Stars N Stuff

ASTRO1004 XCOS1Y
Life and Death of Stars I

Lecture Handouts 2022
Version 1.0.2 — 4 Jan 2022
Robert Bennett
Contents
1 Star Basics I 2
1.1 Star names . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Magnitude system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Distances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 Star Basics 2 7
2.1 Luminosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Inverse Square Law . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.3 Intrinsic brightness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Size and mass of stars; binary stars 14
3.1 Stellar radii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2 Binary stars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3 Masses of stars in binary systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4 Hertzsprung-Russell (HR) Diagram and stellar classification 20
4.1 Hertzsprung-Russell Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2 Mass-luminosity relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3 Classifying stars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.4 Spectrum lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 Proper motions; composition and interiors of stars 26
5.1 Proper motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.2 Inside the Sun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.3 Nuclear energy generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.4 Stellar interiors along the Main Sequence: Convective cores and envelopes . . . . . . . 32
6 Nucleosynthesis I 33
6.1 Elementary particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
6.2 Cosmic abundances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
6.3 Main Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.4 Triple alpha process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.5 Helium capture reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.6 Alpha process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
7 Nucleosynthesis II 37
7.1 Curve of binding energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
7.2 Neutron capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
7.3 Evidence for stellar nucleosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
7.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
8 Visible features of the Sun; equivalents on other stars 41
8.1 Photosphere and granulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
8.2 Differential rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
8.3 Sunspots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
8.4 Upper atmosphere: chromosphere and corona . . . . . . . . . . . . . . . . . . . . . . 44
8.5 Flares and Coronal Mass Ejections . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1
Lecture 1
Star Basics I
This course and part II will take a close look at stars: how we can learn about them, what they are,
why they keep shining, their births and deaths and the sorts of object they leave behind.
It’s obvious that an Astronomy course will spend some time looking in detail at stars. To the naked
eye they are the obvious feature of the night sky. We find planets orbiting them. Galaxies, the building
blocks of the entire universe, are huge assemblages of stars so we need to understand stars for a
complete understanding of galaxies. Since we find them throughout the universe it is not completely
surprising that they are tied up in fundamental issues: the origin of the chemical elements.
Let’s start by noting some obvious features of stars in the sky. Some are bright and some faint. The
faint ones are more numerous. A telescope or binoculars reveals many more, still fainter than the
eye can see. They have different colours: red, orange, yellow, white, blue -these become evident in
binoculars even if they are obvious to the eye only for the very brightest stars. We will start from
these facts, make them more quantitative and work from them towards physical statements about
stars.
First, what do we call them?
1.1 Star names

1.1.1 Constellations
Inevitably our minds “join the dots” and pick out shapes among the stars. The patterns that stars form
in the sky don’t change perceptibly during a human lifetime, so these shapes retain their usefulness.
People in ancient times realised this and saw the shapes of their gods, characters from their legends
among the stars and named these patterns accordingly. Use of these shapes has continued into modern
times. In 1930 the International Astronomical Union adopted 88 of these historical constellations as
standard, drawing boundaries round them so that each point in the sky lies in one and only one
constellation. Some of these are very widely known, e.g. Ursa Major (the Great Bear), Ursa Minor
(the Little Bear), Orion (the Hunter), and of course the constellations of the Zodiac: Taurus, Gemini,
Cancer, Leo, Virgo, etc.
This gives the first method of putting names to the bright stars, ordering them in a constellation by
brightness. We use the Greek alphabet for this, calling the brightest alpha (α), the second brightest
beta (β), the third brightest gamma (γ) and so on. So β Leonis is the second brightest star in the
constellation of Leo. We call these names Bayer letters after the astronomer who first started using
them in 1603.
We also sometimes use Flamsteed numbers. (Naked-eye) stars are numbered from west to east across
the constellation, without regard to brightness. β Leonis, mentioned above is also 94 Leo.
2
LECTURE 1. STAR BASICS I
1.1.2 Proper names

All the brightest stars and some fainter ones have been given their own, unique names, mostly by the
Persian astronomers of the first millennium AD. Again, some of these names are very well-known in
popular culture: Betelgeuse, Rigel, Sirius, Arcturus, Vega, etc. For example, β Leonis is also called
Denebola.
1.1.3 Catalogue numbers

More recently astronomers have drawn up many catalogues, some for specific purposes (e.g. faint,
variable stars within a certain distance from the solar system), others simply aiming at completeness.
The resulting names include an abbreviation for the name of the catalogue and the number of the
particular star. β Leonis is also:
• HD 102647 (no. 102647 in the ”Henry Draper” catalogue - historically important, from the early
20th century)
• SAO 99809 (no. 99809 in the Smithsonian Astrophysical Observatory catalogue)
• HR 4534 (no. 4534 in the Yale Bright Star Catalogue which includes 9110 naked eye stars)
• BD+lS 2383 (entry in the Bonner Durchmusterung catalogue, which refers to entries by position
on the sky rather than simply in a list)
• TYC 870-988-1 (entry in the Tycho-2 catalogue, produced by the European Space Agency in
2000 using measurements from the Hipparcos space mission; includes the 2.5 million brightest
stars)
• And many other names in many other catalogues
1.2 Magnitude system

Some stars are bright and some faint. The magnitude system puts a number to the brightness of a star.
This evolved historically from a loose way of speaking and it is a little bit tortuous in consequence. In the
second century AD stars were classified according to “magnitude”, meaning something like “bigness”
or “importance”. So, the brightest stars were said to be “of the first magnitude.” There were six
magnitude classes, with those of sixth magnitude being the faintest perceptible to the eye.
More recently the magnitude system was made more precise and mathematical, with the following key
features:
• The bigger the magnitude, the fainter the star
• Stars may have any magnitude, bigger than 6 or less than 1 (even negative for very bright
objects), and stated precisely, e.g. 8.91 or 0.04
• A difference in magnitude corresponds a ratio of brightness (i.e. the magnitude scale is a
logarithmic scale)
• A difference of 5 in magnitude corresponds to a ratio of 100 in brightness.
– A star of magnitude 1.0 is 100 times brighter than a star of magnitude 6.0
– A star of magnitude 5.5 is 100 times brighter than a star of magnitude 10.5, etc.
• This means that a difference of 1 in magnitude corresponds to a ratio of approximately 2.5
in brightness
– A star of magnitude 3.4 is 2.5 times brighter than a star of magnitude 4.4.
– A star of magnitude -0.6 is 2.5 × 2.5 = 6.25 times brighter than a star of magnitude 1.4,
etc.
v1.0.2-4 Jan 2022 3

• And mathematics involving powers and logarithms lets us work out the brightness ratio of any
two magnitudes
– A star of magnitude 6.1 is 43.7 times brighter than a star of magnitude 10.2 (you don’t
have to be able to do this; search for Pogson’s equation if you want to)
– Magnitudes of some well-known objects will help with understanding the magnitude scale:
• The Full Moon has a magnitude of -12.5, and the Sun’s magnitude is -26.7
• The brightest star is Sirius, in the constellation of Canis Major, with magnitude -1.5
• The stars Arcturus, Vega and Capella are close to magnitude 0.0
• The planet Jupiter looks to the naked eye like a bright star with magnitude in the region of -2.5
• The planet Venus looks to the naked eye like a bright star with magnitude ranging between about
-4 and -5
• The six brightest stars of the Plough are around magnitude 2, actually from 1.7 to 2.4
• The faintest stars visible to the naked eye, somewhere dark well away from streetlights and towns,
are between about 6.0 and 7.0, dependent on the individual’s eyesight
• The faintest objects ever seen with large telescopes (Keck, Hubble) are about magnitude 30
It is important to emphasise that we are talking here about apparent magnitudes: a number that
expresses how bright the object appears to us. They are also visual magnitudes, ignoring the possibility
that e.g. a hot star might appear much brighter in ultra-violet light than in visible.
A bright star might appear bright to us because it is extremely bright, as stars go, or because it is
closer to us than most stars. The absolute magnitude of a star expresses how bright it is intrinsically.
We remove the distance from the discussion by asking, ”what magnitude would the stars have if they
were all at the same distance from us?” Before we can talk about this, we need to discuss the
distances to the stars.
1.3 Distances
Distances to the stars are enormous by human standards. They must be much farther away than the
6300 km radius of the Earth or we would see changes in the shapes of the constellations when we travel
to other countries. We use the method of triangulation for this: determine the direction to the star
from two locations a known distance apart and use trigonometry (essentially, a scale drawing) to work
out the distance to the star. Astronomers call this method parallax. The distances are so great that
we can only make it work if we use telescopes and exploit the fact that the Earth travels round the
Sun, so that measurements carried out six months apart are separated by 300,000,000 km.
1.3.1 Small angles

Sizes and separations on the sky represent angles. Often, we have to deal with very small angles.
To do this a degree is divided into 60 arc minutes and an arc minute is divided into 60 arc seconds.
We use the symbols ’ for arc minutes and ” for arc seconds, i.e. 1° = 60’ = 60x60” = 3600”. 1”
is the size of a 1p coin held up 2.5 miles away. The Full Moon or the Sun are both 30’ across, half
a degree. The human eye can resolve detail (e.g. separate two stars of roughly similar magnitudes)
down to about 1’.
1.3.2 Distance units

We could use m or km, with scientific notation because the numbers are extremely big, but it’s much
more convenient to adopt a new “yardstick” appropriate to the distances involved. In the solar system
we learned to use the mean Earth-Sun distance, the Astronomical Unit (AU):
1AU = 150, 000, 000km
v1.0.2-4 Jan 2022 4

Even this is a bit small. Instead we use the light year (l.y.), the distance light travels in one year at
a speed of 300,000 km per second. Multiply this speed by 60 seconds in a minute, 60 minutes in an
hour, 24 hours in a day and 365.25 days in a year to get:
1 light year= 9.5 × 1012 km= 63,000 AU
For comparison the Sun is about 8 light minutes away, or about 22.5 years travelling at the speed of
a passenger jet aeroplane. Alternatively, we may use the parsec (pc):
1 pc= 3.26 l.y.
This may seem arbitrary at the moment. We will see that it is closely tied to the method used to
determine stars’ distances.
1.3.3 Parallax
As the Earth travels round the Sun in its annual (almost circular) orbit, a star should appear to trace
out a circle on the sky, as compared to much more distant, background objects. The further away the
star is, the smaller this circle will be. Only Earth’s orbital motion gives a large enough baseline for
changes of apparent direction to be observable. Once we know the size of this tiny little circle (i.e.
Figure 1.1: The principle of stellar parallax
its angular radius, P , measured in arc seconds) we can deduce how far away the star is, using a little
trigonometry. Let’s skip straight to the conclusion:
1
d (pc) =
P ′′
This is why the parsec is such a natural unit of distance: it’s the one that pops out of the method
of parallax. Such small angles can only be determined with a telescope, and indeed great care in
observation. This was only achieved for the first time in the 1830s, more than 200 years after the
introduction of the telescope to Astronomy. F W BesselI, the German astronomer, gets the credit
for the first determination of the distance to another star, 61 Cygni. His announcement was followed
rapidly by a similar measurement by the Scot Thomas Henderson who became the first Astronomer
Royal for Scotland. The nearest star, Proxima Centauri, has a parallax of about 0.75”. Thus its
distance is
1 1
d= = = 1.33pc = 4 light years
P 0.75
v1.0.2-4 Jan 2022 5

Parallax can work up to a few 100 pc but this is only a small fraction of the size of the Milky Way
galaxy. Other, indirect methods have to be used to get further (see other ExCos courses). The Milky
Way galaxy is about 30,000 pc across.
Example A star has a measured parallax of 0.02”. How far away is it? Answer: its distance d is
calculated thus:
1 1
d= = = 50pc = 163ly
P 0.02
1.3.4 Proper motion

