The Evolution of Stars
Evolution of Massive Stars
The upper part of the HR diagram contains massive stars in various
stages of evolution, including blue supergiants (BSG), red supergiants
(RSG) and Of stars (O supergiants with pronounced emission lines). Two
other important classes of stars in this region of the HR diagram are:
Luminous blue variables (LBVs) also known as S Doradus or Hubble-
Sandage variables have high effective temperatures of between 15 000 K
and 30 000 K with luminosities in excess of 10 6 Lsolar. They have masses
in excess of about 85 Msolar. They show clear variability and dramatic
mass loss, most probably due to the fact they exceed the Eddington
luminosity limit. Some of them also show extremely rapid rotation, which
would decrease their effective gravity at the equator. This make mass loss
from these equatorial regions much easier and could lead to the formation
of an equatorial disk. Ejected material from these stars show nuclear
processing, indicating that these are evolved stellar objects. An example
of this class is η Carinae which has undergone dramatic changes from
1837 for about 20 years, reaching an estimated luminosity of about 2 ×107
Lsolar with accompanying mass loss of about three solar masses during
this period.
Formation and Evolution of Stars: Massive Stars 2
Taken from Astronomy Picture of the Day
On the left is an image of η
Carinae, a luminous blue
variable that is estimated
to have a mass of 120
Msolar and is rapidly losing
mass at about 10-3 Msolar
yr-1. It has a luminosity of
about 5 ×106 Lsolar. Each
lobe in this image has a
diameter of about 0.1 pc
and is expanding away
from the central object at a
velocity of about 650 km/s.
The star is about 23000 pc
away.
Formation and Evolution of Stars: Massive Stars 3
Wolf-Rayet stars are very hot stars, with effective temperatures in the
region of 25 000 K to 100 000 K. They are losing mass in excess of 10-5
Msolar yr-1 with wind speeds ranging from about 800 km/s to more that
3000 km/s. The appear to all rotate very rapidly with rotation velocities of
the order of about 300 km/s. They can have progenitor masses as low as
20 Msolar. The spectra show very broad emission lines, rather that
absorption lines seen in other stars. There are three classes of these
stars: WN, WC, and WO. WN spectra are dominated by emission lines of
helium and nitrogen with hydrogen detectable in some stars. WC exhibit
emission lines of helium and carbon, with a distinct absence of nitrogen
and hydrogen. WO stars are much rarer than the WN and WC stars, with
spectra containing prominent oxygen lines, with some contribution from
highly ionized species. This trend is spectral feature is now recognised to
be a direct result of mass loss. WN has lost nearly all their hydrogen
envelope, revealing material synthesized by nuclear reactions in the core.
Convection in the core has brought equilibrium CNO cycle processed
material to the surface. Further mass loss results in the ejection of the
CNO processed material, exposing the helium burning material
generated by the triple alpha reaction. If mass loss continues, much of
this process material will be strip away leaving te oxygen component.
Formation and Evolution of Stars: Massive Stars 4
Taken from Astronomy Picture of the Day
On the right is an image of
nebula M1-67 around the Wolf-
Rayet star 124. The surface
temperature of this star is about
50 000 K. WR 124 is at a
distance of about 4600 pc away.
Formation and Evolution of Stars: Massive Stars 5
M 85 M solar : O Of LBV WN WC SN
40 M solar M 85 M solar : O Of WN WC SN
25 M solar M 40 M solar : O RSG WN WC SN
20 M solar M 25 M solar : O RSG WN SN
10 M solar M 20 M solar : O RSG BSG SN
Above is a general evolutionary path for massive stars. Each massive star
end its evolution in a supernova (SN) explosion (see later). This
qualitative evolutionary scheme has been supported by detailed
numerical evolutionary models.
Very massive stars are extremely rare (only one 100 Msolar star exists on
average for every million 1 Msolar stars.). Despite their relatively small
numbers they do play a major role in the dynamics and chemical
evolution of the interstellar medium (ISM). The enourmous amounts of
kinetic energy via winds from these stars has significant impact on the
kinetis of the ISM. One their formation they have the ability to quench
starformation in their neighbouring regions. UV light from these stars
ionizes gas in surrounding cloud. Also the highly enriched stellar winds
from these stars increases the metal content of the ISM.
Formation and Evolution of Stars: Massive Stars 6
Taken from Meynet & Meader, 2003, A&A, 404, 975
On the left is a
theoretical HR
diagram showing
the evolutionary
tracks for massive
stars with Z = 0.02.
The solid lines are
evolutionary tracks
for models with
initial rotation
velocities of 300
km/s. The dotted
lines are tracks for
models without
rotation. Mass loss
has been included
in these models.
