The lifetimes of stars are many millions of years, much too long for us to study just a few stars and learn about their lives. Recall that, instead, we set out to study many stars at different stages of their lives and from that information, piece together the life story of the stars. We started our discussion in ch. 17, where the H-R diagram was introduced. This was expanded on in ch. 20, where we followed the progress of a protostar on the H-R diagram. We will continue to follow the stars through their lives, and we will chart their progress with the H-R diagram.
Once a protostar evolves to the main sequence and becomes a true star, it begins generating energy by fusing hydrogen. This process can occur for a very long time, typically billions (or at least millions) of years. During this time, very little mass actually "disappears", only 0.7 percent of the mass of hydrogen is converted to energy. The mass remains rather constant during this period.
The luminosity and temperature of a star do change as hydrogen is converted to helium. We plot these changes on the H-R diagram, starting from the left edge of the main sequence band. This left edge is known as the zero-age line. As the amount of helium increases, the temperature and density (pressure) in the core also increase. The increased core temperature and density result in an increased rate of hydrogen fusion (the rate of fusion depends on the temperature and density). An increase in the rate of fusion increases the amount of energy produced, and the star's luminosity will rise.
The time over which these changes occur is slow, but they depend on the mass of the star. Look at Figure 21.4. A star like our Sun barely moves from the zero-age line after 7 billion years. A star that is 15 Msun starts to move substantially away from the zero-age line (and the main sequence band) after only 10 million years. Once again we see that more massive stars evolve more quickly.
The lifetimes on the main sequence for stars of different masses is displayed in Table 21.1. We see that more massive stars have shorter lives as main sequence stars. A more massive star has more hydrogen to fuse, so this may seem counter-intuitive. But, though they have more hydrogen, more massive stars will also have higher core temperatures and densities, and will therefore fuse hydrogen at a higher rate. The rate of hydrogen fusion increases as the fourth power of the core temperature, so doubling the temperature will increase the rate of fusion by 24=16 times. Calculations show that the core temperature increases approximately as the mass, so the rate of fusion increases faster than the increase of mass. Therefore, more massive stars have shorter lives on the main sequence.
Eventually, all the hydrogen in the core is converted to helium. Hydrogen fusion in the core stops. Helium fusion requires higher temperatures, so it can't immediately take over. Instead, without a source of energy to support it, the core contracts. This heats the core and the surrounding regions of the star. The surrounding regions still contain a significant amount of hydrogen, and it will eventually be heated to a temperature where it can begin fusing.
This hydrogen is not in the core. Fusion is now occuring in a layer outside the core of the star. The result of this change is rather dramatic. The star has to readjust itself to this new configuration. Fusion just outside the core can occur more rapidly, producing more luminosity than the fusion that occured in the core. The increased energy outflow heats the outer layers of the star, causing them to expand dramatically in size, and the surface cools down. The star becomes a red giant.
Table 21.2 compares our Sun to the red giant star Betelgeuse in the constellation of Orion. If Betelgeuse were at the center of our solar system, all the planets out to Jupiter would lie within the star (and be consumed by it!). Figure 21.3 shows a Hubble picture of the surface of Betelgeuse. (I said that all stars are just points in our telescopes, well I lied, but just a little.) We can measure the size of Betelgeuse directly. The diameter of a typical red giant is larger than the diameter of the orbit of Mars.
The evolutionary tracks shown in Figure 21.4 are from computer simulations.
In Ch. 20 we saw that molecular clouds can give rise to a number of stars in a relatively short period of time. The result is a star cluster, a close-knit group of stars of approximately the same age. Since the stars in a cluster have different masses, we can use them to test our theory of how stars evolve with age.
That is, within a given cluster, we may find high mass stars that have evolved to become red giants and low mass stars still in their main sequence phase.
There are three types of clusters:
Globular clusters contain tens of thousands to millions of stars. They are located mostly in the halo and nuclear bulge of our galaxy. About 150 globular clusters are known in the Milky Way. Most of them are nearly spherical in shape.
Open clusters contain 50 to 1000 stars. They look quite different from globular clusters, more open and less dense. They are located throughout the disk of the galaxy. There are thousands of open clusters in our galaxy.
Stellar associations contain 5 to 50 extremely hot stars (upper left corner of the H-R diagram) plus other less luminous stars. According to our theory, the hot stars have short lifetimes (1 to 10 million years), and therefore the associations should be very young.
If stellar associations are younger than open clusters which are younger than globular clusters, then the locations of the stars on an H-R diagram for each type should look different. The differences are a result of the different ages.
Figure 21.8 shows the prediction for an H-R plot for a 3 million year old stellar association. What you see is that the more massive stars (upper half of the plot) are on the main sequence (blue line) while the less massive stars are off the main sequence. This is because the less massive stars are still in the protostar stage. They take a longer time to reach the main sequence. (We know this because this is a simulation.)
Now compare this to Figure 21.10, the H-R diagram for the stars in NGC 2264 (shown in Figure 21.9), a stellar association believed to be about 3 million years old. The general features are very similar to the prediction. The number of heavy and light stars in a given association is random, so that doesn't have to look the same as the simulation, but the positions where stars can lie on the H-R diagram agrees.