Stars do physically move through space. The movement of a star across the sky is known as its
proper motion. The distances between the stars are so immense that these motions are not perceptible
without patient, careful observations with a telescope. However, they are measured routinely with
modern, sophisticated large telescopes. Many stars’ parallaxes and proper motions were measured
precisely by the European Space Agency Hipparcos space mission, at the start of the 1990s. Barnard’s
Star is the star with the greatest proper motion, just over 10” per annum.
v1.0.2-4 Jan 2022 6

Lecture 2
Star Basics 2
Introduction
In this lecture we start to learn about stars’ intrinsic properties; not just how they appear to us in
the sky. The starting point will be the knowledge of their distances, dealt with in Lecture 1. This
will allow us to deduce how bright they are “really”, comparing them with their distance from Earth
removed from the discussion. There are two ways of describing stars’ intrinsic brightness: luminosity
and absolute magnitude and we look at the meaning of both, and the range of values they take.
Then we will discuss their colours. First we look at how they can be described systematically, through
the spectrum or the idea of a Colour Index. Then we consider what “colour” is telling us: surface
temperature, through the idea of a blackbody.
2.1 Luminosity
How would we describe how bright a star is, “intrinsically”? I.e., so that we can compare two stars in
a way that does not depend on their distance from Earth.
The most physically sound way is by stating its luminosity, i.e. the rate at which it emits energy
in the form of light. A rate of generating or using energy, a power, is measured in Watts (W). So
we describe the luminosity of a star as a number of Watts - like light bulbs, but with MUCH bigger
numbers.
The Sun’s properties place it in the middle of the full range of stars. Its luminosity is 4 × 1026 W.
It is conventional to express other stars’ properties, including luminosity, in units of the value for the
Sun.
The symbol L is used for luminosity. Also the symbol ⊙ means “the Sun” so L⊙ means “the Sun’s
luminosity”.
2.2 Inverse Square Law

A star radiates light in all directions. As the light travels away from the star it gets spread out
evenly over the surface of a bigger and bigger sphere. The area A of the surface of a sphere is
proportional to the square of its radius r :
A = 4π r 2
So if we look at a star with an object of a fixed area, like a human eye, the fraction of the light from
the star gets smaller and smaller as it gets further and further away; the light has been shared out
over a bigger and bigger sphere, centred on the star. As a result, luminous (i.e. “shining”) objects
appear fainter in inverse proportion to the square of the distance.
7
LECTURE 2. STAR BASICS 2
The following picture may help. It has a wee star in the centre, the turquoise dot, surrounded by two
spheres. The bigger, black sphere has twice the radius of the smaller one. Imagine somebody sitting
on either of these spheres, looking at the star. The same amount of light energy passes through both
of them. On the bigger sphere this energy has been shared over an area that is four times bigger.
Thus the fraction intercepted by an object of fixed area, e.g. a human eye, will be four times less on
the second sphere because its area is four times greater.
Figure 2.1: Illustration of the inverse-square law for light
Example: Two stars have the same luminosity but are at different distances from us. Star A is 5 pc
away and star Bis 30 pc away. Which is brighter, by how much?
Answer: Since the stars have the same luminosity only their distance makes them appear brighter or
fainter. So star B is fainter, because it is further away. It is 6 times further away 6 = 30 ÷ 5 so it is
62 = 36 times fainter.
Example: Consider two identical stars, one 3 pc away and the other 6 pc away.
Answer: The more distant one is twice as far away so it will be 22 = 4 times fainter.
Example: Jupiter orbits the Sun at a distance of 5.2 AU. How much brighter or fainter would the Sun
appear in the sky on Jupiter, than on Earth?
Answer: On Earth the Sun is 1 AU away so it will appear fainter at Jupiter’s greater distance. Precisely
it will be 5.22 = 27 times brighter on Earth (to 2 significant figures).
2.3 Intrinsic brightness

2.3.1 Absolute magnitude
We would like to have a way of comparing stars’ intrinsic brightness, saying for example which are
brighter or fainter than our own Sun. We introduce the idea of absolute magnitude: the magnitude
that a star would have, if it were at a distance of 10pc. Once we know the distance to a star we can
calculate its absolute magnitude via the inverse square law.
When we talk about absolute magnitude we pretend all the stars are at the same distance. Comparisons
of absolute magnitude allow us to make statements like, “star A is brighter than stars B”, not just,
“star A looks brighter than star B from here on Earth”.
We will sidestep the mathematical details and go straight to some of the conclusions. The Sun’s own
absolute magnitude is 4.8. If it were 10 pc away it would be at the fainter end of the naked-eye stars,
v1.0.2-4 Jan 2022 8

faint enough to be lost in light pollution for most city dwellers.

The brightest stars have absolute magnitudes of -8 to -9. All these stars are 1000s of pc away. If we
placed one of these stars at the same distance as the bright star Vega, for instance, it would be almost
100 times brighter than the planet Venus. Clearly none of these extremely luminous stars is anywhere
near us.
The faintest stars have absolute magnitudes of about 15. Most of our nearest stellar neighbours are
of this sort but none is a naked-eye star, in spite of their proximity.
2.3.2 Distance modulus

Further away stars are fainter. If we imagine a whole set of identically luminous stars, at different
distances, the most distant will be the faintest, with the largest apparent magnitudes. Consider a star
with absolute magnitude M . In the sky this same star has an apparent magnitude m. If the star has
an apparent magnitude m greater than its absolute magnitude M , it must be further away than 10 pc.
If its apparent magnitude m is less than its absolute magnitude M it must be closer than 10 pc.
Thus we can view the quantity
m−M
as a measure of the star’s distance. This quantity is called the distance modulus. If it positive, the
star is further away than 10 pc. If it is negative, the star is closer than 10 pc.
We mentioned that the scale of magnitudes is a logarithmic scale. The distance modulus is simply
related to the logarithm of the distance of the star. Those who are not afraid of the mathematics
of logarithms can look in books or online for a more sophisticated treatment of the distance modu-
lus.
2.3.3 Luminosity
A measure of brightness effectively tells us the rate which light energy passes through a given area,
e.g. 1 square metre. If we know the distance r to the luminous object (e.g. star), we know that the
total light energy from the star has been shared out evenly across a huge sphere of area A = 4π r 2 .
Then we can scale up from the area of our small detector to the total area of the whole of the huge
sphere and calculate the total rate at which light energy was leaving the star: its luminosity.
The least luminous stars have about 10−5 (one hundred thousandth) the Sun’s luminosity. The most
luminous — rare, few in number, all very far away! — are 105 -106 times as luminous as the Sun.
2.3.4 Colour
To the naked eye the brightest stars show definite colours: red or orange (Betelgeuse, Aldebaran,
Arcturus), yellow (Capella), white (Sirius), blue (Rigel). Even a pair of binoculars greatly increases the
number of stars that show a definite tint. The colour of a star reveals its surface temperature.
The coolest stars are red, slightly hotter ones are orange, then yellow, white, blue. Most people would
say that the Sun is yellow coloured. However we have to remember that much of the blue light of
the Sun has been scattered across the sky to give us the blue sky. If we put all that blue light back
together with the yellow colour of the Sun’s bright disk, we would end up with a colour that is very
close to white. Viewed from above the atmosphere the Sun would be almost white.
The spectrum
“White light” is light that has all colours mixed together. If we pass light through devices made of
glass, etc., we can separate it out into its component colours. This is what happens in a rainbow.
We now have a more precise idea of “colour”: position in the spectrum. By placing a numerical
scale along the spectrum we can specify individual colours quite precisely. Such a scale has a natural
meaning in terms of a wave picture of light: the wavelength of the light. Visible light has wavelengths
between 400 and 700 nm, where “nm” stands for “nanometres” and “nano” means 10−9 . Infra-red
v1.0.2-4 Jan 2022 9

light has wavelengths longer than 700 nm, ultraviolet shorter than 400 nm. By measuring the amount
of light energy in equal intervals of wavelength we draw a graph of the brightness of the spectrum, in
terms of Intensity, the rate of arrival of energy per unit wavelength. The spectrum of sunlight is shown
in Fig. 2.2:
Figure 2.2: Spectrum of visible light from the Sun, showing spectral lines
The dark spectrum lines, absence of light at well-defined wavelengths, tell us a lot about the star’s outer
layers, as we see shortly. Glossing over the spectrum lines for a while, the broad shape of the curve
shown in Fig. 2.3 tells us about the dominant colour of the star, and the surface temperature.
Black bodies
A blackbody is a hypothetical object that absorbs all the radiation that falls on it. Such a body also
emits light, if it has a finite temperature. The spectrum of light that it emits depends only on
its temperature, not on what it is made of. Many real objects - stars, people, animals - are close to
the ideal blackbody, so that we can use the simple behaviour of the light emitted by a blackbody to
determine the temperature of the real body. Next, we show blackbody spectra for several temperatures,
given in K (Kelvin, the natural temperature scale for physical science). The intervals of temperature
are the same as the Celsius scale but 0K = −273◦ C.
They are brightest at a well-defined wavelength which is determined only by the temperature. At
4000K this lies just in the infra-red, while at 7000K it is just in the ultraviolet. If we can measure
the wavelength of maximum brightness we can determine the temperature.
More broadly, we understand the variation of stars’ colours in terms of the variation of the blackbody
curve with temperature. At lower temperatures most radiation in the visible part of the spectrum is
at the longest - red - wavelengths. At very high temperatures most of the visible radiation is of the
shortest - bluest - wavelengths. In between the net colour is a mix across the various wavelengths
present. As the temperature increases, the blackbody curve also gets higher: the emitting object gets
brighter. Because we know how bright a blackbody is, per unit of area, for a fixed temperature, if we
have measured the total luminosity of a star, we can deduce its surface area and thus its radius - even
although stars are just points of light on the sky, even in the largest telescopes.
Colour Index
To give a single, summary number that represents the colour of a star, and thus the surface temperature
on the assumption of a blackbody, we measure its magnitude when we use a filter that allows through
only light in a restricted range of wavelengths.
For instance, the B filter only allows through light between 380 and 480 nm — at the blue end of the
visible range. The V filter allows through light between 490 and 590 nm, in the middle of the visible
range. We measure the magnitude of the star with the B filter present, called B and the magnitude
with the V filter present, called V. B-V is a Colour Index and it gives a measure of the colour of
the star.
v1.0.2-4 Jan 2022 10

Figure 2.3: Solar radiation spectrum, compared with a best-fit Black-

body spectrum. Also shown are the peak wavelengths of sunlight incident
at the top of the Earth’s atmosphere, and of the (infra-red) radiation
re-emitted at sea level
Figure 2.4: Example Blackbody spectra, including the spectrum of a

blackbody with a temperature of 5777 K which best matches the spec-
trum of the Sun
• Suppose the temperature of the star is 3000 K. The magnitude V will be smaller than the
magnitude B, the star will look red and B-V will be positive.
• Suppose the temperature of the star is 10000 K. The magnitudes V and B will be similar, the
star will look white and B-V will be close to zero.
v1.0.2-4 Jan 2022 11

Figure 2.5: Illustration of the relationship between Blackbody spectra

and the B-V colour index derived from measurements of apparent mag-
nitude using filters
• Suppose the temperature of the star is 30000 K. The magnitude V will be larger than the
magnitude B, the star will look blue and B-V will be negative.
So the Colour Index is a single number that represents the colour of the star. Because it is a well-
defined number we could e.g. use it as one axis of a graph. If we assume we are dealing with a
blackbody we can calculate how the colour index varies with temperature and use this to deduce the
surface temperature of a star, e.g. from this graph:
Figure 2.6: Graph showing the relationship between blackbody temper-

ature and B-V colour index
2.3.5 Range of stars’ surface temperatures

The coolest stars are red. Their surface temperatures are in the region of 3000K, even slightly
lower. Objects with lower surface temperatures exist, but they are not stars in the sense of drawing on
energy generated by nuclear fusion of hydrogen into helium. The hottest stars have a pronounced
v1.0.2-4 Jan 2022 12

blue colour. Their surface temperatures are in excess of 20,000K. The Sun’s surface temperature is
5800K Once again it lies in the middle of the range found among stars. It is natural to ask: are hot
stars always highly luminous or can we have wee, faint hot stars? Can cool stars be very bright? The
answers to these questions are found in the Hertzsprung-Russell Diagram, which we discuss next.
This turns out to be core evidence for understanding what goes on inside a star, and why stars keep
shining over extremely long intervals of time. Before looking at such properties of very many stars
we continue our study of stars’ key parameters by looking at binary stars, which tell us about stars’
masses.
v1.0.2-4 Jan 2022 13