Formation and Evolution of Stars: Massive Stars 7
The AGB and post-AGB evolution of stars more massive than about 8
Msolar is significantly different from lower mass stars. As the helium burning
shell continues to increase the mass of the carbon-oxygen core, and as
the core continues to contract, it eventually starts fusing carbon
generating a variety of by products including 16O, 20Ne, 23Na, 23Mg and
24
Mg.
What follows is a succession of nuclear reaction sequences. Following
carbon burning, the oxygen in the resulting neon-oxygen core will ignite,
producing a new core composition dominated by 28Si. Finally at
temperatures near 3 × 109 K, silicon burning starts through a series of
reactions: 28 4 32
14 Si 2 He ⇔ 16S
32
16S 4 He ⇔ 36 Ar
2 18
.
.
.
52 4 56
24 Cr 2 He ⇔ 28 Ni.
Formation and Evolution of Stars: Massive Stars 8
This silicon burning produces a host of nuclei centered near the 56Fe peak
of the binding energy per nucleon curve (see notes on energy
generation), the most abundant of which is 54Fe, 56Fe and 56Ni. Reactions
that produce nuclei more massive than 56Fe are endothermic, and thus
cannot contribute to the luminosity of the star. Silicon burning is thus said
to produce an iron core (grouping all these products together).
Because carbon, oxygen and silicon burning produce nuclei with masses
progressively closer to the iron peak of the binding energy curve, less
and less energy is liberated per gram of fuel. This results in the time scale
for each succeeding reaction sequence becoming shorter. As an example
a 20 Msolar has a main-sequence lifetime of roughly 107 years, core helium
burning requires 106 years, carbon burning last 300 years, oxygen
burning takes roughly 200 days and silicon burning will be completed in
only 2 days.
At the very high temperatures now present in the iron core, photons
possess enough energy to destroy heavy nuclei. This process is known at
photodisintegration
Formation and Evolution of Stars: Massive Stars 9
Taken from Carroll & Ostlie
The onion-like interior of a
massive star that has
evolved through core silicon
burning. Inert regions of
processed material are
sandwiched between the
nuclear burning shells.
Drawing is not to scale.
Formation and Evolution of Stars: Massive Stars 10
Of particular importance is the photo-disintegration of 56Fe and 4He:
56 4
26 Fe 13 2 H 4 n
4
2 He 2 p 2 n
Photo-disintegration can, in a very short period of time, undo what the star
has done during its entire life, that is producing elements more massive
than hydrogen and helium. It is a highly endothermic process, and
thermal energy is removed from the gas that would otherwise have
resulted in pressure necessary to support the core of the star. The core
masses for which this process occurs vary from 1.3 Msolar for a 10 Msolar
ZAMS star to 2.5 Msolar for a 50 Msolar star.
Under such extreme conditions now present in the core of the star
(Tc ~ 3 × 109 K and ρc ~ 1010 g cm-3 for a 15 Msolar star) the free electrons
Formation and Evolution of Stars: Massive Stars 11
that is assisting the pressure support of the star through degeneracy
pressure are captured by heavy nuclei and by protons that were
produced through photo-disintegration:
−
p e n e
The amount of energy that escapes the star in the form of neutrinos
becomes enormous (for a 20 Msolar star during silicon burning the photon
luminosity is 4.4 × 1038 ergs s-1while the neutrino luminosity is 3.1 × 1045
ergs s-1). Most of the core's support in the form of electron degeneracy
pressure is now gone, and the core starts to collapse extremely rapidly. In
the inner part of the core the collapse is homologous and the velocity of
collapse is proportion to the distance away from the centre.
At the radius where the velocity is exceed the local sound speed, the
collapse can no longer remain homologous, and the inner core decouples
from the supersonic outer core, which is left behind and in nearly free-fall.
Speeds can reach almost 70 000 km s -1 in the outer core (at this speed a
volume the size of Earth will be compressed down to the size of 50 km).
Formation and Evolution of Stars: Massive Stars 12
Homologous collapse of inner core continues, until density exceeds
8 × 1014 g cm-3, roughly three times density of an atomic nucleus. Due to
Pauli's exclusion principle the strong nuclear force suddenly becomes
repulsive. This causes the core to stiffen and rebound somewhat, sending
pressure waves outward into the infalling material of the outer core. When
the velocity of the pressure waves reach the the sound speed, they build
into a shock wave that start to move outwards.
As the shock wave encounters the infalling outer core, the high resulting
temperatures causes further photodisintegration, robbing the shock of
much of its energy (for every 0.1 Msolar of iron that is broken down into
protons and neutrons, the shock loses 1.7 × 1051 ergs).
