Home Search Collections Journals About Contact us My IOPscience

Introduction to stellar evolution

This content has been downloaded from IOPscience. Please scroll down to see the full text.

2016 J. Phys.: Conf. Ser. 703 012002

(http://iopscience.iop.org/1742-6596/703/1/012002)

View the table of contents for this issue, or go to the journal homepage for more

Download details:

IP Address: 155.254.207.48
This content was downloaded on 08/08/2016 at 12:17

Please note that terms and conditions apply.

8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

Introduction to stellar evolution

Scilla Degl’Innocenti
Physics Department, Pisa University, Largo B. Pontecorvo n.3, I-56127 Pisa
INFN, Sezione di Pisa, Largo B. Pontecorvo n.3, I-56127 Pisa
E-mail: [email protected]

Abstract. This contribution is meant as a first brief introduction to stellar physics. First I
shortly describe the main physical processes active in stellar structures then I summarize the
most important features during the stellar life-cycle.

1. Introduction
The quantitative study of the physical structures called “stars” developed later with respects
of other physical topics, as electromagnetism or thermodynamics. This because to understand
the operation of stellar structures one must already be acquainted with fundamental physical
mechanisms as photon-matter interactions or energy production through nuclear fusion.
Moreover precise evaluations have been made possible only by the development of electronic
calculators. At the same time telescopes more and more refined have been developed, also
on board of satellites, to observe the whole electromagnetic spectrum without the limitations
induced by the terrestrial atmosphere. Now the characteristics of stars of different mass, chemical
composition and evolutionary phase are known with a quite good precision; the general scenario
is well defined and in agreement with the most of observations of the Sun and of the stars in
our Galaxy.

2. Stellar Observables
The main stellar observables are luminosity and color. The luminosity is the energy emitted
per unit time, but it is not directly measurable for two main reasons. In general, the adopted
detector (telescope) is not sensitive to the whole electromagnetic spectrum, but only to a given
spectral range, which depends on its characteristics. Most important, due to the fact that a star
isotropically emits its energy in all the directions, what is measured is a flux (energy per unit time
and unit surface of the detector) and not a luminosity which, however, is the physical quantity
directly linked to the stellar characteristics. In more detail, if L∗ is the stellar luminosity, a
detector at a distance d from the star measures the flux:
L∗
Φ= (1)
4πd2
The luminosity can be obtained only if a distance estimate is available; it’s thus easy to
understand why the problem of distance determination of astrophysical objects is of paramount
importance. The color of a star, that is the spectral distribution of the emitted energy, can
be directly connected with the temperature of the stellar surface (photosphere). In fact for

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

4.0

3.0
RGB

log L/LSun
2.0

1.0
SGB PMS
MS
0.0
PMS
-1.0
3.80 3.75 3.70 3.65 3.60 3.55 3.50
log Teff
Figure 1. Balance between pressure
force (FP ) and gravity force (Fg ) on a Figure 2. Evolution in the Hertzsprung-Russell
given infinitesimal element at distance r diagram of a 1 M star from the Pre-Main
from the center of the star with tickness Sequence to the end of the Red Giant Branch
dr and base area dA. phase.

almost the whole stellar structure photon-matter interactions are so efficient that the star can
be assumed at thermodynamical equilibrium. In this case the gas particles follow a Maxwell-
Boltzmann distribution while the photons a Planck one, with the same temperature for particles
and photons. The temperature clearly varies with the position inside the star; in most cases the
temperature monotonic decreases from the center to the stellar surface. Under these conditions
the energy spectral distribution is the so called “black body distribution”. In reality the emission
of a star with a given surface temperature slightly differs from the one of a black body of the
same temperature, due to emission and absorption phenomena by the atoms/molecules/grains of
the stellar atmosphere. For a precise relation between the observed stellar color and the surface
temperature one needs so solve the photons transport equation in the stellar atmosphere, taking
into account all the possible absorption/emission phenomena. It’s easy to understand that these
calculations are very complex and they become more and more difficult as the photospheric
temperature decreases because, due to the higher presence of molecules and grains, the possible
photon-matter interactions increase more an more. These kind of calculations, which are an
example of how much computing capacity is fundamental for precise stellar physics evaluations,
are performed by few groups in the world and they are still affected by not negligible uncertainties
mainly for stars with relatively low photospheric temperature. In absence of high rotation
velocities and magnetic fields (which is supposed to be the situation of the most of the stars) the
stellar equilibrium configuration is a sphere (as we will better discuss in the following); however,
in general, it’s not possible to directly measure the stellar radius. Stellar surface temperatures
are generally expressed as “effective temperature” , that is the temperature which would have
a star if it would exactly emit as a black body. Thus the relation among luminosity (L), radius
(R) and effective temperature (Tef f ) is:

