Home
Research
Cool stars can form molecules in their atmospheres, and these molecules absorb entire chunks of the infrared spectra the stars produce.
The brightest cool stars are known as red supergiants and red giants. The most evolved stars in this group will die soon, ejecting their outer layers until nothing is left but their cores. In the process they will embed themselves in thick dust shells and disappear completely from the optical skies. Studying these stars before they make dust helps us to understand which molecules form, and since the dust is made from the molecules, we can better understand the way they die. Younger red giants are very bright and they are often used by astronomers as calibrators. But to use a star as a calibrator, you have to know what it is supposed to look like, at every wavelength, which means you have to understand which molecules form in their atmospheres.
At the other end of the brightness range we have brown dwarfs, which don't have enough mass to even be stars. These objects are sort of a cross between big Jupiters and tiny stars, and by looking at them in them in the infrared, we can see the molecules in their atmospheres and learn about their composition.
My interest in brown dwarfs began on an observing run in 1995 at the United Kingdom Infrared Telescope (UKIRT). The plan was to study something completely different, but my run was with Tom Geballe, then director of UKIRT, and he had just had a special request from Shri Kulkarni at Caltech. They had recently discovered Gliese 229B, had taken a preliminary infrared spectrum of it at Palomar, and wanted us to take a better spectrum of it at UKIRT before they announced their discovery. So we did. Geballe et al. (1996) published the resulting spectrum.
Spectra of brown dwarfs stand out because the stars are so cool, and molecules like methane (CH4) can survive in their atmospheres. The Sun is too hot for any molecules, but for lower mass stars, it's a different story. Not only did we verify the previous detection of methane, but we also found absorption from steam (H2O) as well.
The discovery of Gliese 229B was only the start. Since 1995, hundreds of brown dwarfs have been discovered. They even have their own spectral classes now. The warmer ones are known as L dwarfs, and the cooler ones are T dwarfs. Davy Kirkpatrick maintains an excellent archive of all known brown dwarfs. I last counted in November, 2012, when he had 918 L dwarfs, 355 T dwarfs, and 15 Y dwarfs listed. Y dwarfs weren't discovered until 2011. They are the coolest of the brown dwarfs, and we knew for years they had to be there, but they are so faint they're very difficult to detect.
My interest in brown dwarfs was renewed with the launch of the Spitzer Space Telescope in 2003. At Cornell, I worked as part of the Infrared Spectrograph team. One of the science teams we formed focused on brown dwarfs. Jeff Van Cleve christened us the Dim Suns team! (He likes Chinese food.) In 2004, we published our first results (Roellig et al. 2004). These were the first mid-infrared spectra ever of brown dwarfs and in them we discovered absorption from ammonia (NH3), another molecule too fragile to exist in normal stars.
Several papers followed. Cushing et al. (2006) looked at how more molecules can be detected in cooler dwarfs. The M and L dwarfs show only H2O absorption. T dwarfs are cooler and also have CH4 NH3 in their spectra. Mainzer et al. (2007) used higher-resolution spectra from Spitzer to look for evidence of clouds (which we didn't see) and evidence of vertical motion (like convection, which we did see).
Red giants are also cool stars, and they also have interesting molecules in their atmospheres. Red giants are bright and plentiful, and they have served as calibrators for infrared spectroscopy since the beginnings of the field in the late 1960s. Of course, if you are using a star as a standard star, your calibration is limited by how well you understand the spectrum of that star.
As part of the Infrared Spectrograph team at Cornell, one of my primary duties was to calibrate our data. I designed a program of observations to ensure that we had good standards and that we thoroughly understood their spectra. Our sample included 30 K giants, giving us the opportunity to study how the spectra changed as the stars got cooler. The dominant absorbers in the spectral regions observed by the IRS are SiO and OH. Sloan et al. (2015a) confirmed that the SiO bands get stronger as the stars get cooler, but unpredictably, making it unwise to assumem an infrared spectrum based on a star's spectral type. They also found OH bands in most of the K giant spectra. It's a testament to the quality of the IRS that we could observe them so clearly in our spectra. Both the SiO and OH bands are stronger than predicted by models, indicating that we need to better understand the physics of the outer atmospheres of these stars.
Late in the Spitzer mission, we observed a sample of 20 M giants with the IRS. We published a paper on these cool red giants in 2015 (Sloan et al. 2015b). The M giants are an interesting group. When most stars die, they are giants twice at the end of their life. The first time is after they run out of hydrogen in their core and are burning hydrogen in a shell around a helium core. Those stars are on the Red Giant Branch (RGB). When stars become giants the second time, that is when they die, ejecting their envelopes and embedding themselves in a dense shell of dust. These stars are on the Asymptotic Giant Branch (AGB). The K giants in our first 2015 paper are all RGB stars, but our sample of M giants includes both kinds of giants: RGB and AGB.
All of our M giants show absorption from SiO and OH like we see in the K giants, but almost all of our M giants also show absorption from water vapor (H2O). The stars with the strongest water bands are also the most variable stars in our sample, which means that they are the ones most likely to be AGB stars. Some of these more variable stars with the stronger water bands even show some dust in their spectra. It looks as though we are observing these stars just as they are starting to end their lives as stars.
Geballe, T.R., Kulkarni, S.R., Woodward, C.E., & Sloan, G.C. 1996, “The near-infrared spectrum of the recently discovered brown dwarf Gliese 229B,” ApJ Letters, 467, L101.
Roellig, T.L., et al. 2004, “Spitzer Infrared Spectrograph (IRS) Observations of M, L, and T dwarfs,” ApJ Supplement, 154, 418.
Cushing, M.C., et al. 2006, “A Spitzer Infrared Spectrograph (IRS) spectral sequence of M, L, and T dwarfs,” ApJ, 648, 614.
Mainzer, A.K., et al. 2007, “Moderate Resolution Spitzer Infrared Spectrograph (IRS) Observations of M, L, and T Dwarfs,” ApJ, 662, 1245.
Sloan et al. 2015a, “Spectral Calibration in the Mid-Infrared: Challenges and Solutions,” AJ, 149, 11.
Sloan et al. 2015b, “Infrared spectral properties of M giants,“ ApJ, submitted.
Last modified 18 May, 2016. © Gregory C. Sloan.