If the remainder of the core is not too massive (not greater than about 1.2
Msolar), the shock will fight its way through rest of the outer core and collide
with the remainder of the nuclear processed material and the outer
envelope. Once the shock forms above the surface of the inner core, the
time it takes to penetrate the outer core is only 20 milliseconds. This
process known as prompt hydrodynamic explosion.
Formation and Evolution of Stars: Massive Stars 13
If the iron core is too large the shock stalls becoming nearly stationary
with the infalling matter accreting onto it. The shock becomes an
accretion shock. Below the shock a neutrinosphere develops from the
process of photodisintegration and electron capture. Since the overlying
material becomes extremely dense, neutrinos cannot easily penetrate this
matter, and about 5% of the neutrino energy is deposited in the matter
just behind the shock. This additional energy heats the material and
allows the shock to resume its march towards the surface. This
temporary stalling of the shock is called a delayed explosion
mechanism.
If the initial ZAMS mass of the star is not too large (perhaps less than 25
Msolar), the remnant in the inner core will stabilize to become a neutron
star, supported by degenerate neutron pressure. If the stellar mass is
much larger, neutron degeneracy is not enough to support the remnant
core against gravity and the final collapse will be complete, producing a
black hole. The total energy released is on the order of the binding
energy of a neutron star, about 3 × 1053 ergs. This is about a 100 times
the more energy the Sun will produce during its entire lifetime.
Formation and Evolution of Stars: Massive Stars 14
As the shock moves outwards toward the surface, it drives the hydrogen
rich envelope and the remainder of the nuclear-processed material in
front of it. The total kinetic energy of the expanding material is on the
order of 1051 ergs, about 1% of the energy liberated in neutrinos.
When this material becomes optically thin (at a radius of about 10 15 cm), it
releases about 1049 ergs of energy win the form of photons with a peak
luminosity of nearly 1043 ergs s-1, or roughly 109 Lsolar.
The catastrophic collapse of an iron core, followed by the generation of a
shock wave with the resulting ejection of the stellar envelope and
tremendous optical display is believed to be the mechanism by which
Type II supernovae are created. Observationally these supernovae are
characterized by a rapid rise in luminosity, reaching a limiting absolute
bolometric magnitude of -18, followed by a steady decrease, dropping 6
to 8 magnitudes a year. Their spectra show lines associated with
hydrogen and heavier elements, with P-Cygni profiles common to many
lines. Type II supernovae can be classified as either Type II-L (linear) or
Type II-P (plateau).
Formation and Evolution of Stars: Massive Stars 15
Taken from Carroll & Ostlie
The formation of a P-Cygni line profile, in for example the outflow from a
supernova explosion.
Formation and Evolution of Stars: Massive Stars 16
A false-colour image of
the Crab Nebula, a
supernova remnant.
This image was created
by combining from data
taken by space-based
observatories,
Chandra, Hubble, and
Spitzer, to explore
the debris cloud in x-
rays (blue-purple),
optical (green), and
infrared (red) light. The
Crab Pulsar, a neutron
star spinning 30 times a
second, is the bright
spot near picture
center. This remnant of
the stellar core powers
the emission at all
wavelengths. The Crab
Nebula is about 12 ly,
and about 6,500 ly
away.
Formation and Evolution of Stars: Massive Stars 17
* Theorized progenitor
Massive star with hydrogen
rich envelope *
Mass acreting white dwarf *
WN star * WC star *
Possibly merging or accreting Core collapse supernovae
white dwarf exceeding
Chandrasekhar limit
Formation and Evolution of Stars: Massive Stars 18
Taken from Carroll & Ostlie
Formation and Evolution of Stars: Massive Stars 19
Taken from Doggett & Branch, 1985, AJ, 90, 2303
Composite blue light curve for
Type I supernovae.
Formation and Evolution of Stars: Massive Stars 20
Taken from Doggett & Branch, 1985, AJ, 90, 2303
The characteristic shapes of Type
II-P and Type II-L light curves.
These are composite light curves,
based on the observations of
many supernovae.
Formation and Evolution of Stars: Massive Stars 21
The source of the plateau seen in Type II-P curves are due to the
radioactive decay of a large amount of 56Ni that was produced by the
shock front during its march through the star (the half life of 56Ni is 6.1
days).
It is expected that the explosive nucleosynthesis of a supernova shock
will produce significant amounts of other radioactive isotopes as well,
such as 57Co (half life of 271 days), 22Na (half life of 2.6 years) and 44Ti
(half life of 47 years). If these isotopes are present in sufficient quantities,
each in turn may contribute to the overall light curve, causing the slope of
the curve to change.