L = 4πR2 σTef
4
f (2)

where σ is the StefanBoltzmann constant, σ = 5.67 · 10−5 dyne cm−2 K−4 . Usually, observed
stars are shown in a plane which resumes the main stellar observables: effective temperature
(or correspondingly the stellar color) on the abscissa and total luminosity (or correspondingly
the luminosity in a given photometric band) on the ordinate, the so called Hertzsprung-Russell
(Color-Magnitude) diagram.

2
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

3. Stellar structure and evolution equations

To “solve” a stellar structure means to theoretically calculate its physical quantities
(temperature, pressure, density, luminosity production etc..) at each point inside the star.
To do this one has to solve the so called “stellar equilibrium equations” which reflect the stellar
characteristics describing its working principles.

3.1. Stellar gas characteristics

Stars are spheres of photons and plasma, that is ionized gas in which electrons are extracted from
the atoms; the plasma is thus constituted by electrons and ions but it is globally neutral. The
ionization state of the atoms which constitute the stellar plasma mainly depends on temperature
and density (and clearly on the ionization energy of the various electrons of the different atomic
species) and thus it varies inside the star. At each point in the stellar interior the ionization
state of each specie can be calculated through the Saha equation [see e.g. 1, , cap.3]. The
thermodynamical properties of the plasma can be calculated as a function of the chemical
composition and of two physical quantities (e.g. pressure and temperature) through the equation
of state (EOS) of a system of plasma and photons at thermodynamical equilibrium. For the
Sun the law of perfect gas holds with a very good approximation, except for the very central
regions in which few percents of the pressure are due to degenerate electrons. For different
conditions of temperature and pressure the EOS can be more complicated; at relatively high
density electrons became degenerate and under specific conditions coulombian effects among ions
are no more negligible. Thus if one knows the chemical composition and two physical quantities
(e.g. temperature and pressure) at a given point inside the star, one can obtain through the
EOS all the thermodynamical quantities at that point. This can be synthetically written as:

ρ(r) = ρ[T (r), P (r), X(r), Y (r), Z(r)], (3)

where X(r),Y(r) and Z(r) are, respectively, the fractional abundance in mass of hydrogen, helium
and elements heavier than helium, which in astrophysics are called “metals”, at a given radial
coordinate. Clearly X(r)+Y(r)+Z(r)=1.

3.2. Hydrostatic equilibrium

Stellar lifetimes are orders of magnitude higher than the human ones, thus it’s impossible to
directly follow a stellar life-cycle. However now one can observe an enormous number of stars at
different ages, obtaining in this way information about the characteristics of the stars at different
phases of their life. Moreover the Sun has been observed in a scientific way from centuries. In
addition the fact that the life is present on the Earth from some millions of years means that
the solar characteristics did not change in a relevant way during this period. The evidence
that most of the stars do not change their luminosity, effective temperature and thus radius
for periods longer than centuries means that they are structures at mechanical equilibrium.
The dynamical timescales (that is the time in which a structure out of dynamical equilibrium
changes its characteristics) in stars are in fact of the orders of hours, [see e.g. 2, cap.1]. At each
point inside a star the gas pressure counteracts the gravity force which would cause the stellar
contraction. The gravity is a central force, thus the equilibrium configuration is a sphere. This
simplifies very much the calculations: for spherical symmetry all the quantities can be expressed
as a function of the radial coordinate. One can imagine a stellar structure composed of several
spherical shells so thin that the physical quantities inside each shell can be assumed constant;
by solving the stellar structure equations one can obtain the value of the physical quantities in
each shell.
The mass, dM, inside a shell of radius dr is then:

dM (r) = 4πr 2 ρ(r)dr; (4)

