Themes › Theme E Nuclear and quantum physics

E.5Fusion and stars

Every star is a giant fusion reactor, held together by its own gravity. This topic explains how fusion powers stars, why a star stays stable for billions of years, how we measure the distances and sizes of stars we can never visit, and how the Hertzsprung–Russell diagram sorts them and reveals their life stories. It brings together E.3, B.1 (black-body radiation) and D.1 (gravity). There is no extra HL content in E.5.

Knowledge and science

Nature of science

MeasurementPatternsCollaborationTechnology

Measurement. Stellar parallax is so small that it was first measured only in 1838, by Friedrich Bessel, for the star 61 Cygni: about a third of an arcsecond. Its absence had long been used as an argument that the Earth doesn't move.

Collaboration. At Harvard, Annie Jump Cannon classified the spectra of around 350 000 stars by hand, creating the O B A F G K M sequence. Cecilia Payne-Gaposchkin showed in 1925 that stars are mostly hydrogen and helium; at first, senior astronomers persuaded her to call the result "almost certainly not real".

Patterns. Around 1910 Ejnar Hertzsprung and Henry Norris Russell each plotted the luminosity of stars against their temperature, and found the stars fell into a few distinct groups.

Technology. The Gaia space telescope (2013–2025) measured the parallax of over a billion stars, with a precision hundreds of times better than is possible from the ground.

ToK: questions to think about

  • Can we understand what we can't imagine? The nearest star is 40 trillion km away. Do numbers like this give us knowledge, or just the feeling of it?
  • How can we study a process we can't watch? A star's life lasts millions or billions of years, but we only ever see each star at one moment. How do snapshots of many stars become a story about one?
  • Who decides what is "real"? Payne-Gaposchkin's correct result was doubted because it went against expert opinion. When should a new result overturn a consensus?
  • Is the light we see the star we think it is? Light from distant stars left them long ago. Some may no longer exist. What does it mean to observe the past?

How do physics, NoS and ToK fit together? →

1. Nuclear fusion

Nuclear fusion is the joining of two light nuclei to form a heavier one. Light nuclei are on the steep left side of the binding energy curve (E.3), so fusing them increases the binding energy per nucleon and releases energy. Per kilogram of fuel, fusion releases even more energy than fission.

The main process in the Sun is the proton–proton chain. Its overall effect is to turn four protons into one helium-4 nucleus:

$$4\,{}^{1}_{1}\text{p} \rightarrow {}^{4}_{2}\text{He} + 2\,{}^{\;\;0}_{+1}\text{e} + 2\nu_e$$

The positrons quickly annihilate with electrons, releasing more energy as gamma rays.

The Sun photographed in extreme ultraviolet light and coloured orange-yellow. Its surface is mottled, with bright active regions, and loops and wisps of glowing gas rise from its edge.
Our nearest star, photographed in extreme ultraviolet light (shown in false colour) by NASA's Solar Dynamics Observatory. The light comes from very hot plasma in the Sun's outer atmosphere. Every second, about 4 million tonnes of the Sun's mass is converted into the energy it radiates, all of it released by fusion in the core. Image: NASA/SDO (AIA), Wikimedia Commons, public domain.

Worked example: energy from the proton–proton chain

Using atomic masses (hydrogen-1 = 1.007825 u, helium-4 = 4.002603 u), find the energy released when four hydrogen atoms become one helium atom.

$\Delta m = 4 \times 1.007825 - 4.002603 = 0.028697$ u, so $E = 0.028697 \times 931.5 = 26.7$ MeV.

Using atomic masses automatically includes the energy from the positrons annihilating. About 0.7% of the original mass becomes energy.

Worked example: deuterium–tritium fusion

Fusion reactors on Earth aim to use $^{2}_{1}\text{H} + {}^{3}_{1}\text{H} \rightarrow {}^{4}_{2}\text{He} + {}^{1}_{0}\text{n}$. The masses are 2.014102 u, 3.016049 u, 4.002603 u and 1.008665 u. Find the energy released.

$\Delta m = (2.014102 + 3.016049) - (4.002603 + 1.008665) = 0.018883$ u, so $E = 17.6$ MeV.

Worked example: how fast is the Sun losing mass?

