Stellar Spectral Types
Published on Feb 6, 2007 at 12:13 am.
6 Comments.
Filed under stars.
If you read Sky and Telescope or Astronomy or any other astronomy magazine, you will occasionally see some reference to “G Type stars” or perhaps “B Type stars.” What do these letters mean?
Well, to answer that we go back over a century to the the first studies of stellar spectral. When the light from a star is passed through a prism (or a diffraction grating), the light is split up into its individual colors. This range of colors is the spectrum of the star. Now, if the light is from a point source, or a narrow slit, then you can notice that certain colors are missing, or at least dimmer, in the spectrum. This was done with sunlight long before it was done with stars by Joseph von Fraunhofer, and so these missing colors are called Fraunhofer lines. It was quickly realized that these lines corresponded to spectral lines of chemical elements. So, studies of the spectrum of the Sun could lead to an understanding of its chemical composition. But stars are dim, and when you spread out the light, it is even dimmer! So, it took a number of years before stellar spectra could be studied. By the 1860’s, though, Angelo Secchi had measured the spectra of a number of different stars. He collected these spectra into five categories based upon the spectral lines that were prominently seen in their spectra.
Secchi’s five spectral types turned out to be too limiting. Later research showed that the main composition of stars was hydrogen. So, by the end of the 19th Century, Williamina Fleming had expanded spectral classification to many more types. She assigned letters A through P to stars, based on the prominence of hydrogen in the spectrum. Later, Annie Jump Cannon realized that many of these types were redundant, and then reordered the remaining ones by temperature. In order from hottest to coolest, she had types:
O B A F G K M
Later, each type was further subdivided into ten finer divisions, with a letter ranging from a 0 to 9. So, A Type stars range from A0, A1, A2, and so forth to A9. Each division is about equally spaced, so one step past A9 is F0, then F1, and so forth. On this system, the Sun is a Type G2 star. A type G1 is slightly hotter, and a type G3 is slightly cooler. A type F9 is a bit hotter, and a type G6 is a bit cooler.
Various subtypes eventually were added. C stars, for stars that had a surplus of carbon in the spectrum. R, N, and S stars were added, with other spectral features. However, all of these stars also had the same basic temperature and dominant spectral characteristics of K and M stars, particular in hydrogen spectral characteristics. They differed in the other spectral lines, due to different chemistries. However, these other chemicals really constitute a tiny percentage of the star, so they could be thought of as special subtypes of K and M stars, leaving the main sequence of letters based on temperature unchanged.
In the 1990’s, though, the technology advanced to the point that a number of very dim cool stars were being found. On the whole, for most stars, the cooler they are, the dimmer they are. So, these extra cool stars were extra dim. They had been there all along, but they simply were too dim to see. So, two new spectral types were added: L and T. These are extremely cool stars. In fact, some of them are “failed” stars — stars that form with too little mass to initiate nuclear fusion. We call these brown dwarfs. So, the sequence of spectral types is now:
O B A F G K M L T
The basic spectral type makes no statement, really, about the star other than its temperature. So, brown dwarfs are spread out among the L and T stars (though most T stars are brown dwarfs). Nor, are all M stars the same. They range from tiny red dwarf stars, physically not much larger than Jupiter (though with a lot more mass) to supergiant stars so large that would encompass the entire inner Solar System out to near the orbit of Jupiter were they to be placed where the Sun is. So, in the 1940’s Phillip Keenan and William Morgan proposed another spectral classification system that included this size effect. This system, the Morgan-Keenan system, or Yerkes system, is the Luminosity Class of the star. It is based upon the luminosity of the star.
The luminosity classes are: Type I (supergiants), Type II (bright giants), Type III (giants), Type IV (subgiants), and Type V (dwarfs). It turns out that the majority of stars are Type V, and these are called main sequence stars. There are other luminosity classes, though: Type 0 (hypergiants), Type VI (subdwarfs), and Type VII (white dwarfs). The hypergiants are a fairly recent addition. Usually white dwarfs get their own WD classification, so Type VII is seldom used.