3
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

where ρ(r) is the corresponding density inside the shell. This is called “continuity equation”.
If at each point of the star one equates the pressure force to the gravity force, as shown in
Fig.1, one obtain the so called “hydrostatic equilibrium equation”:

dP (r) M (r)
= −G 2 ρ(r) (5)
dr r
where P(r) is the pressure and M(r) is the mass inside a sphere of radius r; the above equation
holds at each radial coordinate r.
The hydrostatic equilibrium equation also implies that the most internal regions, which must
“support” the above layers, have an higher pressure. Thus, in general, the pressure is expected
to monotonically decrease from the center to the surface. The two equations written until now,
as a function of the radius, contains three unknowns: P(r), ρ(r) and M(r). P(r) and ρ(r) are
linked through the equation of state, but, except some particular EOS, in this case, also the
temperature (unknown) appears in the EOS. Thus the system is still not resolvable. This is
expected, because the energy production and transport, which play an important role in stellar
structures have still not taken into account.

3.3. Thermal equilibrium and energy production

Stars are very hot structures and thus they radiate, that is they emit photons, losing energy
from the surface. The fact that the stellar effective temperature remains constant for longtime
means that the stellar surfaces are at thermal equilibrium, that is the energy losses from the
photosphere through photons emission are counterbalanced by an energy flux which reaches the
photosphere from the underlying layer. If this would not happen the photosphere would change
its temperature in few minutes. Actually the concept of “thermal equilibrium” is more general
and it must be applied to the whole star; the energy which enters in each shell is equal to the
one which exits from the same shell plus the energy which is possibly produced inside the shell
itself.
The stellar photosphere is continuously supplied of energy by the more internal regions. For
the majority of the stellar life this energy is produced by nuclear reactions; stellar interiors are
so hot that fusion nuclear reactions of nuclei to form heavier elements can happen. The nuclear
reactions most important for energy production are among charged nuclei. For these fusion
reactions high temperatures (higher than about 106 K) are needed because the nuclei must reach
enough energy to pass the Coulom barrier through tunnel effect, reaching nuclear distances
(≈ 10−13 cm) in a way that nuclear reaction can happen, [see e.g. 3, cap.3]. The mechanism
of nuclear fusion in stars has been described in detail in other talks at this school and thus I
will not discuss it here. I only remind that the higher is the charge of reacting nuclei the higher
is the threshold temperature for the ignition of the fusion. However the nuclear fusion is not
the only possible energy source in stars; stars can produce energy through thermodynamical
transformations of the gas linked to contractions and expansions of stellar regions, the so called
“gravitational energy”. Then stars lose energy also through neutrino production. Neutrinos
have a mean free path larger than the stellar radius (except the case of very dense stellar cores
of massive stars at the end of their life) and thus they escape from the stars without interacting
with the stellar matter, that is without releasing their energy to the stellar structure. During
nuclear fusions electron neutrinos are formed; in this case the only effect is that neutrinos
take away a relatively small amount of the energy produced by the fusion, which is no more
available to the structure. Most important for the stellar energy balance is when neutrinos
are produced in regions in which nuclear fusions are not active. Different processes active for
different ranges of density and temperature are responsible for this kind of neutrino production:
neutrino photoproduction, neutrinos from couple production, plasma neutrinos etc...[see e.g. 4].
In these case the neutrino production constitutes for the stars a net energy loss and thus the

4
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

regions in which these kind of neutrinos are produces cool down. Summarizing, in each region
of the star energy production through nuclear fusions and/or thermodynamical transformations
and/or neutrino energy losses can happen. Usually we define ε the energy produced/lost per
unit time and unit mass due to the mechanisms described above, that is:

ε(r) = εnuclear (r) + εgravitational (r) − εneutrinos (r) (6)

Each process of energy production depends on the physical quantities (density/pressure and
temperature) and chemical composition which vary at each point of the star. In particular
the various nuclear reaction rates can be written as analytical expressions, as a function of
temperature, density and reagent abundances [see e.g. 5]. With these information it’s possible
to write the thermal equilibrium equation in each shell of the star:

dL(r)
= 4πr 2 ρ(r)ε(r); (7)
dr
where L(r) is the luminosity (energy per unit time) which enters in a shell of radius r. Another
equation is added to the system but another unknown, i.e L(r), is also added.