Lecture 3
Size and mass of stars; binary

stars
Introduction
In earlier lectures we looked at stars in the sky, learning about their names, about the magnitude
system that puts numbers to their “brightness”, about parallax and stars’ distances, about luminosity
and absolute magnitude, colour, Colour Index, black bodies and surface temperatures. In this lecture
we continue to learn about stars’ basic properties, answering the questions: how big is a star? What is
the mass of a star? (i.e. how much stuff is in a star - it’s a bit difficult to talk about their “weights”).
We answer the second of these questions by looking at so-called binary stars, pairs of stars orbiting
one another under the influence of their mutual gravitational attraction.
I will also continue with an underlying theme of actually looking at stars with your own eyes, to see at
least a little of the evidence for the statements that astronomers make about stars. Visual binary stars
give lovely targets for small telescopes or even binoculars. The brightest eclipsing binaries are regularly
variable stars whose variations can be followed, with some patience, using just the unaided eye.
3.1 Stellar radii

3.1.1 Sun
We know the Sun’s distance, rather precisely in the era of space exploration, and we know how big
it appears on the sky (30’ as discussed already). Thus, we can deduce how big it is. Its diameter is
1,400,000 km, roughly 110 times the diameter of the Earth, about 11 times the diameter of Jupiter,
biggest of the planets. As usual this forms a point of reference for discussing other stars.
It is worth mentioning that almost all other stars are so far away that their disks cannot be inspected
directly, even in the biggest telescopes. In only a handful cases of very large stars has it been possible
to measure the angular size of the disk or look for (big!) features on the star’s surface. So, we have
to reply on less direct methods to determine the size of other stars.
3.1.2 Blackbody radii

As we saw in Lecture 2, a blackbody is a theoretical object whose radiation can be understood using
fairly simple mathematics. Many real objects, like stars, are pretty close to black bodies so we can use
the properties of blackbody radiation to learn about the stars.
In particular, if we assume that we are dealing with a blackbody, a measurement of Colour Index
implies an estimate of the surface temperature of the star.
14
LECTURE 3. SIZE AND MASS OF STARS; BINARY STARS
We recall also that, as blackbodies get hotter, they also glow more brightly. In fact, the luminosity
of a blackbody is given by a simple calculation:
Luminosity = (Stefan-Boltzmann constant) × (star’s surface area) × (fourth power of the temperature)
(where “fourth power of the temperature” means calculating “temperature × temperature × tem-
perature × temperature”). The Stefan-Boltzmann constant is a constant of nature whose value is
well-known:
σ = 5.67 × 10−8 Wm−2 K−4
Thus, if we are told the area and the temperature of an object, we can calculate its luminosity.
Alternatively, if we know the luminosity of a star and its surface temperature (deduced from its colour),
we can calculate its surface area and thus its radius.
Example
Consider a star whose radius is the same as the Sun’s, 700,000 km (= 7 × 108 m), but whose sur-
face temperature is 7,000 K, as opposed to the Sun’s surface temperature of 5800 K. What is its
luminosity?
A star is a sphere (at least to a very good approximation), so its surface area is
4π r 2 = 4 × π × (7 × 108 m)2 = 6.2 × 1018 m2
The fourth power of its temperature is
7000K × 7000K × 7000K × 7000K = 2.4 × 1015 K4
So, its luminosity is
(5.67 × 10−8 × 6.2 × 1018 × 2.4 × 1015 )W = 8.4 × 1026 W
Recall that the Sun’s luminosity is 4 × 1026 W. A fairly slight increase in temperature makes a factor
of more than 2 difference to the luminosity.
Example
We deduce that the luminosity of a star is 0.002 of the Sun’s luminosity, i.e. 8 × 1023 W. Its surface
temperature is 3400 K. What is its radius? The fourth power of its temperature is
3400K × 3400K × 3400K × 3400K = 1.3 × 1014 K4
Thus, the surface area of the star is
Luminosity 8 × 1023
4 = = 1.1 × 1017 m2
(Stefan-Boltzmann constant) × (Temperature) 5.67 × 10−8 × 1.3 × 1014
If we divide the surface area of a sphere by 4π r and then take the square root, this tells us the radius
of the square. So r
area
radius = = 9.4 × 107 m = 94, 000km
4π
This is a much smaller star than the Sun.
Comment: The presence of the fourth power in the blackbody formula means that the luminosity is
very sensitive to the precise value of the temperature. If two stars are the same size but one is just
20% hotter than the other, the hotter one will have twice the luminosity.
v1.0.2-4 Jan 2022 15

3.1.3 Stars’ radii

When we use the blackbody formulae to estimate the radii of many stars, we find a huge variation. The
smallest stars, also usually cooler than the Sun, have just a few percent of the Sun’s radius, around
20,000 km. The largest stars on the other hand are enormous compared to the Sun, with radii that
can be measured in AU. Betelgeuse, in Orion, is an example of a red giant, with a radius of about 760
times the Sun’s radius (530 million km or 3.5AU).
White dwarfs are even smaller than the smallest star size quoted above, typically 5000 km radius
or thereabouts. These are very exotic objects, end points of stars’ lives but very different in physical
character from the hot globes of gas that have given birth to them. More about them elsewhere.
3.2 Binary stars

In a binary star, two stars find themselves in physical proximity. Each feels a pull of gravity towards
the other. Gravity keeps them close together, they move through space together and they orbit around
one another. Because gravity involves mass, we can use the details of the orbit to deduce the
masses of the two stars. So binary stars are immensely important in Astronomy, allowing us to
“weigh” stars. Roughly one half of stars are binary (or multiple - involving more than two) stars, so
that roughly two-thirds of all stars are members of binary or multiple systems.
How do we discover and study binary stars?
3.2.1 Visual binary stars

A visual binary is a binary star in which we can simply see the two stars. In a few cases this is possible
with the naked eye. Most people can distinguish two stars that are more than 1’ apart. In principle a
human eye can resolve down to 20” although this may be an optimistic estimate in most cases. Most
visual binaries are too close for the eye to separate but there are a few examples accessible to normal
eyesight, or at least to binoculars.
The constellation of Draco (the Dragon) lies roughly between Ursa Major and Ursa Minor. In the
dragon’s head, ν (“nu”) Draconis is a binary star whose components are separated by 1’. Both
components are about mag 5, so the star is certainly visible to the unaided eye in dark locations.
Binoculars will certainly show the two stars separately, although close together.
Most binary stars are so close together on the sky that they can only be separated in a telescope;
the human eye is not adequate. A good example is Castor, one of the two brightest stars of the
constellation of Gemini (the Twins). Its two components (A and B) have magnitudes of 1.9 and 3.0,
respectively. They are separated by about 3” at present, increasing as the stars get further apart in
their elliptical orbits. Its period is 445 years, which means that one may see it change significantly in a
human lifetime. Fig. 3.1 shows how the stars have moved relative to one another in the past 50 years
or so, and how they will appear in the coming centuries.
3.2.2 Eclipsing binary stars

In these cases we look close to the plane in which the two stars orbit one another. Normally we see
the light from both stars. Once per orbit, however, one of the stars passes in front of the other,
hiding all or part of its disk, and the two stars together dip in brightness. Usually the two stars are
too close together to be seen separately but the regular occurrence of eclipses reveals the star’s binary
nature.
The star Algol, in Perseus, is a famous example, recognised as for its regular eclipses since 1782.
The stars orbit one another in a time of 2.87 days. An eclipse takes place once per orbit, a couple
of hours under three days, so eclipses take place during darkness for a week or two, then they only
occur when the star is in the daytime sky, then they start to occur during the night again, and so
on. At minimum the star is about one third of its normal brightness, more than a magnitude fainter
than usual. Eclipses last a couple of hours either side of minimum. It is quite possible to follow
v1.0.2-4 Jan 2022 16

Figure 3.1: Apparent relative positions of the binary stars Castor A and
B over the period 1970 to 2400
eclipses of Algol with the naked eye. Predictions of the times of minimum are available online at
http://lackawannaastronomicalsociety.org/varstar.htm
Figure 3.2 illustrates its lightcurve (i.e. the graph of magnitude vs. time), based on measurements
assembled by the American Association of Variable Star Observers (http//www.aavso.org):
Figure 3.2: Light curve of the eclipsing binary Algol
The two stars are dissimilar in luminosity. Most of the time we see the light from both stars:
Once per orbit the fainter star hides part of the brighter one:
v1.0.2-4 Jan 2022 17

Then the brightness of both stars dips. The other eclipse, when the brighter star hides part of the
fainter, makes much less of a difference to the total magnitude of both stars:
and cannot be seen in the (amateur, visually estimated) data shown in the figure. More precise
measurements show that this secondary eclipse does occur, however.
3.2.3 Astrometric binaries

This phrase refers to those cases where only one star is actually seen, but its motion across the sky
involves a “wobble” due to an unseen companion. The most famous example must be Sirius. The
presence of an unseen companion was diagnosed in 1844 from the star’s movement across the sky —
not in a straight line, as expected for normal proper motion, but with a regular back-and-forward
motion superimposed. The existence of this faint companion, called Sirius B, was confirmed 18
years later when it was actually seen in a telescope. Sirius’ magnitude is -1.46 and the companion’s
magnitude is 8.4. Their separation is just 8” so Sirius B is very difficult to spot in the glare of Sirius
A.
Detailed study of Sirius B soon revealed that it was a very unusual object, just a few thousand km in
radius: a prototype of a white dwarf.
3.2.4 Spectroscopic binary stars

We learn that these stars are in a binary system from the Doppler shifts of their spectrum lines,
regularly moving towards and away from us as they orbit one another. Again, this is a method that
will reveal the presence of binaries when the two stars are too close together to separate even in large
telescopes.
3.3 Masses of stars in binary systems

3.3.1 Kepler’s Laws
Kepler’s Laws are familiar from the study of the solar system, where they describe the mathematical
character of the planets’ orbits around the Sun. The planets all have masses much less than the mass
of the Sun but there are still “Kepler’s Laws” for systems where the two bodies have similar masses,
like binary stars. Each star has an orbit in the shape of an ellipse, about a point lying between the
two stars called the centre of mass. The more massive the star, the smaller the ellipse. Both stars
v1.0.2-4 Jan 2022 18

orbit the centre of mass in the same length of time, so the more distant moves faster. This orbital
period is determined by the sum of the masses of both stars. The size of an ellipse is measured by
its semi-major axis, the distance along the long axis of the ellipse. The semi-major axes of the two
stars’ orbits are in inverse proportion to the masses. For example, shown in Fig 3.3 is a picture
of two stars’ orbits in a binary system:
Figure 3.3: Schematic showing the elliptical orbits of two stars, of un-
equal mass, about their common centre of mass
The less massive star has the larger orbit, coloured black. The more massive star has the red orbit.
The more massive star is 1.7 times the mass of the lighter one, so the lighter one’s orbit is 1.7 times
bigger (in semi-major axis) than the more massive one. The blue dot shows the two stars’ centre of
mass. In the solar system, the third of Kepler’s Laws says that the square of the period is proportional
to the cube of the semi-major axis of the orbit. The constant of proportionality involves the mass of
the Sun. In a binary system, once we measure the period and the size of the orbit (semi-major axis),
we know the value of the constant of proportionality for this particular system, and thus the sum of
the masses of the two stars. So: the period together with the size of the orbit (which we know if we
know the star’s distance) tells us the sum m1 + m2 of the masses; and the relative sizes of the two
orbits, which we can learn if we watch the stars over an extended period, tells us the ratio m1 /m2 of
the masses. Thus, we can work out the masses of the two stars separately.
3.3.2 Range of stellar masses