The Sun's luminosity is $3.83 \times 10^{26}$ W. $\dfrac{\Delta m}{\Delta t} = \dfrac{L}{c^2} = \dfrac{3.83 \times 10^{26}}{(3.00 \times 10^8)^2} = 4.3 \times 10^9\ \text{kg s}^{-1}$: over four million tonnes every second. Even so, in its whole lifetime the Sun will turn less than 0.1% of its mass into energy.

2. Conditions for fusion

Nuclei are positive, so they repel. To fuse, they must get close enough (about $10^{-15}$ m) for the strong force to take over. That needs:

In a star, both come from gravity: as a cloud of gas collapses, its gravitational potential energy becomes thermal energy, and the core becomes hot and dense. On Earth, reactors such as tokamaks hold a plasma at over $10^8$ K with strong magnetic fields, because no material container could survive. Getting more energy out than is put in, for long periods, is still being worked on.

3. The equilibrium of a star

A star in equilibrium. Gravity pulls all its layers inwards; the pressure of the hot gas and of radiation from fusion in the core pushes outwards. The two balance. coregravity(inwards)radiation + gaspressure (outwards)
A stable star: gravity pulls inwards; the pressure of radiation and hot gas, kept up by fusion in the core, pushes outwards.

A main-sequence star is stable because two effects balance at every layer:

The balance is self-correcting. If the core contracts slightly, it heats up, fusion speeds up, the pressure rises and pushes it back out. If it expands, it cools, fusion slows, and gravity pulls it back. When the fuel in the core runs out, the balance is lost and the star changes (section 9).

4. Astronomical distances

Metres are inconveniently small for astronomy, so three larger units are used:

An arcsecond is $\frac{1}{3600}$ of a degree. The light year is a distance, not a time: $3.00 \times 10^8 \times 365.25 \times 24 \times 3600 = 9.46 \times 10^{15}$ m.

Worked example: converting units

A star is 8.6 ly away (about the distance of Sirius). Express this in metres, parsecs and AU.

Metres: $8.6 \times 9.46 \times 10^{15} = 8.1 \times 10^{16}$ m.

Parsecs: $\dfrac{8.6}{3.26} = 2.6$ pc.

AU: $\dfrac{8.1 \times 10^{16}}{1.50 \times 10^{11}} = 5.4 \times 10^{5}$ AU.

For scale: light takes about 8 minutes to reach us from the Sun, the nearest other star is about 4.2 ly away, and our galaxy is about 100 000 ly across.

5. Stellar parallax

Hold up a finger and look at it with one eye, then the other: it seems to jump against the background. This is parallax. As the Earth orbits the Sun, a nearby star seems to move slightly against much more distant background stars, and returns to the same place every year.

Stellar parallax. The Earth is shown in January and in July, on opposite sides of its orbit around the Sun, 1 AU from the Sun each time. Lines from each position to a nearby star point to different places on the distant background stars. The parallax angle p is the angle at the star between the line to the Sun and the line to the Earth; d is the distance from the Sun to the star. SunEarth in JanuaryEarth in Julynearby stard1 AUp✦✦✦✦
Six months apart, the nearby star is seen in different directions against the distant stars. The parallax angle p is half the total angle the star appears to shift. The diagram is hugely exaggerated: real parallax angles are under one arcsecond.

In the right-angled triangle Sun–Earth–star, $\tan p = \dfrac{1\ \text{AU}}{d}$. Because $p$ is tiny, $\tan p \approx p$ in radians, so $d = \dfrac{1\ \text{AU}}{p}$. Measuring $p$ in arcseconds and $d$ in parsecs makes this simple:

$$d\,(\text{parsec}) = \frac{1}{p\,(\text{arcsecond})}$$

The smaller the parallax angle, the further away the star.

Worked example: a nearby star

A star has a parallax angle of 0.25 arcseconds. How far away is it, in parsecs, light years and metres?

$d = \dfrac{1}{0.25} = 4.0$ pc $= 4.0 \times 3.26 = 13$ ly $= 13 \times 9.46 \times 10^{15} = 1.2 \times 10^{17}$ m.