3.4. Energy transport

The energy produced in the internal regions must be transferred to the whole star. The only
energy transport mechanism which is always active in stars is the radiative transport, that is the
totality of processes of photon-matter interaction. The main processes in the stellar interiors
are: photon-electron scattering (which, due to the relatively low energies, is mainly a Thompson
scattering), photoionization, inverse bremsstrahlung, photon absorption (and further isotropic
emission) of photons by atoms ,[see e.g. 1, cap.16]. While for the external parts of the stars,
much less dense and with relatively low temperatures, the main opacity sources are scattering
and photoionization on H− ions, absorption on molecules and grains, scattering on molecules and
molecules photodissociation. The total amount of photon-matter interaction processes, which
remove energy from the outgoing flux, is called “opacity”, k, and it is measured as a cross section
(of the various processes) per unit mass, that is in cm2 /g. In this sense the induced emission
process constitutes a negative contribution to opacity. Clearly, each opacity process depends on
the physical quantities (e.g. temperature and density) and on the chemical composition, in a
way that not only the total opacity varies with radius but also the photon matter interaction
process which dominates the opacity is dependent on the position inside the star. The opacity
also depends on photon distribution. By assuming a Planck distribution (which is a valid
guess for the whole star, but the atmosphere) one can mediate the opacity over the photon
distribution obtaining the so called “Rosseland mean opacity” , k̄,[see e.g. 2, cap.4]. It’s easy to
understand that opacity calculations are very difficult because one must evaluate the efficiency
of the described processes for each element of the stellar mixture taking into account all the
possible ionization states. Opacity calculations must thus rely on a equation of state too. Few
groups in the world are able to perform these calculations; the results are made available to
the astrophysical community through tables as a function of temperature, density and chemical
composition. By imposing that photons follow the Planck distribution and assuming that the
photon-matter interaction is the only active energy transport mechanism, it’s easy to find [see
e.g. 6, cap.3], at each point inside the star, a relation between the temperature gradient and
energy flux to be transported, the so called “energy radiative transport equation”:

dT (r) 3 k̄(r)ρ(r) 1
=− Φ(r); (8)
dr 4 ac T (r)3

5
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

where the photon flux Φ at each point of the star is:

L(r)
Φ(r) = , (9)
4πr 2
c is the light speed and a is the radiation pressure constant (a=7.56 · 10−15 erg cm−3 K−4 ). Thus
one sees that the temperature gradient needed to transport a given flux entirely through photon
matter interactions (radiative gradient) is proportional to the flux itself and to the opacity, that
is to the “resistance” of the matter to the photon flux. For completeness one has to mention
that other energy transport mechanisms are possible. If the temperature gradient required to
a stellar structure to transport energy through photon-matter interactions would be too much
high, energy transport through convection becomes efficient too, even if radiative transport is
still present. Briefly, the convective energy transport is due to mass elements which move from
hotter to colder regions releasing their heat similarly to what happens for the water boiling in a
pot. In this case the total energy flux is transported through both the radiative and convective
mechanisms and the temperature gradient expression is different from the one written in eq.8. A
precise treatment of convection is out of the purposes of this talk but it’s useful to remind that
convection is active in the regions where the radiative gradient of eq.8, which is proportional to
the energy flux and to the opacity, is enough high. More precisely, for the convection activation,
the radiative gradient must be higher than the adiabatic gradient, a thermodynamical quantity,
obtained from the EOS. It’s also worth noticing that in very dense stellar regions energy transport
due to degenerate electron conduction [see e.g. 2, cap.4] dominates; in this case too eq.8 is no
more valid.
If one reminds that pressure is, in general, monotonically decreasing from the center to
the surface, taking into account the radiative transport equation and the thermodynamical
characteristics of the photon and gas mixture (e.g. for simplicity one can image a perfect gas
law) generally speaking one can understand that temperature and density too decreases from
the center to the stellar surface.
We have thus written in total four equations as a function of the radius for four unknowns
(pressure, temperature, mass, luminosity), whereas the density is related to pressure and
temperature through the EOS; thus the system is resolvable, once ε(r) and k̄(r) are evaluated.
These differential equations can be numerically solved [see e.g. 6, cap.3] finding, once fixed the
mass and the stellar original chemical composition, the physical quantities in each point of the
star.
As it will be better discuss in the following, the temporal evolution of the stars is generally
driven by chemical composition variations, mainly due to nuclear reactions. Once solved the
equations set at a given time, t0 , taking into account the variations of the chemical composition
after a temporal step Δt, one can integrate again the equations with the new composition at the
time t0 + Δt obtaining the stellar characteristics at that time. A similar procedure is applied
in the phases in which the nuclear reactions are not active when the stellar changes are mainly
driven by the contraction/expansion of the structure. These procedures are iterated for several
temporal steps to follow the stellar evolution.