The least massive stars have about one tenth the mass of the Sun. Below this value the core of the
star does not get hot and dense enough for nuclear fusion of hydrogen. However, there are many less
massive globes of gas, dimly glowing star-like objects that cannot be regarded as huge planets: brown
dwarfs. The most massive stars are in the region of 100-120 times the mass of the Sun. As we shall
see, stars become more and more luminous as they get more massive. Above about 100 M⊙ , a star
would be so luminous that the pressure of its (light) radiation would blow off its outer layers: such
a star could not persist- although one or two observed objects seem to have masses well above this
upper limit.
v1.0.2-4 Jan 2022 19

Lecture 4
Hertzsprung-Russell (HR) Diagram

and stellar classification
Introduction
The first three lectures have shown how we can learn about stars’ physical characters in spite of their
great distances, by looking closely at the light they emit. We know how to determine their distances
from us, their luminosities (or, equivalently, absolute magnitudes), their surface temperatures and radii.
If they are members of binary systems, we can determine their masses (obviously we cannot do this
for every star, but it turns out that stars have enough regularity in their basic properties that a large
subset gives us information that helps us understand all of them).
We can mention a couple of other attributes. If they show proper motion, we can learn about their
velocities perpendicular to the line of sight. If there are systematic (i.e. not varying) Doppler shifts
in their spectra we can determine their velocities towards or away from us. The lines in the spectrum
reveal the chemical composition of at least their outer layers, as well as much other information about
temperatures, pressures etc.
Once we devise means of measuring basic properties, the next stage in any science is to look for
regularities, patterns in the properties of the objects we are studying. In the case of stars, we might
start from simple questions, like, “are all hot stars very luminous?” or “are all cool stars small?” We
find many of the answers to such questions in the Hertzsprung-Russell Diagram which summarises
some of the properties of many stars, and in particular shows the relationships that exist between stars’
luminosity and surface temperature.
This diagram turns out to be a tool central to understanding stars’ workings and life stories. It is
always abbreviated to HR Diagram. In astronomers’ minds, stars live there. It enables major steps in
classifying stars. Once we have described the HR Diagram and its main regions, we mention the Mass-
Luminosity relation, another empirically established relationship between two of the gross parameters
of stars. Then we will take a look at the main stellar classification schemes, which rely on surface
temperature and on luminosity. Finally, we take a closer look at a topic already alluded to, that of
spectrum lines. Stellar classification schemes give us a picture to fit them into, within which they
yield extraordinarily rich and detailed insights into stars’ physical state. In particular they let us answer
the remarkable question: what are stars made of?
4.1 Hertzsprung-Russell Diagram

For as many stars as we can, we determine the parallax, and thus the absolute magnitude, and the
B-V Colour Index. Let’s briefly recall the significance of these quantities.
20
LECTURE 4. HERTZSPRUNG-RUSSELL (HR) DIAGRAM AND STELLAR CLASSIFICATION
4.1.1 Absolute magnitude

Absolute magnitude, M , is the magnitude that a star would have at a distance of 10 pc. There is
nothing fundamental about this distance, but it is typical of stars in our comparative neighbourhood
of the Milky Way. We don’t cover any of the arithmetic of distances and absolute magnitudes in this
course, but that arithmetic is simplified by taking this as the standard distance. The faintest stars have
M = +15. The brightest have M = −9 or thereabouts. The luminosity of a star, the rate at which it
radiates energy in the form of light, is an alternative, more physical way of describing a star’s intrinsic
brightness.
4.1.2 B-V Colour Index

We measure the magnitude of a star with the standard B and V filters in place. The B filter allows
through only light at the blue end of the visible range, and the V filter allows through light in the middle
of the visible range (see earlier notes for the precise wavelength ranges). We also use the symbols Band
V to refer to the values of the magnitude determined with the appropriate filter in place.
Then the difference between the two magnitudes, B-V is a measure of the colour of the star. If
we are happy that the star is well approximated by a black body, we can deduce a surface temperature
from a value of B-V.
Most stars have B-V lying between -0.5 and 2.0 or slightly greater. The hottest stars have the most
negative values of B-V, and the coolest the largest, positive values.
4.1.3 HR Diagram
We draw a graph whose axes are B-V (x -axis) and absolute magnitude, M (y -axis). Each star for
which these quantities have been determined becomes a single point in this diagram. Once we include
as many stars as we can, the HR Diagram becomes a visual summary of stars’ properties, in which
relationships between them become evident. The diagram is named after the two astronomers, Einar
Hertzsprung and Henry Norris Russell, who first studied the correlation between these two quantities
in the 1910s.
Figure 4.1 shows an example HR diagram. Remembering the relationship between B-V and surface
temperature, we realise also that we have arranged the stars in such a way that their temperatures
decrease from left to right across the graph: the highest temperature stars are leftmost. This also
means, if we draw a line horizontally across the picture at a fixed absolute magnitude, that the stars
get bigger and bigger as we look from left to right across the diagram.
4.1.4 Regions of the HR Diagram

The broad band that stretches from upper left to bottom right (i.e. from hot and luminous to cool
and faint) is called the Main Sequence, abbreviated MS. Despite its name it does not represent a
“sequence”. In fact, a star spends the largest part of its life at one point on the MS, changing very
little; a sort of “mature adulthood”.
The coolest MS stars are also very faint. These are the physically smallest stars, a few % of the
radius of the Sun: red dwarfs. In fact, luminosity varies along the MS more rapidly than the changing
temperature alone can account for. The hottest MS stars are disproportionately luminous, telling us
that they are also larger. The largest MS stars are also the hottest, most massive and most luminous
so one can sometimes also come across reference to blue giants.
There is also a large population of red (cool) stars that are very luminous; not MS stars. As we
saw already, in order for these stars to have such high luminosities they must be extremely large.
Betelgeuse is a spectacular example, with a radius of about 760 R⊙ . Inevitably these stars are called
red giants.
We also notice a small, distinct population of stars in the bottom left of the diagram: hot but faint, i.e.
very small. These are white dwarfs, objects with masses comparable to the Sun but radii measured
in thousands of km.
v1.0.2-4 Jan 2022 21

Figure 4.1: Hertzsprung Russell diagram showing absolute magnitude on

the right hand side (based on the measured B filter magnitude) versus
B – V colour index along the bottom [image credit: Richard Powell —
http://www.atlasoftheuniverse.com/hr.html (adapted)]
4.2 Mass-luminosity relation

For MS stars there is a definite relationship between surface temperature and luminosity: the
hotter the star, the more luminous it is. When we combine luminosity estimates with mass estimates,
i.e. using stars in binary systems, we find that there is also a relationship between the mass of a
MS star and its luminosity, expressed in this diagram which has the log of the mass, in units of M⊙
on the x -axis
v1.0.2-4 Jan 2022 22

Figure 4.2: Mass-luminosity relation, showing that, roughly, luminosity

= (mass)3.5
4.3 Classifying stars

4.3.1 Spectral classification
Stars are classified by surface temperature, with each class labelled by a letter of the alphabet. These
letters arose from early attempts to classify stars by their spectrum lines. Once the true significance
of the various classes was understood, in terms simply of surface temperature, the letters wound up
jumbled. From hottest to coolest they are:
OBAFGKM
O type stars have surface temperatures of 20,000K and above. M type stars include the coolest, in the
region of 3,000K. Each class is subdivided and labelled with a number, 0 to 9, from hottest to coolest
(so e.g. F4 is hotter than F5 and cooler than F3). The Sun is a G2 star.
4.3.2 Luminosity classes

As we have seen already, some cool stars are small and faint — those found on the MS. Red giants
on the other hand are extremely large and luminous. A spectral classification, e.g. M3, doesn’t tell
us whether we’re dealing with a dwarf or a giant. So, a spectral classification is supplemented by a
luminosity class. Luminosity class is a little bit tricky. It consists of a Roman numeral from I to V,
i.e. one of I, II, Ill, IV, V. These letters are in decreasing order of luminosity, with V meaning “Main
Sequence” at the end —there are hardly any stars in the HR Diagram below the MS. Luminosity class
I stars are ”supergiants”. The luminosity class is added after the spectral classification, so Betelgeuse
is class Mil. Rigel is class B8I. The Sun is G2V. Proxima Centauri, our nearest stellar neighbour, is
M6V — it has absolute magnitude M = 15. Sirius is A1V — hotter than the Sun, more luminous but
also a MS star.
4.4 Spectrum lines

Splitting sunlight into its component colours we produce a spectrum
v1.0.2-4 Jan 2022 23

Stellar luminosity class Description Examples

Ia Bright supergiants Betelgeuse, Rigel
Ib Supergiants Antares
II Bright giants Polaris
III Giants Arcturus
IV Subgiants Procyon A
V Main-sequence stars and dwarfs The Sun, Sirius A
Table 4.1: Table of luminosity classes
Figure 4.3: Spectrum of visible light from the Sun, showing spectral lines
As well as the underlying black body form the spectrum exhibits many absorption lines at well-defined
wavelengths. Radiation from the Sun’s deeper layers is absorbed by atoms in the Sun’s atmosphere, at
well-defined wavelengths fixed by the interior details of the atoms. Each chemical element has
a characteristic set of wavelengths at which it exhibits lines in the spectrum, either absorption lines,
as here, or emission lines if we observe only a tenuous volume of the gas. So, the two lines marked
D are due to sodium atoms; the same colour found in sodium streetlights. The line marked C is due
to hydrogen atoms. The F and K lines are due to calcium; and so on. As we look at stars across
the spectral sequence, we find regular changes in the strengths of the spectrum lines as shown
in Fig. 4.4. These result from a combination of the physical conditions in the star’s atmosphere, so
Figure 4.4: Typical spectral lines seen across the OBAFGKM spectral
sequence
some lines are more likely in hotter gases. They also reflect the relative proportions of the various
chemical elements. We talk about the abundances of the chemical elements. The abundances found
in the Sun’s atmosphere are believed to be fairly representative of typical cosmic material in our part
v1.0.2-4 Jan 2022 24

of the Milky Way. The most abundant element Is hydrogen (73.9% by mass), then helium (24.0%
by mass), oxygen (1.04%), carbon (0.46%), neon (0.13%), iron (0.1%), nitrogen (0.096%), etc. The
spectral classification OBAFGKM also reflects changes in the strengths of spectrum lines, as shown in
Table 4.5
Figure 4.5: Changes in spectral lines across the OBAFGKM spectral

sequence
v1.0.2-4 Jan 2022 25

Lecture 5
Proper motions; composition and

interiors of stars
Introduction
In the last lecture we looked in detail at the Hertzsprung-Russell Diagram. The axes of this diagram are
surface temperature (or Colour Index) and luminosity (or absolute magnitude). Each star is represented
by a single point in this diagram. We saw that the majority of stars are Main Sequence stars, lying on
a band in this diagram that stretches from hot and luminous (upper left of the HR diagram) to cool
and faint (lower right of the HR diagram). From the Mass-Luminosity relation we will learn that the
most luminous stars are also the most massive. A significant minority of stars, however, are red giants,
cool but very luminous (and thus very large), or white dwarfs (hot, but tiny, 5,000 km radius).
We looked at the spectral classification scheme, which classifies starsby surface temperature: OBAFGKM
from hot to cold. We learned that stars are also classified by luminosity, in classes I-V where Vis Main
Sequence and IV- I are increasingly luminous, “subgiants” to “supergiants”. Thus, the Sun, for in-
stance, is a G2V star. We saw that the spectral classification is also mirrored in which spectrum lines
are strongest.
Today we take a close look at the interior of a star. We will find that it persists in a state of balance
between gas pressure outwards, and the pull of gravity inwards. This state of balance cannot persist
indefinitely because the star is losing energy in the form of light from its surface; however nuclear fusion
in the star’s core enables it to persist in this state of balance for much longer than would otherwise be
the case, about 10 billion years for the Sun. We start to look at nuclear fusion in detail.
Before this we take another quick look at stars’ true movement through space, their proper mo-
tion.
5.1 Proper motion

