We have used the Spitzer satellite to monitor the mid-IR evolution of SN 1987A over a 5 year period spanning the epochs between days \sim 6000 and 8000 since the explosion .
The supernova ( SN ) has evolved into a supernova remnant ( SNR ) and its radiative output is dominated by the interaction of the SN blast wave with the pre-existing equatorial ring ( ER ) .
The mid-IR spectrum is dominated by emission from \sim 180 K silicate dust , collisionally-heated by the hot X-ray emitting gas with a temperature and density of \sim 5 \times 10 ^ { 6 } K and \sim 3 \times 10 ^ { 4 } cm ^ { -3 } , respectively .
The mass of the radiating dust is \sim 1.2 \times 10 ^ { -6 } M _ { \odot } on day 7554 , and scales linearly with IR flux .
Comparison of the IR data with the soft X-ray flux derived from Chandra observations shows that the IR-to-Xray flux ratio , IRX , is roughly constant with a value of 2.5 .
Gas-grain collisions therefore dominate the cooling of the shocked gas .
The constancy of IRX is most consistent with the scenario that very little grain processing or gas cooling have occurred throughout this epoch .
The shape of the dust spectrum remained unchanged during the observations while the total flux increased by a factor of \sim 5 with a time dependence of t ^ { \prime 0.87 \pm 0.20 } , t ^ { \prime } being the time since the first encounter between the blast wave and the ER .
These observations are consistent with the transitioning of the blast wave from free expansion to a Sedov phase as it propagates into the main body of the ER , as also suggested by X-ray observations .
The constant spectral shape of the IR emission provides strong constraints on the density and temperature of the shocked gas in which the interaction takes place .
Silicate grains , with radii of \sim 0.2 \mu m and temperature of T \sim 180 K , best fit the spectral and temporal evolution of the \sim 8 - 30 \mu m data .
The IR spectra also shows the presence of a secondary population of very small , hot ( T \gtrsim 350 K ) , featureless dust .
If these grains spatially coexist with the silicates , then they must have shorter lifetimes .
The data show slightly different rates of increase of their respective fluxes , lending some support to this hypothesis .
However , the origin of this emission component and the exact nature of its relation to the silicate emission is still a major unsolved puzzle .