Evolution of Massive Stars

275-286


In the early morning hours of February 23 1987, Canadian astronomer Ian Shelton was working at the Las Campanas Observatory in Chile and developing photographic plates of a region in the southern hemisphere sky. As he inspected the fresh plate he spotted it immediately - a fifth magnitude star that should not have been there! He stepped outside and looked up into the magnificent southern sky to see this new star hanging overhead in the Large Magellanic Cloud - one of our closest galactic neighbours. A few minutes later, at another telescope at Las Campanas a telescope operator, Oscar Duhalde spotted the new star. Within hours reports began to flood in corroborating Shelton's observation and on February 25, 1987 the International Astronomical Union's Central Bureau for Astronomical Telegrams announced the discovery of the first "naked-eye" supernovae in more than 350 years.

Although novae and supernovae bear a superficial resemblance they are quite different in a number of fundamental ways. Supernovae are rare and violent events in the extreme. The term literally means super - new - (star) and marks the final and troubled chapter in a massive star's life. The star simply explodes and for a few days can outshine the entire galaxy of stars

Supernovae: Past and Present

Ancient SN candidates:

  • Perhaps the earliest recorded supernovae dates from 185 AD. Translating from Chinese records: "on December 7, 185 a guest star appeared in the midst of the constellation Nan-men; it was big as the half of a bamboo mat and showed the five colours in turn ... It diminished in brightness little by little and finally disappeared about July of the next year" Nan-men refers to a region in the constellation Centaurus
  • Other ancient Chinese records suggest that supernovae may have also been observed in 386 AD and 393 AD
  • 1006: One of the most brilliant supernovae in recorded history is recorded by Chinese, Arabic and European astronomers
  • 1054: Chinese astronomers record a SN in the constellation Taurus. Today we see a beautiful nebula (the Crab nebula) in the same spot. Oddly - there is no mention of this in European records.
  • In August 1181 AD Chinese and Japanese astronomers observed a supernova in the Milky way in the constellation Cassiopeia. The supernova was visible for 3 months and is still detectable today.
Figure 10.23 A magnificent HST image of the Crab Nebula which marks the site of the 1054 AD supernova explosion.

Recent SN

  • November 11, 1572 was seen by Tycho Brahe in Cassiopeia. The supernova was doubtless seen by many others and likely detected as early as November 6, 1572 but is known as Tycho's supernova because it was this event that convinced Tycho to devote his life to the study of astronomy. The supernova became as bright as Venus and was visible during the daytime for several weeks. The supernova was visible to the unaided eye for more than 16 months.
  • In the fall of 1604 another supernovae (the last seen in our own galaxy) occurred in the constellation Ophiuchus and was studied in detail by Johannes Kepler. It is commonly referred to as Kepler's supernova.
  Figure 10.24 Chandra space telescope X-ray image of the remnant of Tycho's supernova.
  • Supernova 1987A is the first nearby supernova discovered since the invention of the telescope.
  Figure 10.25 (Left) Visible light image of supernova 1987A in a region of the Large Magellanic Cloud (Right) The progenitor star Sanduleak -69 202a indicated by arrow.

 

Supernovae in other galaxies are routinely observed every year.

The modern designation for supernovae is "SN" + "Year of discovery"+"Letter designation of order of discovery". So, SN1987A refers to the first supernova discovered in 1987. We do this because many supernovae are discovered each year. Figure 10.26 is a King's University College Observatory image of an extremely energetic supernova SN 2004 ET discovered in the fall of 2004 in the constellation Cepheus. While not visible to the unaided eye it was easily see in its host galaxy NGC 6946, 18 million light years distant.

As you will see in the next unit, supernovae are valuable "probes" telling us a great deal about the large-scale structure of the universe.

Figure 10.26 King's University College Observatory image of SN 2004 ET (labeled V) and 5 foreground comparison stars which are in our galaxy.  

Supernovae Come in Two Flavours...