This phrase refers to the movement across the sky of a star, due to its true movement through space
(relative to the Sun). These proper motions cannot be large because the constellations still look more
or less the same as they did 2000 years ago.
Proper motion is measured in arc seconds per annum. Barnard’s star has the largest proper motion,
10.3” per annum.
If we know how far away the star is (its parallax) we can convert the proper motion into a distance
travelled through space per annum, perpendicular to the line of sight (we say transverse) and thus a
transverse speed. For example, this speed is 90 km/s for Barnard’s star.
26
LECTURE 5. PROPER MOTIONS; COMPOSITION AND INTERIORS OF STARS
The star may also be travelling along the line of sight - towards or away from us. The Doppler shift of
spectrum lines reveals this radial velocity.
The Sun travels once round the centre of the Milky Way galaxy in about 2.4 × 108 years, which implies
an average speed of about 230 km/s. Other stars will also have orbits about the centre of the Milky
Way, perhaps in a slightly different plane from ours. Stars further out will be travelling slower, stars
closer to the centre of the galaxy will be travelling faster. Thus, a star picked at random will have
some velocity relative to the Sun’s, which will be detectable as proper motion and/or radial velocity
(depending on how far away it is). Historically, study of many stars’ proper motions was a source of
information about the structure of the galaxy. Also, stars with the largest proper motions are likely to
be among the nearest stars, so these were targets for the earliest attempts to measure parallax.
Figure 5.1: Combined effects of parallax and proper motion for Barnard’s
Star. The straight line shows the average proper motion over several
years and superimposed is the annual to-and-fro motion of the star on
the sky due to the Earth’s orbit around the Sun
5.2 Inside the Sun

The Sun’s chemical composition is only slightly different from the “typical” cosmic abundances quoted
in the last lecture: 74% hydrogen (by mass), 25% helium, 0.57% oxygen, 0.24% carbon, plus smaller
v1.0.2-4 Jan 2022 27

quantities of all the other elements. This material is a gas all through its interior: atoms move around
independently of one another. Almost everywhere in a star (except maybe in the coolest parts of their
outer layers) it is too hot for molecules to survive.
The Sun, and all other stars, are in a state of hydrostatic equilibrium as shown in Fig. 5.2.
Figure 5.2: Schematic of hydrostatic equilibrium balancing gravity and

pressure inside a star
At each point in the star, the gas immediately below is slightly hotter and denser than that above, so
it exerts a net outward pressure (just as the air in the hot oven rushes out when you open the door).
On its own this gas pressure would blow the star apart. On the other hand, the gas in the Sun feels a
pull of gravity towards the centre. It could move in response to this, contracting and becoming denser.
Neither of these things happens because of the other: the inward pull of gravity is balanced by the
outward push of gas pressure all through the interior of the star.
However, this cannot be a permanent state of affairs. The star’s surface is hot, so it will lose energy in
the form of light. This would have to come from the heat energy of the star’s core, so the star should
slowly contract, on a time of about 30,000,000 years in the case of the Sun. This time is called the
Kelvin-Helmholtz timescale. We know from geological evidence that the Sun must have been shining
at a fairly steady rate for much longer than this.
In fact, conditions in the core of the Sun are so hot and dense that nuclear reactions can take place,
leading to the transformation of hydrogen into helium and an associated, steady, continuous release of
energy. This energy from nuclear fusion is sufficient to keep the Sun’s interior in the same hot state,
even although it is continually losing energy as light from its surface, for much longer than the known
age of the solar system.
We know how gravity works and we know how the pressure of a gas depends on temperature and density,
so we can take this mathematical principle of hydrostatic equilibrium, express it mathematically on a
computer and determine the conditions all through the interior of a star. We describe first the results
of this exercise for the Sun. As shown in Fig. 5.3, the inner 25% or so by radius of the Sun is the
core: this is where conditions allow the nuclear reactions to take place. Then there is a region known
as the radiative zone where energy is being carried outwards by radiation, i.e. light. If one were to
look down the gas below would look slightly hotter and brighter than the gas above, and the result is
a net flow outward of radiation. The gas does not move. In the outer 30% by radius the gas is in a
state of convection: continual circulation with bubbles of hot gas rising, cooling to their surroundings
and falling back down again. In the convection zone, movement of gas is the main means by which
energy is carried outwards, towards the surface. In the case of the Sun there is visible evidence: the
solar granulation, the mottling of the Sun’s surface, consisting of “granules” of about 1000 km size
with a characteristic lifetime of about 15 minutes.
v1.0.2-4 Jan 2022 28

Figure 5.3: Schematic of the Solar interior, showing the dimensions of

the core, radiative zone and convection zone
A more detailed picture is provided by the graph shown in Fig. 5.4, which shows how temperature
changes with distance from the Sun’s core. Note that the temperature in the core is about 15 million
K (1.5 × 107 K). Recall that the temperature at the surface is about 6000K, and that the Sun’s radius
is about 700,000km.
Figure 5.4: Graph of the temperature as a function of radius, from a

computer model that solves the equations (including hydrostatic equi-
librium) describing the Solar interior
To understand, particularly, some of the differences between stars of different masses (and thus different
location in the Main Sequence), we want to look in more detail at the business of nuclear fusion.
v1.0.2-4 Jan 2022 29

5.3 Nuclear energy generation

5.3.1 Constituents of the atom
An atom is the smallest bit of a chemical element. It is composed of a nucleus, which has positive
electrical charge, and electrons, which have negative electrical charge. The nucleus in turn is composed
of protons and neutrons. Protons have a positive electrical charge and neutrons have no charge. The
number of positive charges in the nucleus determines which chemical element we are dealing with.
Nuclei with the same number of protons but different numbers of neutrons are called isotopes. These
represent the same chemical element (the atom will still have the same number of electrons), but with
different atomic masses.
Electron mass = 9.11 × 10−31 kg

Proton mass = 1.673 × 10−27 kg
Neutron mass = 1.675 × 10−27 kg
Note that the neutron and proton have almost the same mass - protons and neutrons are both hadrons,
similar sorts of particle, although one has a positive charge and one doesn’t.
The simplest atom has just one proton in the nucleus. When this acquires an electron and becomes
electrically neutral it is an atom of hydrogen. Helium is the next “biggest” atom, with two protons
and two neutrons in the nucleus, and thus two electrons. If we remove one or more electrons from an
atom it has a net electrical charge and is called an ion. The atomic number, always called Z , is the
number of protons in the nucleus, so, for example, Z = 1 for hydrogen, Z = 2 for helium, Z = 6 for
carbon, Z = 10 for neon, Z = 26 for iron and Z = 92 for uranium - the heaviest naturally occurring
element.
5.3.2 E = mc 2 ; hydrogen fusion

The equivalence of mass and energy is one of the spectacular consequences of Albert Einstein’s Special
Theory of Relativity. These two previously unconnected ideas are seen to be two sides of the same
coin in the famous equation:
E = mc 2
This calculates the amount of energy (E, measured in Joules) associated with a mass m (measured as
always in kg). Here c is, as usual, the speed of light c = 3 × 108 ms−1 . Normally mass is mass and
energy is energy. But, because c is such a large number, however, if we can turn even a small fraction
of the mass of an object into energy, we get much more energy out than via any possible chemical
operation.
In particular, one proton is a nucleus of hydrogen. If we can persuade four protons to stick together
as two protons and two neutrons, we have made a nucleus of helium (a so-called alpha particle). We
say that fusion of hydrogen into helium has occurred. From above, the mass of four protons is
4 × 1.673 × 10−27 kg = 6.692 × 10−27 kg
However, the mass of one helium nucleus (well known from experiments in the laboratory) is 6.692 ×
10−27 kg, less than the mass of four protons. So, if we succeed in turning four protons into one helium
nucleus, what happens to the difference in the masses, the 0.047 × 10−27 kg? It is converted into
energy:
E = mc 2 = 0.047 × 10−27 kg × (3 × 108 ms−1 )2 = 4.26 × 10−12 J
Coming from the creation of a single helium nucleus this is a tiny amount of energy on everyday scales.
The point is that it is an enormous amount of energy per unit mass, compared with anything we could
get out of forming or breaking molecules (the usual business of chemistry).
v1.0.2-4 Jan 2022 30

How much energy do we get from 1 kg of hydrogen? We answer this question by calculating how many
helium nuclei we will form, i.e. working out 1 kg ÷ (mass of four protons)= 1.49 × 1026 . So in total
1 kg of hydrogen gives us
(4.26 × 10−12 J per helium nucleus formed) × 1.49 × 1026 = 6.37 × 1014 J
This is a lot of energy! For comparison, burning 1 kg of high quality coal (anthracite) gives something
like 3.5 × 107 J. The evidence on Earth for the reality of this super-powerful energy source is the
horrifying destructive power of the hydrogen bomb.
We know that the Sun’s luminosity is 4 × 1026 W (i.e. it needs 4 × 1026 J every second to keep it going)
and its mass is 2 × 1030 kg. How long could it keep shining at this rate powered by hydrogen fusion?
The sum to calculate this is
(2 × 1030 kg) × (6.37 × 1014 J per kg)
= 3.2 × 108 s = 1011 years
4 × 1026 J per s
We can keep the Sun shining steadily for a period much longer than its known age if it is powered by
nuclear fusion of hydrogen into helium.
5.3.3 Proton-proton chain and CNO cycle

The Sun is made primarily of hydrogen. If hydrogen nuclei collide and stick together to form helium
nuclei, we can keep the Sun shining steadily for tens of billions of years. In the Sun’s deep core, the gas
is hot enough and dense enough for the necessary nuclear reactions to occur. However, four protons
colliding with each other simultaneously would appear to be far too unlikely an event ever to occur on
its own, and in fact a sequence of simpler nuclear reactions has to take place, the end result of which
is the conversion of hydrogen into helium. Exactly how this happens turns out to be important for
understanding stars’ interiors.
In the Sun, this takes place via the proton-proton chain, which we can write as follows:
p + p → d + e+ + ν
p + d → 3 He + γ
3
He + 3 He → 4 He + p + p
Here p stands for a proton, d for a deuteron, a nucleus of hydrogen’s isotope deuterium (one proton +
one neutron), e + is a positron, identical with an electron except for its positive charge, γ is a photon (a
particle of light, in this instance in the gamma-ray part of the spectrum) and ν is a neutrino, an almost
massless particle that often accompanies such nuclear reactions. The neutrino escapes completely from
the Sun. The positron goes a short distance before annihilating when it meets an electron. The two
particles together cancel one another out and their mass is converted into gamma-ray photons, which
are absorbed and also go to heating the gas.
At the end of this process four protons have been converted into one helium nucleus. The energy
released goes into heating the gas of the solar core.
Protons all have positive electrical charge. Bodies with electric charge of the same sign (i.e. positive
or negative) repel one another. So, protons cannot get near enough to one another for these nuclear
reactions to take place, unless the temperature of the gas is very high so that all the particles are
moving very fast. Then they are able to approach one another closely enough - about 10−15 m - that
the electrostatic repulsion can be overcome by the attractive strong nuclear force.
In stars hotter than the Sun the CNO cycle becomes more important. Carbon, nitrogen and oxygen
v1.0.2-4 Jan 2022 31

nuclei play the role of catalysts; hence the name:

