ABSTRACT
The lunar picritic volcanic glasses have been identified as being quenched samples of primary magmas extruded onto the lunar surface via fire-fountains. The study of the composition of these glasses is of extreme importance for the understanding of the Moon's mantle composition and petrogenetic processes.
Based on their chemical signature (low-Ti and high Mg#) and physical characteristics, the lunar picritic glasses are believed to represent primary magmas. Experimental data suggest that these melts formed at 360-500 km depth (18-25 kbar) in the lunar mantle, and were transported onto the lunar surface through a "channel" network (McKenzie, 1985b) that isolated these magmas and minimized fractionation. The aim of this work is to obtain high precision electron microprobe analysis in order to more confidently model melting processes that may have occurred deep within the lunar mantle.
For this study, Apollo 14 volcanic green glasses A and B were analyzed. These glasses (A and B) show large enrichments of incompatible elements (e.g. K, Na, Ti) and an intragroup trend that does not follow olivine (olivine is known to be the liquidus phase for all pressures less than about 20 kbar), low-Ca pyroxene, augite or plagioclase fractionation trends. Although batch melting, mantle source hybridization and/or assimilation of KREEP in the magma source region have been considered, these processes fail to explain the behavior of major elements.
More recently, it has been suggested (Delano, 1996; Delano and Fernandes, 1998) that the trends observed reflect deep-seated magmatic processes resulting from dynamic melting of an ascending mantle diapir (Delano and Fernandes, 1998). This model, the dynamic partial melting process, involves the differential flow of melt and residual matrix (Eggins, 1992). In the melt region, the degree of melting will increase as a function of decompression, therefore of height above the adiabat and peridotite (source) solidus intersection. The amount of melt present (i.e. porosity) at a particular height, however will be less than the degree of melting due to the more rapid buoyancy-driven ascent of melt compared to the matrix. The magnitude of this melt-filled porosity (phi) will depend upon the velocity of the melt relative to the matrix and upon the rate of melting. The amount of melt created will dictate the abundance of elements in a specific melt fraction.
Based on dynamic melting modeling, the degree of partial melting involved in the origin of these melts (the glasses) is within the range of values (5 to 25%) expected for conventional petrogenetic processes (e.g. batch melting). The model results for Al, Ca, K, Na, and Ti show a wide range of porosity and distribution coefficients that may have been involved in the origin of these melts. The little sensitivity to the model shown by K and Na suggest that these two elements are important in the identification of the degree of partial melting undergone by the source mantle diapir. These two elements may provide important information that will allow all of the other elements, including Ti, to be better constrained in the model.

Vera Assis Fernandes, V.A., 1999. Major- and minor-element analysis of Apollo 14 volcanic green glasses B, and petrogenic modeling of Apollo 14 green glasses A and B. Unpublished MSc. thesis, State University of New York at Albany. 147 pp., +xi
University at Albany Science Library call number:  SCIENCE Oversize (*) QE 40 Z899 1999 F47

Return to MS Theses completed in the Geological Sciences Program, University at Albany