Let's compare a supernova with a nova. Like a nova, a supernova is a star that suddenly increases dramatically in brightness, then slowly dims again, eventually fading from view. The exploding star is commonly called the supernova's progenitor. In some cases, supernovae light curves can appear quite similar to those of novae, so a distant supernova can look a lot like a nearby nova—so much so, in fact, that the difference between the two was not fully appreciated until the 1920s. But novae and supernovae are now known to be very different phenomena. Supernovae are much more energetic events, driven by very different underlying physical processes*.
(Note that when discussing novae and supernovae, astronomers tend to blur the distinction between the observed event (the sudden appearance and brightening of an object in the sky) and the process responsible for it (a violent explosion in or on a star). The two terms can have either meaning, depending on context.)
Well before they understood the causes of either novae or supernovae, astronomers knew of clear observational differences between them. The most important of these is that a supernova is more than a million times brighter than a nova. A supernova produces a burst of light billions of times brighter than the sun, reaching that brightness within just a few hours after the start of the outburst. The total amount of electromagnetic energy radiated by a supernova during the few months it takes to brighten and fade away is roughly 1043 J—nearly as much energy as the Sun will radiate during its entire 1010-year lifetime! (Enormous as this energy is, however, it pales in comparison with the energy emitted in the form of neutrinos, which may be 100 times greater.)
A second important difference is that the same star may become a nova many times, but a star can become a supernova only once. This fact was unexplained before astronomers knew the precise nature of novae and supernovae, but it is easily understood now that we understand how and why these explosions occur. The nova accretion—explosion cycle described earlier can take place over and over again, but a supernova destroys the star involved, with no possibility of a repeat performance.
In addition to the distinction between novae and supernovae, there are also important observational differences among supernovae. Some supernovae contain very little hydrogen, according to their spectra, whereas others contain a lot. Also, the light curves of the hydrogen-poor supernovae are qualitatively different from those of the hydrogen-rich ones, as illustrated in Figure 21.7. Based on these observations, astronomers divide supernovae into two classes, known simply as Type I and Type II. Type I supernovae, the hydrogen-poor kind, have a light curve somewhat similar in shape to that of typical novae. Type II supernovae, whose spectra show lots of hydrogen, usually have a characteristic "plateau" in the light curve a few months after the maximum. Observed supernovae are divided, roughly, equally between these two categories.
Figure 21.7 The light curves of typical Type I and Type II supernovae. In both cases the maximum brightness or intensity can reach nearly 10 billion Suns, but there are characteristic differences in the falloff of the luminosity after the initial peak. Type I light curves somewhat resemble those of novae (Figure 21.4). Type II curves have a characteristic bump in the declining phase.
What is responsible for these differences among supernovae? Is there more than one way in which a supernova explosion can occur? The answer is yes. To understand the alternative supernova mechanism, we must return to the processes that cause novae and consider the long-term consequences of their accretion—explosion cycle.
Nova explosions eject matter from a white dwarf's surface, but they do not necessarily expel or burn all the material that has accumulated since the last outburst. In other words, there is a tendency for the dwarf's mass to increase slowly with each new nova cycle. As its mass grows and the internal pressure required to support its weight rises, the white dwarf can enter into a new period of instability—with disastrous consequences.
Recall that a white dwarf is held up not by thermal pressure (heat) but by the degeneracy pressure of electrons that have been squeezed so close together that they have effectively come into contact with one another. 
(Sec. 20.3) However, there is a limit to the pressure that these electrons can exert. Consequently, there is a limit to the mass of a white dwarf, above which electrons cannot provide the pressure needed to support the star. Detailed calculations show that the maximum mass of a white dwarf is about 1.4 solar masses, a mass often called the Chandrasekhar mass, after the Indian astronomer Subramanyan Chandrasekhar, whose work in theoretical astrophysics earned him a Nobel Prize in physics in 1983.
If an accreting white dwarf exceeds the Chandrasekhar mass, the pressure of degenerate electrons in its interior becomes unable to withstand the pull of gravity, and the star immediately starts to collapse. Its internal temperature rapidly rises to the point at which carbon can fuse into heavier elements. Carbon fusion begins everywhere throughout the white dwarf almost simultaneously, and the entire star explodes in another type of supernova—a so-called carbon-detonation supernova—comparable in violence to the "implosion" supernova associated with the death of a high-mass star, but very different in cause. In an alternative and (some astronomers think) possibly more common scenario, two white dwarfs in a binary system may collide and merge to form a massive, unstable star. The end result is the same—a carbon-detonation supernova.
We can now understand the differences between Type I and Type II supernovae. The explosion resulting from the detonation of a carbon white dwarf, the descendant of a low-mass star, is a supernova of Type I. Because this conflagration stems from a system containing virtually no hydrogen, we can readily see why the spectrum of a Type I supernova shows little evidence of that element. The appearance of the light curve (as we will soon see) results almost entirely from the radioactive decay of unstable heavy elements produced in the explosion itself.