12
C → 13 N + γ
13
N → 13 C + e + + ν
13
C + p → 14 N + γ
14
N + p → 15 O + γ
15
O → 15 N + e + + ν
15
N + p → 12 C + 4 He
Here 12 C denotes the most common isotope of carbon with six protons and six neutrons, 13 C is a
nucleus of the isotope that has seven neutrons in the nucleus, etc. On the left-hand sides we added
four protons, one at a time; we started out with one 12 C nucleus and at the end of the process
we got a 12 C nucleus back, plus a 4 He nucleus. Again, four protons have been fused into a helium
nucleus.
This is important because we need higher temperatures still for the CNO cycle to operate (because it
involves nuclei with greater electrical charge, a greater electrostatic repulsion has to be overcome). It
becomes more important than the p-p chain for stars more massive than about 1.3M⊙ .
A star is a Main Sequence star while hydrogen fusion is taking place in its core. As we see later, fusion
of heavier elements than hydrogen is possible, but only when higher temperatures are attained in the
core of the star.
5.4 Stellar interiors along the Main Sequence: Convective cores

and envelopes
Above we drew a picture of the Sun’s interior. Nothing moves in the core and the radiative zone. The
temperature falls off only slowly as one comes out from the centre of the Sun, until we reach about
70% of the Sun’s radius. At this point the temperature is low enough for helium atoms to start to
retain at least one electron. This makes the stellar material more opaque so the temperature falls more
rapidly with distance. When the temperature falls rapidly the gas starts to move more readily and
convection sets in. The Sun has an outer convection zone.
Stars cooler than the Sun have bigger and bigger convection zones, in proportion to the size of the
star. At spectral type M (red dwarfs) the convection zone may stretch all the way through the star’s
interior.
Stars hotter than the Sun have smaller convection zones. Stars hotter than spectral type A no longer
have an outer convection zone at all. However, the CNO cycle has become dominant in energy
generation at this point. The rate at which the CNO cycle generates energy depends very strongly on
temperature, and the result is that the temperature falls off rapidly in the cores of these stars, as one
comes out from the core. The result of this steeply falling temperature in the core is convection again,
but in the core rather than an outer convective envelope.
Convective outer envelopes can amplify magnetic fields, producing “starspots”, filaments and promi-
nences, and flares. Convective cores mean that fresh hydrogen gets mixed in to the material of the
core from the outer parts of the star, making a difference to the rate at which the star evolves and
changes.
In Main Sequence stars we see a sort of “alchemy” taking place: the chemical element hydrogen is
transformed by nuclear reactions into the chemical element helium. The chemical composition of the
star is changing and eventually this drives more visible changes in the star. It also indicates that we
may find in the interiors of the stars the answer to a rather fundamental question: the origin of the
chemical elements. We look in more detail at this question in the next lecture.
v1.0.2-4 Jan 2022 32

Lecture 6
Nucleosynthesis I
Introduction
We have taken a look at stars’ insides, discussing how they differ along the Main Sequence. We noted
the key role of nuclear fusion, releasing energy that keeps the interior of a star hot. As a result, stars
are able to look more or less the same for very long periods of time (10 billion years in the case of the
Sun) even although they are losing energy from their surfaces in the form of light.
In the process we began to look closely at the structure of atoms, and the details of the atomic nucleus.
We have also begun to look closely at the entities involved in nuclear fusion, the constituents of the
nucleus and the nature of the chemical elements. We looked at the detailed chains of reactions that
can achieve hydrogen fusion: the proton-proton chain and the CNO cycle.
Protons and neutrons are elementary particles. Why do we find them in a dizzying array of combina-
tions, giving the richness of the periodic table and the associated families of isotopes? How did nature
manage to combine them in these many ways? Perhaps the answer lies in the interiors of stars. We
know now that conditions in their cores are hot and dense enough for nuclear fusion to occur, and we
can imagine that fusion processes involving nuclei heavier than hydrogen can occur. Could the origin
of the chemical elements lie in the cores of stars?
We shall look in detail at the nuclear reactions that can occur inside stars after hydrogen burning and
see how they account for the cosmic abundances of the elements. This process overall is known as
nucleosynthesis, i.e. the building up of larger nuclei from smaller ones.
6.1 Elementary particles

First, we take a slightly closer look at a couple of the entities involved in the nuclear reactions that
power the stars. We’ve already mentioned protons, neutrons and electrons. Most of this section is for
background; no need to remember the details, beyond the occurrence and nature of the positron in
fusion reactions.
Positrons are products of some of the nuclear reactions, or radioactive decays, involved in the p-p
and CNO processes. The positron is a particle of anti-matter. Anti-matter particles are a sort of
mirror image of normal matter. A positron is identical to an electron except for its electrical charge:
positive rather than negative (hence the name). If a positron and an electron meet, they annihilate
one another. The rest mass energy (mc 2 ) of both particles is converted into photons.
(Detail you do not need to remember: before annihilating the electron and positron may briefly form a
sort of atom called positronium, like a hydrogen atom but made of an electron and a positron instead
of an electron and a proton. The detailed state of the positronium atom determines whether two
photons are produced, each with an energy equal to the electron rest mass, or three, with the total
energy of 2mc 2 divided up between them.)
33
LECTURE 6. NUCLEOSYNTHESIS I
In the core of a star, a positron only travels a few cm before annihilating. The photons produced in
turn travel only cm before they collide with electrons, giving up their energy to the gas in this way.
So, the rest mass energy of the positron contributes to heating the gas in the star’s core.
The opposite process can happen if a photon has more energy than twice the value of mc 2 for an
electron. If the photon comes near an atomic nucleus some of its energy can go to producing an
electron-positron pair. Cosmic rays, energetic particles from the cosmos, continually bombard Earth’s
atmosphere, producing a cascade of radiation through the atmosphere including many electron-positron
pairs.
Positrons are not only found in stars’ cores and in cosmic rays. Some naturally occurring radioactive
substances decay by emitting a positron. Positrons are used deliberately in nuclear medicine. Anyone
who has had a PET scan in a hospital has encountered them (guess what the “P” stands for!). The
main point of mentioning these details is: the various particles that interact in the cores of stars
are also found, studied and even put to work here on Earth. The cores of stars are permanently
inaccessible, but their constituents are well known and understood.
The neutrino is the other, rather ethereal particle produced in these reactions. Neutrinos have a very
small mass. They interact very weakly with other particles so most of them escape freely from the
Sun, carrying off a small fraction of the energy released in the fusion reactions (2% in the case of the
p-p chain).
Neutrons, as mentioned, are very similar to protons in mass. They have no electrical charge, how-
ever.
The nuclear reactions that result in fusion do not occur easily. The elementary particles that collide
have electric charge, so there is an electrostatic force between them. Two particles will attract one
another if one has positive electrical charge and the other negative. However, if both are positively
charged (e.g. two protons) or negatively charged (e.g. two electrons), they repel one another, with a
force that gets greater the closer together they get. If two (positively charged) nuclei manage to
get as close as 10−15 m, the strong nuclear force can overcome the electrostatic repulsion between
them and glue them together. They need to have high energies for this to be possible, however, so it
can only happen in conditions of very high temperature, as we find in the cores of stars. The necessary
conditions also occur briefly in the hydrogen bomb, or in devices that attempt to achieve controlled
nuclear fusion in the laboratory, as an energy source, e.g. the Joint European Torus in Oxfordshire,
JET: https://www.euro-fusion.org/devices/jet/
As we see below, the neutron’s lack of electrical charge means that it can play an important role in
building up some of the chemical elements.
A last “particle” that plays an important role in these reactions is the alpha particle. An alpha
particle is a nucleus of helium, i.e. not an elementary particle like a proton. Many naturally occurring
radioactive substances decay by emitting an alpha particle — it is a natural unit that is favoured by
the workings of the nuclear world. Once we have helium in the core of a star, reactions involving alpha
particles play increasingly important roles.
6.2 Cosmic abundances

By studying spectrum lines in stars, and also by analysis of samples of material from meteorites, we
build up a picture of the relative abundances of the chemical element, summarised in Fig. 6.1
Hydrogen is the most abundant substance and we regard it as primordial — protons (i.e. hydrogen
nuclei) were present shortly after the Big Bang. The conditions following the Big Bang enabled
hydrogen fusion, for a brief period, and most cosmic helium is also primordial. We look in the nuclear
factories of stars to account for the rest of the elements. We note some features of the abundance
figure. In general, elements get rarer as the atomic number Z gets bigger — although lithium
(Z = 3), beryllium (Z = 4) and boron (Z = 5) are notable exceptions. Some elements are significantly
more abundant than their near neighbours e.g. nitrogen, atomic number Z = 7, is less abundant
v1.0.2-4 Jan 2022 34

Figure 6.1: Relative abundances of the chemical elements
than carbon, Z = 6, and oxygen, Z = 8. Fluorine (Z = 9) is much less abundant than oxygen or neon
(Z = 10).
We would like to be able to account for these features.
6.3 Main Sequence

Most stars are found on the Main Sequence because this is where they spend the greater part of their
lives. So, if we are looking at a sample of stars of random ages, and we pick one of them at random,
it is most likely to be in the Main Sequence stage of its life.
More physically, the Main Sequence stage of a star’s life is the period when it draws on hydrogen
fusion as its energy source. The processes described below belong to later stages of a star’s life
(and, as you will learn in LDS II, these stages do not occur for all stars).
6.4 Triple alpha process

The core of the Sun is at a temperature of 1.5 × 107 K, i.e. 15 million K. This is hot enough for the
reactions of the proton-proton chain to take place. Let’s think about a reaction involving two helium
nuclei. Both have charge Z = 2, so the repulsive force between them is 2 × 2 = 4 times greater than
for protons. So alpha particles will not get close enough together, in the core of the present-day Sun,
for fusion reactions of helium to occur to any degree.
However, helium fusion can occur at higher temperatures. If the core temperature of a star reaches
108 K, the triple alpha process can occur:
4
He + 4 He + 4 He → 12 C
12
This again generates energy from fusion, and results in synthesis of C nuclei.
6.5 Helium capture reactions

We can envisage fusion reactions involving heavier nuclei still, e.g. 12 C + 12 C → 24 Mg but these
need very high temperatures indeed (600 million K in this case), again because of the charges of
the colliding particles. Reactions involving nuclei with smaller charge are more likely, and particularly
v1.0.2-4 Jan 2022 35

reactions involving helium nuclei, e.g.

12
C + 4 He → 16 O (above 200 million K)
16
O + 4 He → 20 Ne (Neon)
20
Ne + 4 He → 24 Mg (Magnesium)
24
Mg + 4 He → 28 Si (Silicon)
The importance of these alpha capture reactions explains the successive peaks and dips in the abun-
dances, as we look at steadily increasing Z and in particular the jumps in two of the most abundant
substances. To get fluorine, for instance, we have to rely on much rarer reactions involving single
protons or neutrons, so much less fluorine is produced in stars than oxygen or neon.
6.6 Alpha process

In the core of star there is gas, of a chemical composition depending on the stage of the star’s life, and
radiation, i.e. photons. The photons are continually exchanging energy with the particles (electrons),
so the radiation and the gas reach a state of thermal equilibrium. The radiation will have more or less
a blackbody spectrum with the same temperature as the gas. At hundreds of millions of Kelvin this
means there will be significant numbers of gamma-ray photons, some of them with enough energy
to make nuclei break up. We call this photodisintegration. So more and more alpha particles are
available and larger nuclei can be built up. The next nuclei produced are 32 S (sulphur), 36 Ar (argon),
..., 52 Fe (iron) and 56 Ni (nickel). However,56 Ni is unstable. It decays to 56 Co (cobalt-56), which in
turn decays to ’56 Fe, the most common, stable isotope of iron. To get past iron we need to consider
other sorts of reaction.
v1.0.2-4 Jan 2022 36

Lecture 7
Nucleosynthesis II
Carrying directly on from Lecture 6, a nucleus can be pictured – very informally! – as a spring. The
electrostatic force is the spring, trying to push the protons apart but the strong force acts against the
spring keeping it compressed at a certain length. If we somehow switched off the strong force the spring
would spring back to its normal length and the energy stored in the coiled spring would be manifested
as energy of motion (in nuclear fusion, as heat energy of the surrounding gas).
Now suppose we combine two of these springs, bang them together. The result would be a new
combination of the two springs, but in a state of balance at a different length because of its new
circumstances. If the new spring is slightly less compressed than the two originals, then some
of the stored energy has been released as motion (heating).
7.1 Curve of binding energy