We can do the same for open clusters. Figure 21.12 shows the H-R plot for M41 (Figure 21.11). This cluster is believed to be older than the association shown earlier. This cluster has mostly high mass stars, heavier than the Sun. On the H-R diagram, the lower mass stars are on the main sequence and the higher mass stars are off the main sequence. This would agree with our prediction that higher mass stars have relatively short lives on the main sequence before evolving into red giants. A few of the highest mass stars are already in the red giant phase.
Now we can look at what will happen in an older globular cluster. Figure 21.13 shows the prediction for a cluster at 4.24 billion years. Figure 21.14 shows the measurements for the globular cluster 47 Tucanae (Opening figure of Ch. 18). Note that the horizontal and vertical scales have changed. The basic resemblance between the simulation and the measurements is clear. The low mass stars -- down to the faintest stars -- are on the main sequence. The high mass stars have moved off the main sequence in a rather characteristic pattern.
Everything discussed until now applies to all stars. They start as balls of gas that contract to protostars, heat up, begin hydrogen fusion, and when the hydrogen in the core is depleted, the core begins to contract, hydrogen fusion begins outside the core, and the star expands to a giant. After this, the paths of stars differ depending on their mass. We'll go through some of the possible paths.
First we look at stars with initial masses less than about 3Msun. In these stars, the helium core continues to collapse until the temperature reaches 100 million K. At this temperature it becomes possible for 3 helium nuclei to join to make a carbon nucleus (4He has 2 protons and 2 neutrons, 12C has 6 protons and 6 neutrons). This process occurs rapidly in what is known as a helium flash.
It is a coincidence of nature that this process involves 3 helium nuclei and produces carbon. The fusion of 2 helium nuclei produces beryllium-8 (8Be), and unstable nucleus with a lifetime of about 10-16 seconds. Lucky for us carbon-based life forms, since beryllium is a toxic metal!
The star readjusts to the new conditions again, becoming less luminous (see points B and C in Figure 21.15). The star soon burns all the helium in the core, converting it to carbon and oxygen. The core contracts again, heating up the helium outside the core. This helium is soon able to fuse. The star goes through several similar oscillations of contraction and more fusion.
The star now has a structure like an onion: a core of carbon and oxygen; a layer surrounding that core where helium is fusing into more carbon and oxygen; and somewhere beyond that a layer where hydrogen is fusing to form more helium. For light stars (like our Sun), this is as far as fusion can go. The star cannot get hot enough to go beyond helium fusion producing carbon and oxygen. We'll discuss the death of such stars in chapter 22.
The H-R diagrams for older clusters give us confidence in the calculations that lead to this picture. As predicted, they show that larger stars have moved off the main sequence to become giants while smaller stars are still on the main sequence.
Recall that all this fusion occurs in just the core of the star. The outer part of the star doesn't participate in fusion.
When stars become giants, they can lose a substantial amount of material into space. While the mechanism isn't completely understood, it is observed as objects known as planetary nebulae (which has nothing to do with planets). These objects can be quite beautiful (Figure 21.16).
The material lost by aging stars is returned to interstellar space where it can be incorporated into new objects -- stars and planets. The star began its life as mostly hydrogen and helium, but the material it returns to space will be a mixture containing the old hydrogen and helium plus heavier atoms, such as carbon and oxygen, formed in the star. And, as we'll learn in coming chapters, the more massive the star, the greater and heavier are the atoms that it puts out.
Our theory for the beginning of the universe (the Big Bang, or what is now called the Standard Model of Cosmology) says that the universe began with Hydrogen, Helium, and a little bit of Lithium (the first, second, and third elements in the periodic table). All the heavier elements are synthesized in stars! Lower mass stars can produce Carbon and Oxygen. Higher mass stars are required to produce the heavier elements.
Higher mass stars evolve more quickly than the lower mass stars. Their evolution up to the formation of a Carbon-Oxygen core is roughly the same as for lower mass stars, so we'll pick up from there. (Note differences: slower helium burning without the helium flash, and larger size, dubbed supergiants.)
A star larger than about 8 solar masses can fuse carbon to form oxygen, neon, sodium, magnesium, and silicon. In fact, physicists have worked out the reactions that allow the production of virtually all the elements up to iron. This process of building larger nuclei from smaller nuclei is called nucleosynthesis.
Careful observations of spectral lines show that our Sun, nearby stars, and stars in open clusters have between 1 and 4 percent of elements heavier than helium (by mass). The amount of heavier elements in globular clusters is 1/10 to 1/100 this amount. This makes sense if the stars in globular clusters are first generation, and therefore contain mostly the hydrogen and helium initially available. As these stars age, they eject material back into space that contains heavier elements. Second (and later) generation stars will use this material, and therefore contain a higher percentage of heavy elements.
Note that this implies that first generation stars (called Generation III stars by astronomers for some unknown reason) couldn't have formed with a planet like Earth, composed primarily of heavy elements. Later generations of stars will presumably contain even greater fractions of heavy elements.
Stars spend most of their lives (about 90%) on the main sequence. Their time off the main sequence is relatively short. The fusion of heavy elements that takes place when a star moves off the main sequence occurs at such high temperatures that once started, the fuel is rapidly used up. Some of the most dramatic events in the universe occur when a star finally dies, and this is the topic of the next chapter.