Astronomy, like all sciences advances through painstaking observation. Although there has been only one bright ("naked-eye")supernova since the invention of the telescope, literally hundreds have been studied in other galaxies. This has led to the realization that there are two distinctly different kinds of supernovae. This has come by studying both how the intensity of light varies with time during a supernovae event as well as looking at the spectra of supernovae.

How the light varies - Looking at Light Curves

Figure 10.27 compares two supernovae (SN 1937C and SN 1987A). SN 1937C is a "classic example" of a Type I supernova. It shows a rapid decline and then a leveling off and more gradual decline. SN 1987 is a Type II supernova and shows a briefer decline then a plateau followed by another rapid decline.

Figure 10.27 Comparison of the light curves produced by Type I and Type II supernovae.

Differences in the Brightness and Spectra of Type I and Type II Supernovae

Type I
  • the most brilliant class, luminosity approximately 4 billion solar luminosities (M = -19)!
  • spectrum shows NO HYDROGEN lines but has many broad emission and absorption features
Type II
  • hydrogen lines are present
  • less luminous than the Type I event, luminosity approximately 0.6 billion solar luminosities (M =-17)!

Example 10.10 Suppose you were writing a research proposal to study supernovae. How many galaxies would you need to observe annually to observe 20 supernova events?

Solution: Supernovae occur with an estimated frequency of 1 event per galaxy every 50 years. By a simple scaling argument, if we observe 50 galaxies for 1 year we would expect to see at least one event. So, if we want to see 20 events we will need to observe 20 X 50 = 1000 galaxies. This is a bit too optimistic however. For one, we can't count on observing the galaxy for the entire year - it will either be lost in the glare of daytime or, if it is circumpolar, poorly placed in the sky for half the year. So, we should increase our sample size to 2000 galaxies! Second, we can't count on clear skies all the time so we should revise our estimate slightly higher again. This will depend on the location of your observatory and could ultimately determine the viability of your research proposal At best you should plan on observing at least two thousand galaxies, as often as possible annually to achieve your objective of detecting 20 supernovae per year.

As example 10.10 suggests, observing supernovae is arduous work. Prior to the 1990's most supernovae were discovered by dedicated amateur astronomers who had visually "memorized' what the star fields around faint galaxies look like. Since the 1990s robotic telescopes have been able to observing tens of thousands of galaxies annually. This creates massive amounts of data which is automatically analyzed ("reduced") by computer to alert astronomers if a supernova has been detected. The world wide web has ensured that this information is distributed immediately for critical follow-up observations.



Conditions that favour a Type II Supernova Explosion

The existence of two classes of supernovae points to two different ways in which a supernova can occur. Common to both, however, is the fundamental process of stellar collapse and the violent conversion of matter into energy. There are two key conditions that will lead to the development of a supernova:

  1. star starts out massive and retains mass more than 3 solar masses of matter when it reaches shell burning, or
  2. star is massive and member of a close binary system

Nuclear Fusion - Not Always a Good Idea

As you now understand, stars "work" by using the energy released by nuclear fusion - a process that combines lighter elements to form heavier ones and usually releases energy. As a star ages it accumulates - via fusion - increasingly heavy elements in its core. First helium, then carbon , neon, oxygen and so on. The usual "scenario" is this:

The accumulation of heavy fusion products effectively "shuts off" core burning. Unable to support its weight, the star collapses a bit and cashes in some of its gravitational energy reserve . This causes heating which eventually enables the core to begin a new fusion reaction which stabilizes the star from further collapse A new shell burning source develops around the core.