The implosion—explosion of the core of a massive star, described earlier, produces a Type II supernova. Detailed computer models indicate that the characteristic shape of the Type II light curve is just what would be expected from the expansion and cooling of the star's outer envelope as it is blown into space by the shock wave sweeping up from below. The expanding material consists mainly of unburned gas—hydrogen and helium—so it is not surprising that those elements are strongly represented in the supernova's observed spectrum. (See Interlude 21-1 for an account of a recent Type II supernova that confirmed many basic theoretical predictions while also forcing astronomers to revise the details of their models.)
Figure 21.8 summarizes the processes responsible for the two different types of supernovae. We emphasize that, despite the similarity in the total amounts of energy involved, Type I and Type II supernovae are unrelated to one another. They occur in stars of very different types, under very different circumstances. All high-mass stars become Type II (core-collapse) supernovae, but only a tiny fraction of low-mass stars evolve into white dwarfs that ultimately explode as Type I (carbon-detonation) supernovae. However, there are far more low-mass stars than high-mass stars, so by a remarkable coincidence the two types of supernova occur at roughly the same rate.
Figure 21.8 Type I and Type II supernovae have different causes. These sequences depict the evolutionary history of each type. (a) A Type I supernova usually results when a carbon-rich white dwarf pulls matter onto itself from a nearby red-giant companion. (b) A Type II supernova occurs when the core of a more massive star collapses, then rebounds in a catastrophic explosion.
We have plenty of evidence that supernovae have occurred in our Galaxy. Occasionally, the explosions themselves are visible from Earth (see Interlude 21-2). In many other cases we can detect their glowing remains, or supernova remnants. One of the best-studied supernova remnants is known as the Crab Nebula, shown in Figure 21.9. Its brightness has greatly dimmed now, but the original explosion in the year A.D. 1054 was so brilliant that manuscripts of ancient Chinese and Middle Eastern astronomers claim that its brightness greatly exceeded that of Venus and—according to some (possibly exaggerated) accounts—even rivaled that of the Moon. For nearly a month, this exploded star reportedly could be seen in broad daylight. Native Americans also left engravings of the event in the rocks of what is now the southwestern United States.
Figure 21.9 This remnant of an ancient supernova is called the Crab Nebula (or M1 in the Messier catalog). It resides about 1800 pc from Earth and has an angular diameter about one-fifth that of the full Moon. Because its debris is scattered over a region of "only" 2 pc, the Crab is considered to be a young supernova remnant. In A.D. 1054 Chinese astronomers observed the supernova explosion itself. The center frame shows the Crab in visible light. The left and right frames, to the same scale, show the Crab Nebula in the radio and ultraviolet, respectively.
The Crab Nebula certainly has the appearance of exploded debris. Even today, the knots and filaments give a strong indication of past violence (and continuing activity—see Interlude 21-3). In fact, astronomers have proved that this matter was ejected from some central explosion. Doppler-shifted spectral lines indicate that the nebula—the envelope of the high-mass star that exploded to create this Type II supernova—is expanding into space at several thousand kilometers per second. A vivid illustration of this fact is provided by Figure 21.10, which was made by superimposing a positive image of the Crab Nebula taken in 1960 and a negative image taken in 1974. If the gas were not in motion, the positive and negative images would overlap perfectly, but they do not. The gas moved outward in the intervening 14 years. Running the motion backward in time, astronomers have found that the explosion must have occurred about nine centuries ago, consistent with the Chinese observations.
Figure 21.10 Positive and negative photographs of the Crab Nebula taken 14 years apart do not superimpose exactly, indicating that the gaseous filaments are still moving away from the site of the explosion.
The nighttime sky harbors many relics of stars that blew up long ago. Figure 21.11 is another example. It shows the Vela supernova remnant, whose expansion velocities imply that its central star exploded around 9000 B.C. It lies only 500 pc away from Earth. Given its proximity, it may have been as bright as the Moon for several months. We can only speculate what impact such a bright supernova might have had on the myths, religions, and cultures of Stone Age humans when it first appeared in the sky.
Figure 21.11 The glowing gases of the Vela supernova remnant are spread across a large 6° of the sky. The inset shows more clearly some of the details of the nebula's filamentary structure. (The long diagonal streak was caused by the passage of an Earth-orbiting satellite while the photo exposure was being made.)
Although hundreds of supernovae have been observed in other galaxies during the twentieth century, no one has ever observed with modern equipment a supernova in our own Galaxy. A viewable Milky Way star has not exploded since Galileo first turned his telescope to the heavens almost four centuries ago (see Interlude 21-2). Now known as Tycho's supernova, this last supernova observed in our Galaxy caused a worldwide sensation in Renaissance times. The sudden appearance and subsequent fading of this very bright object in the year 1572 helped shatter the Aristotelian idea of an unchanging universe.
Based on stellar evolutionary theory, astronomers calculate that an observable supernova ought to occur in our Galaxy every 100 years or so. Because the brilliance of a nearby supernova might rival that of a full Moon, it seems unlikely that astronomers could have missed any since the last one nearly four centuries ago. Our local neighborhood of the Milky Way seems long overdue for a supernova. Unless massive stars explode much less frequently than predicted by the theory of stellar evolution, we should be treated to a (relatively) nearby version of nature's most spectacular cosmic event any day now.