4. Stellar evolution
In the following I will describe the main stellar features during the different evolutionary
phases, as obtained by solving the stellar structure and evolution equations; some of the main
characteristics can also be roughly obtained by general evaluations discussed above.

4.1. Pre-Main Sequence and Main Sequence stars

Stars born inside a molecular cloud of gas and interstellar dust; the gas is mainly composed
by H and He with traces of other elements. These molecular clouds, neglecting, as a first

6
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

approximation, possible effects due to rotation and magnetic fields, are in equilibrium between
the gravitational force and the pressure force. However, due to stochastic motions, the colder
and more dense regions of the cloud, can start to contract. As soon as the collapse proceeds the
region fragments more and more; until each part in which the original region of the cloud has
been divided will form a star. The way in which the gas of the original cloud falls onto the star
until the achievement of the stellar final mass is still an open problem, because hydrodynamical
protostellar models are still largely debated and not yet settled. The duration of the protostellar
phase too is not known with precision and it varies with the stellar mass, but it is supposed to
range from few 105 to about 106 years, [see e.g. 7, for details].
Anyhow, at the end of the protostellar phase the radiation pressure of the new born star
sweep away the residual gas of the original cloud. The star is now in the “Pre-Main Sequence
phase (PMS)”. Moreover during the first phases of their life stars are supposed to be fully
convective; thus they are completely mixed and possible chemical inhomogeneities of the original
cloud are canceled. During the protostellar phase the star contracted and warmed up until the
achievement of the hydrostatic equilibrium: the gas force exactly counterbalances the gravity
force. The stellar gas pressure is mainly due to its high temperature. However the star is warm
and radiates, loosing energy from the surface; thus, after a while, the star cools down and its
pressure decreases becoming insufficient to completely sustain the star against the gravity. The
star slightly contracts increasing its temperature and reaching again the hydrostatic equilibrium
until the radiation losses from the surface cools down again the star. The PMS phase is
characterized by these consecutive stellar contractions when the star passes from one “quasi-
equilibrium” state to another one, while its internal temperature increases. The timescale of
these processes is the one at which the stars cools down, that is the timescale in which a photon
produced in the stellar interior can reach the surface. This timescale that can be seen also
as the “relaxation time” for departure of a star from the thermal equilibrium (the so called
“Kelvin-Helmholtz” time) varies with the stellar mass but it is of the order of 106 -107 years.
While the star contracts its density increases too and the radius significantly decreases, then
also the luminosity strongly decreases as shown in Fig.2. During the PMS evolution a star reaches
first the temperature for deuterium and light elements (Li, Be, B) burning, then it reaches the
temperature for hydrogen burning into helium in its core, which is the warmest region of the
star. The hydrogen fusion into helium is the first nuclear reaction able to produce an energy
which can replace the radiative losses from the surface. Hydrogen is the first abundant element
to be burned because the coulombian barrier for the fusion of two protons is the lowest one and
H is the most abundant element in the Universe. When the hydrogen fusion is fully active the
star is in equilibrium, with the nuclear reactions which supply to the star the energy lost by
photon emission from the photosphere. Sometimes one hears on TV broadcasts that the stellar
luminosity depends on the efficiency of the nuclear reactions, but actually the exactly opposite
is true. Stars are luminous because are warm and, as all the warm bodies, they irradiate. As
a first approximation, the stellar temperature depends from the fact that the stellar gas must
be warm to have enough pressure to counteract the gravity force; then the surface temperature
also depends on the temperature gradient inside the star (see eq.8). Thus, roughly speaking,
the higher is the mass of the star, the higher is its effective temperature and its luminosity.
When, at the end of the PMS phase, hydrogen fusion ignites, it adjust its efficiency until the
energy losses from the surface, proportional to the effective temperature and the stellar radius,
are completely counterbalanced. This because in stars a feedback mechanism is efficient: if
the produced energy is lower than radiative losses the star still cools down and contract, even
if with a longer timescale. The result is the increase of the temperature of the central regions
together with the burning efficiency, which is very sensitive to the temperature. On the contrary
if the produced energy is higher than the one lost from the surface, the central regions warm
up and expand; this leads to a temperature decrease and thus to a reduction of the burning