We can summarize the fusion steps for a massive star in the following table:

Fusion Steps
Result
PP or CNO chains, H —>He
yields energy GOOD IDEA!
triple-alpha converts He —>C
yields energy GOOD IDEA
C—>Ne, O, Mg
yields energy GOOD IDEA
O ---->> Si,S,P
yields energy GOOD IDEA
Si ---->> Fe
yields energy GOOD IDEA
Fe --->> Co, Ni
costs energy
VERY BAD IDEA!
Nature pulls the ultimate practical joke on our increasingly desperate star, for iron is the most stable of all nuclei. Fusion beyond iron takes energy rather than releasing it. So, instead of halting collapse, iron fusion hastens it!
Table 10.3 Fusion steps that a massive star could proceed through.

 

 

Figure 10.28 Animation of the evolution of a 25 Mo star

 

The following table summarizes the temperatures and densities exhibited by a 25 solar mass star:

Stage Temp. (K) Density (g/cc) Duration of Stage
Hydrogen burning 4 x 107 5 7 million years
Helium burning 2 x 108 700 700 thousand years
Carbon burning 6 x 108 2 x 105 600 years
Neon burning 1.2 x 109 4 x 106 1 year
Oxygen burning 1.5 x 109 107 6 months
Silicon burning 2.7 x 109 3 x 107 1 day
Core collapse 5.4 x 109 3 x 109 0.25 seconds
Core bounce 2.3 x 1010 4 x 1012 milliseconds
Explosion about 109 . 10 seconds
Table 10.4 Stages of nuclear burning for a 25 solar mass star.
 

The Point of No Return

Once central temperatures and densities initiate carbon burning the star is "done for". Its end will come in two ways
  1. The carbon detonation may be so violent that the star explodes at this stage. This would be the case for lower mass stars (3 M - 9 M).
  2. If the star has a greater mass then the core will contract and the nuclear fusion process will continue to produce heavier elements up to and including iron (Fe). When the iron core achieves a mass of 1.3 - 2 M collapse ensues because of a number of disastrous complications:
    • The density of the core is now approaching the density of the nucleus of an atom! This means that, electrons are literally "squeezed" into the nuclei in a process called electron-capture. Since the electrons were helping provide support pressure this destabilizes the star even more and accelerates collapse.
    • The gamma rays being produced in the core are now energetic enough to tear the nuclei apart in a process called photodisintegration. This represents loss of yet another source of support pressure - again hastening collapse.
    • Neutrino Losses: Neutrinos are produced in copious quantities. Initially they escape from the core and cause an additional cooling.
Eventually, collapse occurs so rapidly and the density increases so dramatically that this flood of neutrinos reverses the collapse and violently ejects the material above the core. The core itself may undergo additional rebound compression and end up in a very peculiar way!

 

Conditions that favour a Type I Supernova Explosion

If a massive star is a member of a close binary and has a white dwarf companion it will eventually produce a Type I supernovae. As figure 10.29 shows, as the massive star evolves into a blue supergiant it will fill its Roche lobe and mass transfer to the white dwarf will begin.

The white dwarf is almost entirely composed of a hot but no longer nuclear burning core of carbon and oxygen. As matter (mostly hydrogen) from the supergiant falls onto its surface the white dwarf's mass begins to increase. However, nature strictly enforces the Chandraskehar limit! Recall from the last section, the Chandrasekhar limit of 1.4 Mo represents the largest mass for which the degenerate pressure produced in the core of a white dwarf can support the star's weight. If a white dwarf exceeds this mass it will collapse with catastrophic consequences. It will heat and compress very quickly to the point at which Carbon detonation will occur

with a devastating effect. The entire star will be destroyed in the ensuing explosion. Figure 10.29 Blue supergiant is losing mass onto a companion white dwarf. If the white dwarf accretes enough mass to exceed the Chandrsekhar limit it will explode as a Type I supernova.

 

Nucleosynthesis

In his greatest work Leaves of Grass (1855), American poet Walt Whitman summed up a profound truth with the line quoted above. This "truth" is that you and I and all living creatures are in fact "star stuff". One of the final acts of a supernova is to cobble together all of the heavy elements that make up our universe. The intense nuclear reactions occurring during core collapse and core bounce provide the energy need to fuse neutrons together in a dizzying array of combinations to produce the hundreds of elements and isotopes that make up our universe. The iron so necessary for the hemoglobin in your blood or the calcium in your bones were born in the cores of dying stars! This is one of the greatest discoveries of the 20th century and is called nucleosynthesis.