We talk about the binding energy of nuclei, the energy stored in the nucleus that can be released in
nuclear reactions. Its experimentally determined value, per nucleon (i.e. proton or neutron) is shown
in Fig. 7.1. If we combine nuclei in a way that involves moving upwards on the curve, then we get
Figure 7.1: Binding energy per nuclear particle as a function of atomic

mass number, indicating a peak value at the chemical element iron
37
LECTURE 7. NUCLEOSYNTHESIS II
energy out.
At the far left-hand side, fusing protons to make helium releases energy. At the far right-hand side,
breaking nuclei up (called fission) is the way to release energy. In between, at iron nuclei, we have a
most stable state. We cannot break iron nuclei up and get energy out; we cannot fuse iron nuclei to
make heavier nuclei and get energy out. If we want to make nuclei heavier than iron, we will need to
put energy in, i.e. actually take it from the heat energy of the core of the star. So, we can no longer
rely on simple fusion reactions to synthesise the elements heavier than iron.
Because iron is the most stable of the nuclei, eventually iron builds up in the core of a massive star
and it will develop an onion-like internal structure, with layers of different chemical composition
(see Fig. 7.2)
H
He
C
Ne
O
Si
Fe
Figure 7.2: Schematic of the internal structure of a star
7.2 Neutron capture

Instead we add nuclei, one neutron at a time. Neutrons are produced in many nuclear reactions, so
they are plentiful in the cores of stars. Because the neutron has no charge it is easier to capture them
in this way — there is no electrostatic repulsion to overcome. Adding neutrons does not change
the atomic number, only which isotope of the particular species we are dealing with. However, if we
add enough neutrons, eventually the resulting nucleus is radioactive; when it decays, we get a new
chemical species, with a different atomic number.
For example:
• 56
Fe captures neutron → 57 Fe
• 57
• 58
• 59
Fe is unstable (half-life of about a month), decays to 59
Co
• 59
Co captures neutron → 60 Co
• 60
Co decays to 60
Ni
• etc...
In typical stellar core conditions, each of the above steps takes about a year. Thus, these se-
quences of reactions occur in a fairly leisurely way and they are called s-process reactions;“s” stands
for“slow”.
s-process reactions can get us as far as 209 Bi (bismuth) — the heaviest, stable, naturally occurring
nucleus. To account for the heaviest nuclei found in nature, e.g. 232 Th (thorium), 238 U (uranium),
242
Pu (plutonium), we need to suppose that more extreme conditions occur, in the events of a supernova
v1.0.2-4 Jan 2022 38

explosion. Neutrons are much more plentiful at such times, conditions are generally more extreme and,
for instance, the 59 Fe would acquire many more neutrons before it has time to decay to 59 Co, opening
up new sequences of atomic and mass numbers. We call these sequences of neutron capture reactions
r-processes,“r” for“rapid”.
7.3 Evidence for stellar nucleosynthesis

Why do we believe these stories of stellar core nuclear physics? There are several sorts of evidence,
e.g.
7.3.1 Nuclear reaction and decay rates

In the laboratory we can isolate individual nuclei, accelerate them with electric fields to high energies,
typical of 100 million K gases, and bang them into targets to determine the rates at which reactions
occur. Then we take these known reaction rates, and the lifetimes of radioactive nuclei, combine them
with mathematical models for the interiors of stars and the ways in which they change, together with
the known age of the Milky Way galaxy, and calculate how many of each type of nucleus we expect
to create over the history of the galaxy. The results of this lengthy and involved process agree pretty
well with the graph of abundances shown above.
7.3.2 Technetium-99
Technetium (Tc) is element Z = 43. It has no stable isotope. Its longest-lived isotope, 98 Tc has a
half-life of 4.2 million years. 99 Tc, the longest-lived isotope produced in s-process reactions, has a
half-life of just 213,000 years. Any technetium present initially in the rocks of the Earth will have
decayed eons ago and it does not occur naturally on Earth in any measurable quantity.
Spectrum lines of Tc are definitely identified in spectra of red giant stars, however, indicating that
99
Tc must have been created in the star in the last few hundred million years — very short compared
to any star’s lifetime. For these lines to exist at all, synthesis of this isotope must be more or less
continual in these stars.
7.3.3 Supernova light curves.

The brightness of a supernova (the explosion that marks the end of the life of a massive star) decays
at first fairly rapidly and then more gradually over the months following the supernova explosion. The
character of the decay does not depend on details of the supernova, its luminosity etc.
Figure 7.3: Schematic light curve for a particular type of supernova,

showing the rapid nickel 56-driven decay followed by the more gradual
decline driven by cobalt 56
This can be explained if the luminosity is driven by the decay of radioactive nuclei created in the
explosion: first 56 Ni, which has a half-life of 6 days, and then its decay product, 56 Co, which has a
half-life of 77 days.
v1.0.2-4 Jan 2022 39

7.4 Conclusion
Hydrogen and helium in the cosmos are primordial but almost all the rest of the elements, with
all their detail of isotopic abundances, have been synthesised in nuclear fusion reactions taking
place in the interiors of stars, throughout their lives, and throughout the history of the whole of
the Milky Way galaxy (the exceptions as lithium, beryllium and boron, rapidly destroyed by nuclear
reactions in stars’ interiors, and present here on Earth only because of reactions of cosmic rays in the
interstellar medium).
We are able to observe the universe and think about it only because it is 13.7 billion years old, it
has stars in it, powered by nuclear fusion, and there has been enough time for these circumstances
to produce the mix of chemical elements that enabled our emergence as conscious life forms here on
Earth.
v1.0.2-4 Jan 2022 40

Lecture 8
Visible features of the Sun;

equivalents on other stars
Introduction
Among the stars, the Sun’s proximity allows us to inspect it in unequalled detail. For only a handful
of other stars can we get any sense at all of the appearance of their disks, sparse information on the
edge of what is possible with our largest and most advanced telescopes. These stars are all giants,
of course. We have no images of the surfaces of other Main Sequence stars, so the Sun once more
assumes great importance. In visible light its “surface”, the photosphere, does not display a huge array
of features but what’s there is important. The Sun’s hot, outer layers, the chromosphere and corona,
may be studied in particular emission lines, but our view of them is greatly enhanced when we study
their ultraviolet and X-ray emission. They are revealed as highly structured, dynamic regions in which
strong magnetic fields play a key role. Similar phenomena certainly exist in other stars to lesser or,
sometimes, much greater degrees. We take a close look at the Sun’s outer layers and at what can be
learned about similar phenomena in other stars.
Since we will be talking about the Sun, we first repeat the usual warning: NEVER look directly at
the Sun through a telescope or binoculars. You will collect the sunlight from a much larger area than
your eye and focus it all onto your retina. You risk serious damage to eyesight. Expensive solar filters
make it safe to use telescopes (these should go over the front of the telescope, not at the eyepiece),
or one can project an image of the Sun onto a piece of card and view that safely.
8.1 Photosphere and granulation

The Sun is a globe of gas, not a solid body like a planet, so in one sense it does not have a well-defined
surface. If one were to travel inside it on a spacecraft one would not have much sense of crossing a
boundary, only of the gas around becoming denser and harder to see through (eventually becoming
hot and dense enough to crush any feasible spacecraft).
Nonetheless, if we look through its tenuous outer layers towards the Sun there is a depth at which the
gas becomes cumulatively opaque. This happens in a fairly narrow layer, with some width depending
on the precise wavelength we look at and the viewing angle. This layer is the photosphere, the Sun’s
visible “surface”. It has a width of about 500 km (to put this in context, remember that the Sun’s
radius is R⊙ = 700, 000km). The gas there has a temperature of 5800 K and a density of about 10−4
kg per cubic metre (about 0.001 of air at sea level).
The photosphere shows little in the way of features but what there is holds intense interest. First there
is the solar granulation, a fine mottling across the whole of the solar surface, imperceptible except
when the Sun is viewed through telescopes. As we mentioned earlier, this is the visible manifestation
41
LECTURE 8. VISIBLE FEATURES OF THE SUN; EQUIVALENTS ON OTHER STARS
of the Sun’s outer convection zone.

Figure 8.1 shows part of the Sun’s surface about 30,000km across. The granulation is very clear to
Figure 8.1: Solar surface granules, taken by the Daniel K. Inouye Solar
Telescope
see, individual granules are bright in the middle where hot gas is rising upwards, and dark round the
edges (in the lanes of granulation) where the gas, now cooler, is falling back down. Individual granules
are 800-1500km across. Since it is the top of the layer of the sun in a state of convection (informally
“bubbling”), the granulation is consistently changing. Individual granules have a lifetime of about 15
minutes.
Granulation results from the presence of an outer convective envelope in the Sun. Recall that
convective envelopes shrink as we move up the Main Sequence stars, disappearing completely when
we reach stars about 1.3 times the mass of the Sun (i.e. 1.3M⊙ ) - up to about spectral type A. Stars
hotter than spectral type A have a convective core but no outer convective envelope. Thus, we expect
that something resembling granulation would exist in Main Sequence stars from spectral type M to
type A.
How would we test this? No telescope in existence can possibly resolve granulation on the surface of
another star. However, the gas in the granules is coming towards the observer or moving away,
so spectrum lines in its radiation will be blue- and red-shifted. Gas in some parts of the star’s
surface will be coming towards us, gas from other parts will be moving away. The result will, be a line
that is not definitely shifted to longer or shorter wavelengths but is broadened compared to the line
observed from gas at rest. This will be mixed in with broadening due to the finite temperature of the
gas (because emitting atoms have random speeds in all directions, with an average determined by the
temperature of the gas), and due to the rotation of the star (if we look anywhere close to the star’s
equatorial plane. It can still be unravelled, however, and has been used to study convection in other
stars.
Figure 8.2 schematically illustrates a graph of the brightness of a spectrum line as a function of
wavelength. The blue curve represents a line seen in the laboratory, from gas that is not moving
towards or away from the observer. The line always has some finite width in wavelength, but this will
be quite narrow. The green curve shows the same line but from gas moving away from the observer
so there is a Doppler shift. The orange curve represents the same line from a source that includes
a range of velocities, both towards and away from the observer (e.g. a star whose disk involves
granulation).
v1.0.2-4 Jan 2022 42

Figure 8.2: Schematic illustration of broadening of spectral lines
8.2 Differential rotation

The Sun does not rotate like a solid body (e.g. a football). It is a globe of gas and is free to rotate
differently at different latitudes. It rotates most rapidly at the equator, taking 25 days to go around
once. Near the poles it takes more like 35 days to rotate once.
8.3 Sunspots
8.3.1 The Sun
Sunspots are the other feature of the Sun’s photosphere seen in visible light. These are dark patches
up to 50,000 km across. They appear dark because they are about 2000 K cooler than the surrounding
photosphere. They are regions of strong magnetic field: about 0.3 Tesla, as opposed to the 0.003
Tesla found on average across the Sun’s disk (or the roughly 0.0005 Tesla found on Earth, that makes
the compass needle point north). The gas in the Sun’s outer atmosphere is a plasma (a gas with
a significant proportion of ions, in which electric and magnetic forces play important roles). This
strong magnetic field interferes with convection and prevents hot gas from arriving as easily at the
solar surface, so the material of the sunspot is cooler.
Small sunspots may appear and disappear in the space of a day or so; large ones may persist for a few
weeks. We can watch spots as the Sun rotates, possibly appearing over the limb (i.e. visible edge) of
the Sun, crossing the visible disk as the Sun rotates and, in the slightly unusual case of the biggest,
disappearing over onto the far side for a couple of weeks before reappearing again.
We cannot be certain very far in advance how spotty the Sun will be on any given day. However, the
number of sunspots does show some regular behaviour, reaching a maximum roughly every 11 years
— see e.g. http://sidc.oma.be/silso/monthlyssnplot. This (fairly) regular behaviour is known
as the sunspot cycle.
Does the Sun actually get fainter when there are spots on the visible disk? It is not obvious this will be
the case. Energy generation in the Sun’s core is totally unaffected by these surface phenomena. The
reduction in light from sunspots might be slightly compensated by other parts of the disk brightening
very slightly. Precise measurements to answer this question are only possibly from space, above the
variable screen of the atmosphere. They were first successfully carried out in 1980, by R Willson and H
Hudson; Hudson now works here in the University of Glasgow. These measurements show that the Sun
does indeed get very slightly fainter, at the few 10ths of a percent level, when there is a big sunspot
on the disk. However, they also show the Sun is on average slightly brighter (at the same 10ths of a
percent level) at the maximum of the sunspot cycle. It is also more variable because there are more
often sunspots.
It is worth mentioning that these very modest variations are too small to play any important role in
Earth’s climate. We need to look elsewhere for an explanation of climate change.
v1.0.2-4 Jan 2022 43