Limits. From the ground, the atmosphere blurs images, so angles below about 0.01″ can't be measured: parallax works only up to about 100 pc. Space telescopes like Gaia reach thousands of parsecs, but that is still only a small part of our galaxy. Other methods are used further away.

6. What starlight tells us

7. Finding the radius of a star

Stars are too far away to measure their size directly. But a star's luminosity depends on its surface area and temperature, by the Stefan–Boltzmann law (B.1):

$$L = \sigma AT^4 = 4\pi R^2\sigma T^4 \quad\Rightarrow\quad R = \sqrt{\frac{L}{4\pi\sigma T^4}}$$

The method: find $T$ from the peak wavelength (Wien's law), find $L$ from the apparent brightness and distance, then find $R$.

Worked example: a hot star

A star's spectrum peaks at 290 nm, and its luminosity is $1.5 \times 10^{28}$ W (about 40 times the Sun's). Find its radius.

$T = \dfrac{2.9 \times 10^{-3}}{2.90 \times 10^{-7}} = 1.0 \times 10^4$ K.

Bottom of the fraction: $4\pi\sigma T^4 = 4\pi \times 5.67 \times 10^{-8} \times (1.0 \times 10^4)^4 = 7.1 \times 10^{9}$.

$R = \sqrt{\dfrac{L}{4\pi\sigma T^4}} = \sqrt{\dfrac{1.5 \times 10^{28}}{7.1 \times 10^{9}}} = 1.5 \times 10^9$ m, about twice the Sun's radius.

Worked example: comparing with the Sun

A red giant is 100 times as luminous as the Sun, with a surface temperature of 4000 K (the Sun's is 5800 K). How many times bigger is it?

From $L \propto R^2T^4$: $\dfrac{R}{R_\odot} = \sqrt{\dfrac{L}{L_\odot}} \times \left(\dfrac{T_\odot}{T}\right)^2 = \sqrt{100} \times \left(\dfrac{5800}{4000}\right)^2 = 10 \times 2.1 = 21$.

It is cooler than the Sun but much brighter, so it must be much bigger.

8. The Hertzsprung–Russell diagram

The HR diagram plots each star's luminosity (vertical axis, usually relative to the Sun) against its surface temperature (horizontal axis). Both scales are logarithmic, and temperature increases to the left.

A Hertzsprung–Russell diagram on a black background. Luminosity compared with the Sun runs up the vertical axis from 10 to the minus 5 to 10 to the 6, and surface temperature runs along the bottom from about 40 000 K on the left to below 3000 K on the right, over a colour bar from blue to red. Stars are drawn as coloured balls, bigger for bigger stars. The main sequence runs diagonally from large blue stars at the top left, through the Sun, to small red stars at the bottom right, ending with the red dwarf AB Doradus C. Large orange and red giants lie above the main sequence on the right, very large supergiants lie across the top, and white dwarfs lie in a pale band below and to the left.
The HR diagram. Temperature increases to the left, and the star sizes are drawn to show that giants and supergiants are much bigger than main-sequence stars, and white dwarfs much smaller. Image: ESO, Wikimedia Commons (original at eso.org), CC BY 4.0. Resized.

Lines of constant radius. From $L = 4\pi R^2\sigma T^4$, stars of the same radius lie along straight lines on the HR diagram, sloping down from top left to bottom right (you may be asked to sketch them). Moving up and to the right means bigger stars: that is why giants and supergiants are so large and white dwarfs so small.

An HR diagram with luminosity compared with the Sun on the vertical axis, from 10 to the minus 4 up to 10 to the 7, and temperature on the horizontal axis, from 50 000 K on the left to 2000 K on the right. Parallel dotted lines slope down from top left to bottom right, labelled R equals 0.001, 0.01, 0.1, 1, 10, 100 and 1000 solar radii, with the bigger radii further up and to the right. A thick black curve, the main sequence, runs from the top left to the bottom right, with tick marks giving star masses from 40 solar masses at the top down to 0.1 solar masses at the bottom. Named stars are marked: HD 93129 A, Beta Scorpii, Vega, Sirius, Altair, the Sun, 61 Cygni and Barnard's star on the main sequence; Polaris, Aldebaran, Mira, Mu Cephei and Betelgeuse above and to the right, between the 10 and 1000 solar-radius lines; Sirius B and Procyon B at the bottom left, near the 0.01 solar-radius line.
The same axes with lines of constant radius (dotted) and star masses along the main sequence. The Sun sits on the $R = 1\,R_\odot$ line. Betelgeuse lies between the 100 and 1000 $R_\odot$ lines, and the white dwarfs Sirius B and Procyon B lie near the 0.01 $R_\odot$ line, about the size of the Earth. Notice that the more massive a main-sequence star is, the hotter and more luminous it is. Ignore the right-hand scale (absolute magnitude), which is not in the course. Image: AstroOgier, Wikimedia Commons, CC0 (public domain). Resized.