 

Some Likely Supernova Candidates

Are there any stars we should "keep an eye on"? Let's address this question with the following worked examples:

Example 10.11 The binary system IK Pegasi consists of an A8V star of mass 1.67 Mo orbiting a white dwarf of mass 1.15 Mo. The stars are separated by a distance of 44 Ro. Is this a possible supernova candidate and if so, what type?

Solution: Use Roche to inspect the Roche lobes for this system. The stars are certainly close enough for mass transfer to occur. The critical radius for the A8V star is 17 Ro and it will certainly evolve to this size so mass transfer is very likely. Since the white dwarf is currently 1.15 Mo it is reasonable to expect that it will gain enough mass to become a Type Ia supernova. Stellar models suggest that this will occur about 2 billion years from now. Keep-an -eye on it - but don't hold your breath!
  Figure 10.30 Roche lobe for potential supernova candidate IK Pegasi

Example 10.11 helps to underscore an important caveat on whether or not a star system with a white dwarf companion will become a Type I supernova. The system must be a "close system" so that the size of the donor star is roughly the same order of magnitude as the size of its Roche lobes (ie - it must be able to fill its Roche Lobe at some time in its evolution).

Example 10.12 Look up any relevant data that you need to help you to determine whether or not Betelgeuse will become a supernova and if it will determine what type.

Solution: Betelgeuse would appear to be an excellent candidate for a Type II supernova event. It is a lone star with a mass of 20 Mo and appears to be highly evolved - likely in helium-core burning stage. Its current radius is almost 1000 Ro!

 

 

 

 

  Figure 10.31 Betlegeuse - a familiar red star in our winter sky located in the upper, eastern corner of Orion.

 

 

Supernova Remnants - Remembrances of Things Past

In its dramatic demise a supernova leaves behind a shimmering, ethereal monument to itself. Supernova Remnants (SNR's) can be breathtakingly beautiful. We also think that they can be participants in the seeding of the galaxy via nucleosynthesis and in the initiation of star collapse as they plow through the interstellar medium as a shock wave. Figure 10.32 is a zoomable image of the Crab Nebula. Spend some time exploring this amazing object.

 
Figure 10.32 A zoomable Hubble image of the Crab Nebula (M1). Drag with the mouse to pan, click left mouse button to zoom or use controls on the applet.

In young supernova remnants (snr's) the gases expelled by the dying star can move at a considerable fraction of the speed of light with velocities on the order of 10 000 km/s to 20 000 km/s. Nearly 1000 years after the Crab Nebula snr was created, gases are still moving outward at 1500 km/s.

Practice

  1. An astronomer uses the Doppler shift in the hydrogen lines in the spectrum of a distant supernova to conclude that gases are expanding outward at 15 000 km/s. What type of supernova is this?
  2. You have been hired as scientific advisor to the screen writers for a popular sci-fi series. In one episode a rescue mission is launched to save the inhabitants of a planet circling a red dwarf star that is hours away from "going supernova". What's wrong with the science here?
  3. Discuss whether or not Sirius will ever become a supernova. (Hint use the applets Roche and starEvolve to answer this). Here is some salient data:

    Sirius A: Sp. Type: A1V mass = 2.1 Mo
    Srius B: Sp. Type: WD mass = 1.0 Mo
    Separation of stars 23 AU
  4. List three essential conditions that must exist in a binary star system for it to ever produce a Type Ia supernova.
  5. Type Ia supernovae make excellent "standard candles" since they all have approximately the same absolute magnitude of -19 at their maximum brightness. Explain how a Type Ia supernova occurs and speculate why they are all so uniform in brightness.

 

To understand the evolution of massive stars

Chp 13.3