The normal way of referring to the spectral type of a star is to combine these two systems. For example, the Sun is a G2 V star. Betelgeuse is a M2 Ia (the supergiants, and giants, have subclasses, so supergiants can be Ia, Iab, or Ic). Capella is a G6 III, and Polaris is about an F7 Ib (note: Polaris is a variable star, so this changes a bit).
So, the next time that you are reading about stars, and they say that a star is some particular type, this is what they mean. Of course, astronomers add on to the spectral classification system, and there are a whole host of letters that can be added after the spectral type to denote specific spectral features, so what I’ve given here is just a taste of what is really used.
-Astroprof
Astroprof’s Page » Canopus on February 16, 2007 at 10:48 am: 1
[…] But, what type of star is Canopus that it shines so brightly? Long time readers will recall that I normally post some physical data about these stars when I do a star post. However, reliable data is tough to get on Canopus. You’d think that such a bright star would be exceedingly well studied. Well, it is well studied, but that doesn’t mean that we understand it well! Canopus is a distended star that has evolved off of the main sequence. Normally such stars are cool red giants or supergiants. But, Canopus is hot — nearly 7500K, in fact. Stars of this type are very rare, and very poorly understood. Looking through tables of data on Canopus I find is classified in many different ways, from spectral type A9 to F2. F0 seems about the most common. I have also seen it categorized as a bright giant (luminosity class II), a low luminosity class supergiant (Ib), an intermediate luminosity supergiant (Iab), and even a bright supergiant (Ia). What gives? Well, it turns out that at this brightness and temperature, there is very little distinguishing these different types of luminosity classes, so there is a lot of interpretation that goes on in classifying them. For cooler stars, spectral types K and M, the distinctions are far greater, and we can generally agree on them. But, Canopus is at at temperature where all the types sort of run together, and Canopus is such a rare type of star that we really don’t know how to tell the different types apart well in this temperature and luminosity range. This type of star is perhaps evolving either towards or away from being a red giant (or red supergiant). About all we can be sure of is that Canopus is not on the main sequence, so it is a dying star of some sort, but we can’t even tell at what stage in its dying it happens to be. We can most all also agree, though, that Canopus is much hotter than the Sun, vastly brighter, and much larger. It is likely about 75 times the diameter of the Sun. We really don’t know much about this type of star, so it is hard to say what its mass is, but I’d hazard a guess that it may be only a few times the mass of the Sun. Canopus does not have enough mass to become a supernova. It will still die and yield a white dwarf, but it will likely produce a very massive white dwarf. Canopus likely has sufficient mass to begin to fuse carbon into oxygen and neon (Main sequence stars fuse hydrogen into helium, and they begin to die when they start to run out of hydrogen to fuse. They then fuse helium into carbon. Most stars stop with fusing helium into carbon, but the really big ones can fuse the carbon into heavier elements). […]
Jeff Murdoch on February 10, 2008 at 10:06 pm: 2
I’ve run across a spectral classification not mentioned above:
“M2 D”.
This classification note of “D” leads me to think it means “dwarf” or equivalent to “Type V” or possibly “Type VI”.
Can you confirm?
Thanks!
Astroprof on February 11, 2008 at 12:40 am: 3
Hmm. Type “D” is usually reserved for white dwarf stars. None of them should be as cool as an M2, though. So, I suspect that it may have been type V or VI as you suggest. I’d have to see the classification system that you mention in context to be sure.
none on April 23, 2008 at 5:02 pm: 4
I think you should actually tell what temperature the letters represent. But, I learned a lot from this site; you rock, Astroprof!
Stephen Brandon on January 26, 2010 at 3:26 pm: 5
How are elements with a greater atomic weight than oxygen formed if most stars can’t fuse past carbon?
Astroprof on January 27, 2010 at 10:07 am: 6
Stars can fuse past carbon, but only the higher mass stars. Stars somewhat higher mass than the Sun can fuse to neon. Stars of very high mass fuse to iron, and then all the rest of the elements are created in the supernova explosion as they die.