The HR diagram also lets us estimate the distance of a main-sequence star too far away for parallax: its spectrum gives its temperature, the diagram gives its luminosity, and its apparent brightness then gives its distance.

9. How stars evolve

Every star begins as a cloud of gas and dust (a nebula), which collapses under gravity. As it shrinks it heats up, forming a protostar. When the core reaches about $10^7$ K, hydrogen fusion starts and the star joins the main sequence, where it spends about 90% of its life. What happens next depends on its mass.

More massive stars burn out faster. A massive star has much more fuel, but its core is hotter and denser, so it fuses far faster and is enormously more luminous. The Sun will spend about 10 billion years on the main sequence; a star of 25 solar masses only about 7 million years.

Low-mass stars (up to about 8 solar masses), like the Sun

  1. Hydrogen in the core runs out. The core contracts and heats up; hydrogen fuses in a shell around it, and the outer layers swell and cool. The star becomes a red giant.
  2. The core becomes hot enough to fuse helium into carbon and oxygen.
  3. The outer layers drift away as a glowing shell of gas, a planetary nebula.
  4. The carbon–oxygen core is left as a white dwarf. It is held up against gravity by electron degeneracy pressure: electrons resist being squeezed into the same states. A white dwarf can only exist if its mass is below about 1.4 solar masses (the Chandrasekhar limit).

High-mass stars (above about 8 solar masses)

  1. The star becomes a red supergiant. Its core is hot enough to fuse heavier and heavier elements in layers, like an onion: carbon, neon, oxygen, silicon, and finally iron.
  2. Iron is at the peak of the binding energy curve, so fusing it releases no energy. The core can no longer support itself and collapses in less than a second.
  3. The outer layers rebound off the collapsed core in a supernova, which can briefly outshine a whole galaxy. Elements heavier than iron are made in the explosion and scattered into space.
  4. The core is left as a neutron star (held up by neutron degeneracy pressure, with the density of a nucleus) if its mass is below about 2–3 solar masses (the Oppenheimer–Volkoff limit). Above that, nothing can stop the collapse, and it becomes a black hole.

On the HR diagram, the Sun will move from the main sequence up and to the right to the red giants, then across and down to the white dwarfs.

The elements in your body were made in stars: the hydrogen in the Big Bang, the carbon and oxygen in red giants, and heavier elements in supernovae (and in collisions between neutron stars).

10. Common mistakes

11. Check your understanding

Why does fusion need such a high temperature?

Nuclei are positively charged and repel. Only very fast nuclei can get close enough (about $10^{-15}$ m) for the strong force to bind them, and high temperature means high kinetic energy.

A star has a parallax angle of 0.050″. How far away is it in light years?

$d = \frac{1}{0.050} = 20$ pc $= 20 \times 3.26 = 65$ ly.

Two stars have the same surface temperature, but star X is 16 times as luminous as star Y. Compare their radii.

$L \propto R^2$ at fixed $T$, so $R_X = \sqrt{16}\,R_Y = 4R_Y$.

Where on the HR diagram are stars that are very hot but very dim, and what are they?

Bottom left: white dwarfs. Being hot but dim means they must be very small.

Why can't a star produce energy by fusing iron?

Iron has the highest binding energy per nucleon (near the peak of the curve), so fusing it into heavier nuclei would absorb energy, not release it.

What decides whether a star ends as a white dwarf, a neutron star or a black hole?

Its mass. Low-mass stars leave white dwarf cores below 1.4 solar masses. High-mass stars explode as supernovae, leaving a neutron star or, if the core is above about 2–3 solar masses, a black hole.

Practise E.5 questions