleted Moon, remains a lingering puzzle. The resource potential, if any, of the surface-correlated volatile sublimates on the volcanic glass spherules had not been systematically addressed prior to this workshop.

The intrinsic oxygen fugacity (IOF) of olivine separates from the Salta, Springwater, and Eagle Station pallasites was measured between 850 and 1150 C using oxygen-specific solid zirconia electrolytes at 100,000 Pa. Thermodynamic calculations of redox equilibria involving equalibrium pallasite assemblages are in good agreement with the experimental results and provide a lower limit to pallasite redox stability; others involving disequilibrium assemblages, suggest that pallasites experienced localized, late-stage oxidation and reduction effects. Consideration of the redox buffer metal-olivine-orthopyroxene utilizing calculated Eucrite Parent Body (EPB) mantle phase compositions indicates that small redox gradients may have existed in the EPB. Such gradients may have produced strong compositional variation within the EPB. In addition, there is apparently significant redox heterogeneity in the source area of Eagle Station Trio pallasites and Bocaiuva iron meteorites.

A three-component model for the moon's bulk composition was developed on the basis of the results of Delano (1986) on lunar pristine glasses. This suite of models is based on the following two assumptions: (1) that the early moon differentiated into two primary reservoirs, a FeO-rich magma ocean and an olivine-rich residuum of about Fo(90); and (2) that this magma ocean then differentiated into an olivine-dominated cumulate, about equal to or greater than Fo(80), and a primitive liquid composition. Because serious uncertainties remain concerning the moon's complex differentiation history, this three-component model does not yield a single unique composition for the bulk silicate moon; instead, it yields a range of restricted compositions that are permitted within the conceptual framework of the model.

Apollo 14 regolith breccias (14313, 14307, 14301, 14049, 14047) have been found to have different populations of nonagglutinitic, mare-derived glasses. These variations appear to not only reflect different source regoliths but also different closure ages for these breccias. Based upon these different glass populations, 14301 is inferred to have a closure age sometime during the epoch of mare volcanism. All of the other four breccias were formed after the termination of mare volcanism with a possible age sequence from old to young of the following: 14307, 14313, 14049, 14047. Due to the relative simplicity of acquiring high-quality chemical data on large numbers of glasses by electron microprobe, mare glass populations allow: (1) classification of regolith breccias with respect to provenance and (2) estimation of their relative and absolute closure ages. The determination of (Ar-40)-(Ar-39) ages on individual glass spherules within breccias using the laser probe should in the future prove to be a promising extension of the present study.

The trace abundances of Ni have been determined in 20 high-Mg magmas produced by partial melting of the lunar mantle. Associated with the eruption of the high-Mg magmas was the release of indigenous volatile elements. Data suggest that these volatiles were derived from primordial blocks of debris from the outer solar system that became entrained in the circumterrestrial accretion disk. The survival of these reservoirs within the lunar interior places constraints on the moon's thermal history.