CONDENSATION IN THE SOLAR NEBULA: EFFECTS OF PARTIAL ISO- LATION OF CONDENSATES FROM THE RESIDUAL GASES; Michail

The equilibrium condensation sequence is the tool universally employed to understand the effects of temperature excursions on protoplanetary materials in the solar nebula. Thermodynamic calculation of the details of condensation in a system of cosmic composition [e.g., 1-31 reveal the appearance with falling temperature of refractory minerals, then (partly by reaction with earlier minerals) olivine, pyroxene, and plagioclase. Most Fe condenses initially as Fe,Ni metal, which reacts with the gas at lower temperatures to produce troilite and ferrous and ferric iron incorporated into silicates and oxides. At temperatures c-400 K silicates react with water vapor to form phyllosilicates. All these minerals are present in differing proportions in the various classes of chondrites. The inconsistency of all equilibrium condensation models with nature is that the meteoritic high temperature assemblage of refractory condensates and the lower temperature suite of magnesian silicates and plagioclase are not in equilibrium with one other. This discrepancy is so obvious that it was immediately suggested [1,2] that the high temperature condensates in carbonaceous chondrites had somehow become isolated from the residual gas. Beginning with that necessity, some degree of isolation of condensates from the residual gas or, in other words, incomplete equilibrium between gas and solids, became implicit in all models of chondrite formation. The systematic withdrawal or isolation of some fraction of the condensates means that the bulk chemical composition of the system still in reactive contact changes as condensation proceeds. The degree to which the change of bulk composition affects the condensation sequence depends upon the rate of isolation of condensing material, and how it varies with declining temperature. Only the two end-member cases have been studied: zero isolation of condensate (the classic condensation sequence [I-3]), and removal of all condensate as fast as it forms (the "inhomogeneous accretion model" [5,6]) . However, simple consideration of chondrite mineralogy shows that both cases are unrealistic. Under complete equilibrium, high temperature condensates must disappear completely by the time of condensation of ferromagnesian silicates and plagioclase. On the other hand the complete isolation of condensates as they form would leave no A1 to form plagioclase and no Fe to form troilite. As in so many issues, the truth must lie somewhere in between. The objective of the present study is to explore how the partial isolation of condensates would affect the condensation sequence. Model. To study the effects of partial isolation of solids on the condensation sequence we assume that as condensation proceeds a specified fraction of the existing condensate is withdrawn from reactive contact with the residual gas (e.g. , by being enveloped in a layer of later condensate). This changes the composition of the remaining reactive system. We define the isolation degree 5 to be the relative amount of condensed matter (% of the total amount present) that is withdrawn from the reactive system in a given computational step (temperature interval). We expect the isolation degree to be a complex function of temperature, the amount of solids condensed, etc., and it may vary from mineral to mineral. In this study, however, we explore the simplest case where the isolation degree is the same for all condensates at all temperatures. A related parameter is the isolation rate (dMildT), which defines the absolute amount of inert solids, Mi, that is withdrawn in each temperature interval (AT=-1 K). At any given temperature the isolation rate equals the isolation degree multiplied by the total amount of reactive solids in the system. In our calculations we employed the PHEQ condensation code developed by [3]. To account for withdrawal of condensates, the mass-balance equations were modified so that after each computational step, which adds condensates produced in a 1 K temperature interval, a percentage 6 of all the elements present as condensate is subtracted from the total system composition. Our preliminary calculations assumed an initial sys-