A SPITZER SURVEY OF NOVAE IN M31

We report the results of the first infrared survey of novae in M31. Both photometric and spectroscopic observations of a sample of 10 novae (M31N 2006-09c, 2006-10a, 2006-10b, 2006-11a, 2007-07f, 2007-08a, 2007-08d, 2007-10a, 2007-11d, and 2007-11e) were obtained with the Spitzer Space Telescope. The observations, which were obtained between similar to 3 and similar to 7 months after discovery, revealed evidence for dust formation in two of the novae: M31N 2006-10a and (possibly) 2007-07f, and [Ne II] 12.8 mu m line emission in a third (2007-11e). The Spitzer observations were supplemented with ground-based optical photometric and spectroscopic data that were used to determine the speed classes and spectroscopic types of the novae. After including data for dust-forming Galactic novae, we show that dust formation timescales are correlated with nova speed class in that dust typically forms earlier in faster novae. Our failure to detect the signature of dust formation in most of our M31 sample is likely a result of the relatively long delay between nova eruption and our Spitzer observations. The two novae for which we found evidence of dust formation were the two "slowest" novae in our sample. Finally, as expected, we found that the majority of the novae in our sample belong to the Fe II spectroscopic class, with only one clear example of the He/N class (M31N 2006-10b). Typical of an He/N system, M31N 2006-10b was the fastest nova in our sample, not detected with the IRS, and just barely detected in three of the IRAC bands when it was observed similar to 4 months after eruption.