7
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

efficiency. At the end of the process the energy produced by nuclear reactions is adjusted to
exactly counterbalance the surface energy losses. Thus, in this phase, the stellar luminosity is
mainly due to the stellar mass (with a smaller dependence on the chemical composition) and the
nuclear reaction efficiency is tuned to exactly produce the energy needed to the star to irradiate.
When nuclear reactions are so efficient to counterbalance the energy losses from the surface,
the star leaves the PMS phase starting its Main Sequence (MS) evolution. The name “Main
Sequence” is due to the fact that stars in this phase occupy in the Hertzsprung-Russell diagram
a mono-parametric sequence (neglecting the secondary effect of the chemical composition) whose
parameter is the stellar mass. It’s worth noticing that stars with masses lower than about 0.08
M do not succeed to reach the central temperatures needed to hydrogen ignition. They remain
in the Pre-Main Sequence phase continuing to contract while the density increases until the star
is sustained by the pressure of the degenerate electrons; these kind of stars are called “brown
dwarfs”. The star is now in equilibrium thanks to the nuclear energy, but the reaction efficiency
decreases according to the reduction of the fuel (hydrogen) abundance. Thus, after a while, the
nuclear energy is no more enough to counterbalance the radiative losses and the star contracts
again, with the timescale of the fuel abundance reduction, that is of the nuclear burning rate.
Thus the temperature of the central regions slightly increases until the energetic equilibrium
is reached again, with a slightly different configuration of the stellar structure; that is the star
evolves with nuclear timescale.
Nuclear timescales strongly vary with the stellar mass; the higher is the mass, the higher
is the luminosity and thus the higher is the rate of the nuclear reactions which counterbalance
the energy losses due to the luminosity. Moreover the higher is the fusion rate, the lower is the
time in which the fuel is exhausted. To be more precise for the most of the stars the luminosity
increases with the mass to a power of 2÷4, while the H abundance linearly increases with the
mass. As an example, the MS time for the Sun is about 10 Gyr while for a 15 M star of the
same chemical composition is about 107 years.

