THE EVOLUTION OF DUST MASS IN THE EJECTA OF SN 1987A

We present a new analysis of the infrared (IR) emission from the ejecta of SN 1987A covering days 615, 775, 1144, 8515, and 9090 after the explosion. We show that the observations are consistent with the rapid formation of about 0.4 M-circle dot of dust, consisting of mostly silicates (MgS103), near day 615, and evolving to about 0.45 M. of composite dust grains consisting of similar to 0.4 M-circle dot of silicates and 0.05 M-circle dot of amorphous carbon after d circle dot ay similar to 8500. The proposed scenario challenges previous claims that dust in supernova (SN) ejecta is predominantly carbon, and that it grew from an initial mass of similar to 10(-3) M-circle dot, to over 0.5 M-circle dot by cold accretion. It alleviates several problems with previous interpretations of the data: (1) it reconciles the abundances of silicon, magnesium, and carbon with the upper limits imposed by nucleosynthesis calculations, (2) it eliminates the requirement that most of the dust observed around day 9000 grew by cold accretion onto the similar to 10(-3)M(circle dot) of dust previously inferred for days 615 and 775 after the explosion, and (3) establishes the dominance of silicate over carbon dust in the SN ejecta. At early epochs, the IR luminosit% of the dust is powered by the radioactive decay of Co-56, and at late times by at least (1.3-1.6) x 10(-4)M(circle dot) of Ti-44.