The 56Ni is transformed into 56Co through a beta-decay reaction:
56 56
28 Ni. 27 Co e e
The energy released by this decay is deposited in the optically thick
expanding material and then radiated away from the supernova's remnant
photosphere. This “holds up” the light curve for a time resulting in an
observed plateau.
Formation and Evolution of Stars: Massive Stars 22
56
Co, the product of the radioactive decay of 56Ni, is itself radioactive with
a longer half-live of about 78 days:
56 56
27 Co 26 Fe e e
As the luminosity of a supernova diminishes over time, it should be
possible to detect the contribution to the light made by 56Co.
Radioactive decay is a statistical process, and as such the decay rate
must be proportional to the number of atoms remaining in the sample:
dN
= − N
dt
where is a λ constant. Integrating this equation gives
− t
N t = N 0 e
where N0 is the original number of radioactive atoms in the sample.
Formation and Evolution of Stars: Massive Stars 23
Taken from Carroll & Ostlie
The radioactice decay of
56
Ni with a half life of 6.1
days. There is a 50%
chance that any given
56
Ni atom will decay
during a time interval of
6.1 days. If the original
sample is entirely
composed of 56Ni, after n
successive half-lives the
fraction of 56Ni atoms
remaining is 2-n.
Formation and Evolution of Stars: Massive Stars 24
The constant λ is related to the half-life, τ1/2 by:
ln 2
=
1/2
Since the rate at which decay energy is deposited into the supernova
remnant must be proportional to dN/dt, the slope of the bolometric light
curve is given by
d log10 L
= −0.434
dt
or
d M bol
= 1.086
dt
By measuring the slope of the light curve, one can verify the presence of
large quantities of a specific radioactive isotope.
Formation and Evolution of Stars: Massive Stars 25
Taken from Suntzeff et al., 1992, ApJL, 384, L33
The bolometric light curve
of SN 1987A through the
first 1444 days after the
explosion. The dashed
lines show the
contributions expected
from the radioactive
isotopes produced by the
shock wave. The initial
mass of 56Ni (later 56Co)
produced is estimated to
be 0.075 Msolar.
Formation and Evolution of Stars: Massive Stars 26
An important success of stellar evolution theory is the ability to explain
most of the observed abundance ratios of elements. Hydrogen is believed
to be primordial, synthesised immediately following the Big Bang with
much of present day Helium also being formed at that time. Most of the
remaining elements were formed via nuclear processes in stellar
environments.
Relative to hydrogen and helium, lithium, beryllium and boron are very
under-abundant in the universe. As mentioned before, the abundance of
lithium is particularly low in the solar surface as compared to that in
meteorites (solar lithium problem), while the abundance of beryllium is
comparable.
Peaks in the relative abundances for elements occur at carbon, nitrogen,
oxygen, neon and so on because they prominent end products of nuclear
fusion reactions in stellar interiors. Type II supernovae are also
responsible for the generation of significant quantities of oxygen while
Type I supernovae (believed to be due destruction of carbon-oxygen
white dwarf exceeding the Chandrasekhar mass limit) are responsible for
the creation for most of the iron observed.
Formation and Evolution of Stars: Massive Stars 27
Taken from Carroll & Ostlie
The relative abundances
of elements in the Sun's
surface. All abundances
are normalized to 1012
hydrogen atoms.
Formation and Evolution of Stars: Massive Stars 28
The production of higher Z elements, becomes more difficult due the
existence of a high Coulomb barriers and thus requiring very high
temperatures for such fusion reactions. In contrast nuclear reactions
involving neutrons can occur at relatively low temperatures. Reactions
with neutrons
A A1
Z X n Z X
result in more massive nuclei that are either stable or unstable against the
beta-decay reaction,
A1 A1 −
Z X Z 1 X e e
If the beta-decay half life is short compared the the time scale for neutron
capture, the neutron capture reaction is said to be a slow process or an s-
process. s-process reactions tend to yield stable nuclei, directly or
secondarily via beta decay and tend to occur in normal phases of stellar
evolution.
Formation and Evolution of Stars: Massive Stars 29
If the beta-decay half life is long compared to the time scale for neutron
capture the reaction is termed as a rapid process or r-process. r-process
reactions result in neutron-rich nuclei and can occur during a supernova
when a large flux of neutrinos exist.
Both these reactions do not play significant roles in energy production, but
they do account for the abundance ratios of nuclei with A > 60.
Formation and Evolution of Stars: Massive Stars 30