4.2. Advanced evolutionary phases

The history of a star after the central H exhaustion can be roughly seen as a succession of
central burnings of elements with charges more and more high and central regions contractions
to possibly reach the temperatures needed for the various burning ignitions. In the phases in
which the stellar core contracts nuclear burnings in a shell out of the central regions are often
present. One has to remind that the higher is the nuclear charge of the reacting nuclei the higher
is the coulombian barrier and thus the higher is the temperature needed for the fusion ignition.
But the higher is the stellar mass the higher is its central temperature. The logical consequence
is that only the most massive stars (M ≈ 10 M ) can reach the temperatures needed to ignite
all the exoenergetic reactions up to the production of the elements of the iron peak. These stars
can thus produce all the elements of the periodic table, through fusions of charged particles and
neutrons and protons captures on nuclei. The death of these stars is through an explosion in
which the most of the stellar matter is ejected in the space (type II supernova) and thus it will
constitute the cradle for the birth of a following stellar generation. Stars with lower masses have
different final fates in dependence of their mass.
After the central H burning the stars experience a phase in which H is burned in a shell
around a central inert He core. In this phase the stellar envelope expands and cools down and
the star moves toward the red in the HR diagram (Sub Giant Branch, SGB, phase), see Fig.2.
During the MS evolution the burning core slightly contracted and its temperature increased, so
that during the Sub Giant Branch H burning mainly occurs through the CN-NO bi-cycle, much
more sensitive to the temperature than the concurrent proton-proton chain [see e.g. 3, cap.6].
When CNO hydrogen burning reaches the equilibrium of secondary elements in the cycle, the
stars enter in the Red Gian Branch, RGB, phase (see Fig.2). Stars now have an extended and

8
8th European Summer School on Experimental Nuclear Astrophysics (Santa Tecla School) IOP Publishing
Journal of Physics: Conference Series 703 (2016) 012002 doi:10.1088/1742-6596/703/1/012002

low density envelope; their luminosity and lifetime are mainly governed by the dense He core and
the surrounding H burning shell, while the temperature gradient in the envelope determines the
stellar effective temperature. In particular the RGB luminosity increases with the He core mass
which grows due to the burning in the H shell which surrounds the He core. The RGB phase
ends when the central regions of the star reach the temperature for helium burning into C and
O (≈ 108 K), thanks to the He core contraction due to its mass growth. Stars with masses lower
than about 2.3 M (the so called “RGB transition” mass), whose interiors did not reach very
high temperatures, remain longtime in the RGB phase because the reaching of the He burning
temperature is more difficult. These stars reaches very high luminosities along the Red Giant
Branch and thus the mass losses by stellar winds, driven by the radiation pressure, increase too.
At the end stars with masses greater than about 0.5 M succeed to ignite central He burning.
The evolutionary theory predict that lower masses will remain in the RGB phase loosing more
and more mass until the only remain will be the dense, degenerate He core, surrounded by a
very thin H shell with all the nuclear burnings switched off. A similar structure, called He
white dwarf (He WD), is maintained in equilibrium by the pressure of degenerate electrons
which counteracts the gravity and it will cools down until it will reach the cosmic background
temperature. However the Main Sequence lifetime of stars with M < 0.5 M is higher than
the present age of the Universe, thus no one of these stars still reached the white dwarf phase.
However He white dwarfs are observed; their progenitors are stars of mass higher that 0.5M ,
which during the RGB phase loose so much mass that the H shell switches off and the stars
leave the RGB phase ending their life as He WDs. Stars which ignite He in the center but did
not explode as supernovae end their life as white dwarfs with a C/O core or with a Ne core if
they succeed to ignite carbon too.

Acknowledgments
I would like to acknowledge the collegues with which I am working in Pisa from many years in
the stellar physics field: M. Dell’Omodarme, P.G. Prada Moroni, E. Tognelli and G. Valle.

References
[1] Cox J P 1968 Principles of stellar structure - Vol.1: Physical principles; Vol.2: Applications
to stars (New York: Gordon and Breach)
[2] Hansen C J, Kawaler S D and Trimble V 2004 Stellar interiors : physical principles, structure,
and evolution (New York: Springer-Verlag)
[3] Rolfs C E and Rodney W S 1988 Cauldrons in the cosmos: Nuclear astrophysics (Chicago,
IL: University of Chicago Press)
[4] Raffelt G G 1996 Stars as laboratories for fundamental physics : the astrophysics of
neutrinos, axions, and other weakly interacting particles (Chicago: University of Chicago
Press)
[5] Fowler W A, Caughlan G R and Zimmerman B A 1967 Annual Review of Astronomy and
Astrophysics 5 525
[6] Salaris M and Cassisi S 2006 Evolution of Stars and Stellar Populations (Wiley-VCH)
[7] Stahler S W and Palla F 2005 The Formation of Stars (Wiley-VCH)