8.3.2 Other stars

Magnetic phenomena, like sunspots, are intimately tied up with the presence of a convection zone.
Differential rotation also plays an important role. The action of convection and differential rotation on
the magnetic field, together with the action of the magnetic field back on the ionised gas, combine to
generate the sunspot cycle.
Convection zones get bigger, relative to the star, as we look down the Main Sequence to stars cooler
than the Sun. Thus, it is not completely surprising that we find very large spots on some of these stars.
The key to finding them is in the discovery that sunspots make the Sun slightly fainter. The much
larger spots on these stars make a significant difference to brightness. A graph of the star’s magnitude
against time is called a lightcurve. The lightcurve of a star can reveal the presence of starspots and
computer methods can analyse the light curve to determine what the spots must look like. Figure 8.3
is from an article Roettenbacher et al., Astrophysical Journal 767 1 (2013). At the left-hand side is a
Figure 8.3: Spots on KIC5110407 inferred by brightness changes [from

Roettenbacher et al., Astrophysical Journal 767 1 (2013)]
graph of the light curve of a star, KIC5110407. The x -axis shows “phase”, i.e. what fraction of one
rotation we are looking at. At right are computer generated pictures of the star at two points in its
rotation, showing the presence of very large spots — much bigger than found on the Sun.
This sort of thing has been done for many stars. Such large spots are found on stars with deep convec-
tion zones, e.g. M type stars, particularly when they are young and still rotating rapidly, as expected
from our ideas of how convection and differential rotation produce strong magnetic fields.
8.4 Upper atmosphere: chromosphere and corona

8.4.1 Sun
Historically, solar eclipses have been very important in extending our knowledge of the Sun’s outer
layers. A very thin layer seen above the solar surface at eclipses was named the chromosphere on
account of its red colour. The chromosphere sits above the photosphere. It has a thickness of about
v1.0.2-4 Jan 2022 44

2000 km — still a very thin layer compared to R⊙ . The temperature increases very rapidly as we
move up through the chromosphere. A value of 20,000 K might be “typical”. Much of the light from
the chromosphere occurs in the hydrogen-alpha (referred to as Hα) spectrum line. Its wavelength is
656.3 nm so this line has a red colour, accounting for the appearance of the chromosphere. Telescopes
equipped with filters that accept only this line show detail of the chromosphere that is invisible in
normal white light. The chromosphere is revealed to be highly structured and irregular.
Hα images also reveal the existence of prominences, loop-like structures seen above the solar limb
like in Fig. 8.4 These represent cool material organised and supported above the surface by magnetic
fields. The same objects appear dark against the solar disk when they are known as filaments.
Figure 8.4: Hα prominence (Image credit: Jean-Pierre Brahic —

www.astrosurf.com/luxorion/research4.htm). Earth is shown to
indicate the scale of the feature
At its upper edge the chromosphere merges with the corona, the Sun’s tenuous outer atmosphere.
The corona is extremely tenuous, less than one millionth the density of the photosphere, but hot: the
gas has a temperature of 1— 2 million K. The existence of the hot corona was a surprise, historically,
and the question, “what heats the corona?” is a very important one in astrophysics.
The corona is apparent to the eye at the time of a total eclipse of the Sun but is otherwise far too faint
to be seen so close to the brilliant glare of the Sun (it is about one millionth as bright as the Sun’s
disk). Such high temperature gas does not really glow in visible light but light from the photosphere
scatters from electrons in the corona, so that corona! structures and features are seen in visible light
during the few minutes of a totality. The visible corona has a distinctive pearly white colour, and a
highly structured, irregular appearance with streamers protruding up to two or three solar radii from
the surface.
Such high temperature gases radiate mostly in the ultraviolet and X-ray parts of the spectrum, so we get
the best views of the chromosphere and corona by using telescopes and cameras that operate in these
parts of the spectrum (which must be placed about Earth’s atmosphere). The NASA Solar Dynamics
Observatory spacecraft routinely gives fabulous images of the Sun at several ultraviolet wavelengths
(http://sdo.gsfc.nasa.gov). Such images, like the one in Fig. 8.5, reveal that the Sun’s magnetic
fields play an essential role in structuring the atmosphere. The chromosphere and corona glow most
brightly above strong magnetic field regions (called active regions), which will usually be associated
with sunspots. Loop structures are ubiquitous. These are magnetic in nature. On Earth the strongest
magnetic fields are found near the poles. The Sun is different. The strongest magnetic fields are
found in the sunspots, or in the associated active regions. The high temperature gas of the corona
is almost completely ionised: all of the constituents of the gas have electrical charge. Magnetic fields
are important for charged particles. They can travel along the magnetic field easily but not across it.
One footpoint of the loop structures is a north magnetic pole and the other south. The field lines that
v1.0.2-4 Jan 2022 45

represent the properties of the magnetic field take on the character of structures because of the way
they channel and organise the ionised gas of the corona.
Figure 8.5: Ultraviolet image of the Sun. Note the bright active regions
at the right-hand side and the well-defined magnetic loops.
The ultraviolet brightness and extent of the corona go up and down with the sunspot cycle.
The corona is certainly dominated and structured by the presence of strong magnetic fields, and almost
certainly owes its existence to them. It might be heated by: magnetic waves that can travel from the
deeper atmosphere and dissipate in the corona; electrical currents flowing in the magnetic loops and
heating the gas; or lots of small, barely detectable solar flares. Solar flares are explosions, above
magnetic active regions, discussed in a little more detail below.
8.4.2 Other stars

X-ray astronomy started in the early 1960s. The first cosmic X-ray sources discovered were very
exotic objects indeed: binary X-ray sources including black holes or neutron stars; the Crab nebula.
X-ray astronomers at the start of the 1980s, with ever-improving instruments, were slightly surprised
to discover that many normal stars were also X-ray sources, because of their hot corona. So other
stars also have hot coronae — again, as long as they have outer convection zones, so Main Sequence
stars like the Sun and cooler. Stellar chromospheres also exist, revealed primarily by their spectrum
lines.
8.5 Flares and Coronal Mass Ejections

8.5.1 Sun
Solar flares are explosions in the solar atmosphere, above the active regions with their strong magnetic
fields. They last tens of seconds to minutes and involve the release of up 1026 Joules from the current
system flowing in the magnetic fields, which has been “wound up” and stressed by the gas moving in
the convective zone. Viewed in the Hα spectrum line, patches on the solar surface are seen to brighten
up and the corona and chromosphere glow brightly in X-rays and ultraviolet. Gas is heated to tens
of millions of K, and subatomic particles (electrons and ions) are accelerated by electric fields to high
energies. Flows of gas are set in motion. Radiation is emitted across the electromagnetic spectrum, all
the way from radio to X-rays and, sometimes, gamma-rays. Energetic ions and electrons are released
v1.0.2-4 Jan 2022 46

into interplanetary space. The largest flares could constitute a radiation hazard for astronauts working
in space.
Flares are often accompanied by Coronal Mass Ejections (CMEs). These are clouds of gas, accom-
panied by magnetic fields, ejected from the Sun into interplanetary space. If they arrive at Earth, they
may trigger disturbances in the Earth’s magnetic field that produce the aurora borealis and aurora
australis (”Northern/Southern Lights”).
8.5.2 Other stars

Some of the most active M type stars experience flares that are up to 100,000 times as big as those seen
on the Sun (in terms of energy radiated). The stars are known to be variable, for instance because
of the presence of large spots, so astronomers monitor them closely enough to simply see the star
brighten when these flares occur. Some of the largest have been detected in X-rays by space missions.
These huge flares would constitute a significant radiation hazard for life on planets around such stars.
Occurring on stars much smaller and fainter than the Sun (remember that the faintest M dwarfs are
as faint as 100,000th of the Sun’s luminosity), they may make a much bigger difference to the total
brightness of the star.
On 23 April 2014 a flare occurred on the star DG Canes Venatici. In just a few tens of seconds the
star brightened from magnitude 11 — invisible in binoculars — to about 7, easily visible in binoculars.
This increase would have been obvious and dramatic to anybody watching this particular star in a small
telescope. Such a “superflare” is on a much greater scale than anything that has ever happened on
the Sun.
8.6 Conclusion
The Sun’s outer convection zone and its differential rotation combine to generate strong magnetic
fields, in a cyclical way. Similar phenomena occur in other stars, especially when these rotate rapidly
and when they have deep outer convection zones. The understanding of magnetic fields in plasmas
gained from the Sun serves as basic knowledge for understanding more spectacular objects further
away.
v1.0.2-4 Jan 2022 47

Revision History
Revision Date Author(s) Description

1.0.0 12 Dec 2021 Robert Bennett Initial 2022 version
1.0.1 2 Jan 2022 Robert Bennett Accessibility improvements, typos corrected
1.0.2 4 Jan 2022 Robert Bennett Typos corrected
48

Stars N Stuff

Uploaded by

Copyright:

Available Formats

Stars N Stuff

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stars N Stuff

Uploaded by

Copyright:

Available Formats

ASTRO1004 XCOS1Y

Life and Death of Stars I

1.1 Star names

1.1.2 Proper names

1.1.3 Catalogue numbers

1.2 Magnitude system

v1.0.2-4 Jan 2022 3

1.3.1 Small angles

1.3.2 Distance units

1AU = 150, 000, 000km

v1.0.2-4 Jan 2022 4

Figure 1.1: The principle of stellar parallax

v1.0.2-4 Jan 2022 5

1.3.4 Proper motion

v1.0.2-4 Jan 2022 6

2.2 Inverse Square Law

Figure 2.1: Illustration of the inverse-square law for light

2.3 Intrinsic brightness

v1.0.2-4 Jan 2022 8

faint enough to be lost in light pollution for most city dwellers.

2.3.2 Distance modulus

v1.0.2-4 Jan 2022 9

v1.0.2-4 Jan 2022 10

Figure 2.3: Solar radiation spectrum, compared with a best-fit Black-

Figure 2.4: Example Blackbody spectra, including the spectrum of a

v1.0.2-4 Jan 2022 11

Figure 2.5: Illustration of the relationship between Blackbody spectra

Figure 2.6: Graph showing the relationship between blackbody temper-

2.3.5 Range of stars’ surface temperatures

v1.0.2-4 Jan 2022 12

v1.0.2-4 Jan 2022 13

Size and mass of stars; binary

3.1 Stellar radii

3.1.2 Blackbody radii

4π r 2 = 4 × π × (7 × 108 m)2 = 6.2 × 1018 m2

The fourth power of its temperature is

7000K × 7000K × 7000K × 7000K = 2.4 × 1015 K4

So, its luminosity is

(5.67 × 10−8 × 6.2 × 1018 × 2.4 × 1015 )W = 8.4 × 1026 W

3400K × 3400K × 3400K × 3400K = 1.3 × 1014 K4

Thus, the surface area of the star is

v1.0.2-4 Jan 2022 15

3.1.3 Stars’ radii

3.2 Binary stars

3.2.1 Visual binary stars

3.2.2 Eclipsing binary stars

v1.0.2-4 Jan 2022 16

Figure 3.2: Light curve of the eclipsing binary Algol

v1.0.2-4 Jan 2022 17

3.2.3 Astrometric binaries

3.2.4 Spectroscopic binary stars

3.3 Masses of stars in binary systems

v1.0.2-4 Jan 2022 18

3.3.2 Range of stellar masses

v1.0.2-4 Jan 2022 19

Hertzsprung-Russell (HR) Diagram

4.1 Hertzsprung-Russell Diagram

4.1.1 Absolute magnitude