Size-selective concentration of particles in a weakly turbulent protoplanetary nebula may be responsible for the initial collection of chondrules and other constituents into primitive body precursors. This paper presents the main elements of this process of turbulent concentration. In the terrestrial planet region, both the characteristic size and size distribution of chondrules are explained. "Fluffier" particles would be concentrated in nebula regions which were at a lower gas density and/or more intensely turbulent. The spatial distribution of concentrated particle density obeys multifractal scaling, suggesting a close tie to the turbulent cascade process. This scaling behavior allows predictions of the probability distributions for concentration in the protoplanetary nebula to be made. Large concentration factors (> 10 5) are readily obtained, implying that numerous zones of particle density significantly exceeding the gas density could exist. If most of the available solids were actually in chondrule sized particles, the ensuing particle mass density would become so large that the feedback effects on gas turbulence due to mass loading could no longer be neglected. This paper describes the process, presenting its basic elements and some implications, without including the effects of mass loading.

No abstract

In many Type B Allende inclusions, melilite is reversely-zoned over restricted portions of each crystal. Textural relationships and the results of dynamic crystallization experiments suggest that the reverselyzoned intervals in these Type B melilites result from the co-precipitation of melilite with clinopyroxene from a melt, prior to the onset of anorthite precipitation. When clinopyroxene begins to precipitate, the All Mg ratio of the melt rises, causing the crystallizing melilite to become more gehlenitic, an effect which is negated by crystallization of anorthite. Because the equilibrium crystallization sequence in these liquids is anorthite before pyroxene, melilite reverse zoning can occur only when anorthite nucleation is suppressed relative to pyroxene. This has been achieved in our experiments at cooling rates as low as 0.5°C/hour. Our experiments further indicate, however, that reverse zoning does not form at cooling rates ~50°C/hour, probably because the clinopyroxene becomes too Al-rich to drive up the Al/Mg ratio of the liquid. Type B inclusions with reversely-zoned melilites must have cooled at rates greater than those at which anorthite begins to crystallize before clinopyroxene but <50°C/hour. Such rates are far too slow for the Type B droplets to have cooled by radiation into a nebular gas but are much faster than the cooling rate of the solar nebula itself. One possibility is that Type B's formed in local hot regions within the nebula, where their cooling rate was equal to that of their surrounding gas. Other possibilities are that their cooling rates reflect their movement along nebular temperature gradients or the influence of a heat source. The sun or viscous drag on inclusions as they moved through the nebular gas are potential candidates for such heat sources.

No abstract

A new titanium‐bearing calcium aluminosilicate mineral has been identified in coarse‐grained calcium‐aluminum‐rich inclusions (CAIs) from carbonaceous chondrites. The formula for this phase, which we have temporarily termed “UNK,” is Ca3Ti(Al, Ti)2(Si, Al)3O14, and it is present in at least 8 of the 20 coarse‐grained CAIs from the Allende CV3 chondrite examined as part of this project. The phase occurs in Types A and B1 inclusions as small tabular crystals oriented along two mutually perpendicular planes in melilite. UNK crystallizes from melts in dynamic crystallization experiments conducted in air from four bulk compositions modeled after Types A, B1, B2 and C inclusions. Cooling rates resulting in crystallization of UNK ranged from 0.5 to 200 °C/h from maximum (initial) temperatures of 1375 to 1580 °C. Only below 1190 °C does UNK itself begin to crystallize. To first order, the presence or absence of UNK from individual experiments can be understood in terms of the compositions of residual melts and nucleation probabilities. Compositions of synthetic and meteoritic UNK are very similar in terms of major oxides, differing only in the small amounts of trivalent Ti (7–13% of total Ti) in meteoritic samples. UNK crystallized from the Type A analog is similar texturally to that found in CAIs, although glass, which is typically associated with synthetic UNK, is not observed in meteoritic occurrences. A low Ti end‐member of UNK (“Si‐UNK”) with a composition near that of Ca3Al2Si4O14 was produced in a few samples from the Type B1 analog. This phase has not been found in the meteoritic inclusions.