Eruption conditions and mechanism of A17 Orange Glass eruption: petrology and modeling data

The eruption of the orange glass magma to form the deposits sampled at the Apollo 17 site is considered to have been a gas-rich, fire-fountain eruption. The deposit consists entirely of small 1-2 mm spherical beads of the orange glass created during fragmentation of the magma. The data on the gas phase components in the pre-eruption A17 orange glass magma are constrained by C-O-S system phase equilibria and the compositions of glass and phenorysts in the erupted samples. Glass-rich melt inclusions trapped in 74002 olivine exposed in orange glass thin sections have been analyzed for major elements as well as sulfur and other volatiles, correcting for the crystallization of olivine and ilmenite around the margins of the inclusions to obtain the pre-eruption composition. The sulfur present in the melt inclusions indicates there has been a loss of 560 ppm of S from the pre-eruption melt. The sulfur fugacity would not initiate the nucleation of a gas in this magma; it had to be created by the oxidation of reduced C in the magma. The depth of equilibrium C-O gas generation by graphite oxidation is at 20 MPa or 4 km, and the S would partition into the CO-rich gas. The modeling calculations constrain the amount of C necessary to achieve fragmentation of the orange glass melt prior to its reaching the surface. Introduction. Primitive picritic lunar glasses such as the Apollo 17 orange glass, appear to be the product of lunar fire-fountaining (1,2). The pre-eruption conditions in the A17 Orange glass magma, and the mechanism by which this magma erupted in a fire fountain event remain interesting, and to some extent, unanswered questions of lunar volcanism. The problem is particularly interesting because the eruption was clearly driven by the development of a significant gas phase, and clues to the composition and abundance of this gas phase should result from an accurate model of the eruption (3). To review, the A17 orange glass was collected on the surface near Shorty crater (74220) and in the 68 cm core (74001/2) taken at the same site, and samples have been extensively studied (4,5). The upper 5 cm of the core and the 74220 surface material were affected by surface reworking, but the rest of the core represents essentially pure volcanic ejecta of orange glass composition. The base of the core is composed primarily (93%) of black beads rich in skeletal olivine and ilmenite; these beads represent melt fragments that cooled slowly enough for crystallization to take place (4). The top of the core consists of orange glass beads that cooled fast enough to preclude crystallization. The presence of olivine phenocrysts enclosed in orange glass beads of samples 74220 and 74002/1 reveals that olivine was present on the liquidus of the melt from which the volcanic eruption was derived. From their size, shape, and unzoned character, it is inferred that these olivines crystallized prior to the eruption (5). The olivines trapped samples of the melt (melt inclusions) as the phenocrysts grew at depth. Some of these inclusions may have leaked through cracks in the host crystal during magma ascent and eruption, but others probably represent undegassed samples of pre-eruption melt, possibly modified by added crystallization of host olivine. The presence of Fe-Ni metal blebs within the olivine phenocrysts and in the orange glass, indicate that these metals were also present as the orange glass melt began to crystallize. These metal blebs have been used to infer the oxidationstate of the melt prior to eruption (5,6). The composition, abundance, and origin of the volcanic gas phase that was present during the orange glass fire-fountaining has yet to be definitively determined. Butler and Meyer (7) studied the surface coatings on orange and black beads and determined that sulfur was present in significant quantities as well as Cl and F. The coatings are assumed to represent the condensable portion of the gas that drove the firefountaining. Studying the orange glass and the oxidative behavior of the glass upon heating, (8) inferred that a self-reduction took place within the melt during ascent. Graphite was present in the source region of the orange glass (carbon in the form of graphite would be stable at depth in the moon if the oxygen fugacity in this region is close to the I-W buffer as is generally assumed). The C-O gas formed may have been the propellant for lunar pyroclastic eruptions. Fogel and Rutherford (9) studied magmatic volatiles in relation to Apollo 15 green and yellow glasses, looking for traces of carbon exsolution that would cause the lunar fire fountaining. They concluded that exsolution of C-O gas from the melt didn't cause the fire fountaining, but oxidation of graphite was sufficient. Weitz et al., (5) studied volatiles in the orange glass by examining the metal spherules found in the olivine phenocrysts and orange glass beads. The size, shape, location and composition of the blebs led them to conclude that the metal was formed by the oxidation of graphite and the corresponding reduction of iron, nickel and cobalt oxides from the melt. The oxidation-state indicated by the metal-melt equilibrium is IW 1.3 log units. At the orange glass liquidus (1320oC) and this oxygen fugacity, the carbon oxidation process would occur at 200 bars or about 4 km depth in the lunar crust. The implication is that production of the lunar volcanic gas Lunar and Planetary Science XXXIV (2003) 1322.pdf