Origins of the Elements
How radiation became matter after the Big Bang, how Hubble showed the Universe is expanding, how stars turn mass into energy and light elements into heavier ones, and how spectra and the Hertzsprung–Russell diagram tell us what each star is doing.
75 min · Module 8: From the Universe to the Atom
Assumes you have read The Electromagnetic Spectrum, Light: Quantum Model and Energy from the Nucleus.
The full teaching. Start here.
The short version
The Big Bang. About billion years ago the Universe was extremely hot and dense, and almost entirely radiation. As it expanded it cooled. Energetic photons made particle and antiparticle pairs by ; once it was cool enough, the particles survived. Quarks joined into protons and neutrons within a millionth of a second, these fused into helium and a little lithium within about minutes, and at years electrons joined nuclei to form atoms. The radiation released then is the cosmic microwave background.
Hubble. Using Cepheid variable stars for distance and red shift for velocity, Hubble found in 1929 that galaxies recede with speeds proportional to their distance, . The Universe is expanding.
Stars. A main sequence star fuses hydrogen into helium in its core. The helium is per cent lighter than the hydrogen that made it, and that mass becomes the star's energy, . Low mass stars like the Sun use the proton–proton chain; stars above about mostly use the CNO cycle, where carbon is a catalyst. After the main sequence, helium fuses to carbon by the triple alpha process.
Reading stars. A star's spectrum gives its temperature (from the peak, by Wien's law, and from which lines appear), its composition, and whether it is a giant or a dwarf (from line widths). The HR diagram plots luminosity against temperature. Its regions show a star's stage: main sequence, giant, supergiant or white dwarf.
From radiation to matter
The Big Bang theory describes how the Universe expanded and cooled from an extremely hot, dense state. It does not describe an explosion into empty space: space itself was expanding everywhere at once. It also does not tell us what happened at the very first instant, where the physics it is built on stops working.
Energy becomes particles
In the first instants the Universe was filled with photons of enormous energy. By , mass and energy are equivalent, so two photons with enough energy can turn into a particle and its antiparticle: a quark and an antiquark, or an electron and a positron. The reverse also happens: a particle and antiparticle meet and annihilate into two gamma rays.
At first both happen constantly. Particles form and are destroyed almost at once. What changes everything is expansion: as space stretches, so do the photons' wavelengths, and their energy falls. Once typical photons can no longer make a particle pair, or break apart a structure that has formed, whatever particles exist survive.
Very slightly more matter than antimatter formed, about one extra particle per billion pairs. When the pairs annihilated, that small excess was left over, and it is all the matter in the Universe today.
The timeline
Each step needs a cooler Universe than the step before, because each builds something more fragile.
| Time after the Big Bang | Temperature | What happens |
|---|---|---|
| First s | above K | The four forces separate; quarks, leptons and their antiparticles form and annihilate |
| About s | about K | Quarks combine into protons and neutrons |
| Seconds | about K | Most matter and antimatter has annihilated, leaving the small excess of matter |
| to minutes | about K | Protons and neutrons fuse into deuterium, helium-3, helium-4 and a trace of lithium-7 |
| years | about K | Electrons join nuclei to form neutral atoms; light travels freely |
| million years | The first stars form from clumps of hydrogen and helium | |
| million years | The first galaxies |
Three features of this table explain what the Universe is made of.
Why the elements stopped at helium. Nuclei could only survive once photons were too weak to break them up, a few minutes in. But free neutrons decay into protons, with a mean life of about minutes, and the falling temperature and density soon made fusion too slow. In that short window almost every neutron ended up in helium-4. To go further, helium-4 would have to add a proton, giving mass , or another helium-4, giving mass . No stable nucleus has or nucleons, so the chain stalled. The result was about per cent hydrogen and per cent helium by mass, which is what we measure in the oldest stars and gas clouds.
Why the early Universe was opaque. Until years the Universe was a plasma: nuclei and free electrons. Free electrons scatter photons very strongly, so light was bounced around like headlights in fog. When the temperature fell to about K, electrons were captured by nuclei. Neutral atoms scatter light far less, so the Universe became transparent.
Where the cosmic microwave background comes from. The radiation released at that moment was black body radiation at about K, peaking in the near infrared. Since then space has stretched about times, and so have the wavelengths. It now arrives from every direction as an almost perfect black body spectrum at K, peaking in microwaves.
CheckpointAnswer before reading on.
Put these events after the Big Bang in order, earliest first: first stars form, atoms form and the Universe becomes transparent, quarks combine into protons and neutrons, helium nuclei form, quarks and leptons appear.
- first stars form
- helium nuclei form
- quarks and leptons appear
- atoms form and the Universe becomes transparent
- quarks combine into protons and neutrons
Hint 1Each step needs the Universe to be cooler than the one before, because each builds something more fragile.
Hint 2Protons and neutrons must exist before they can fuse into helium.
Hint 3Nuclei can only hold electrons once photons are too weak to knock them off, which took years.
CheckpointAnswer before reading on.
Nucleosynthesis in the first minutes after the Big Bang produced hydrogen, helium and a trace of lithium, but almost nothing heavier. Which statement best explains why?
Hint 1Free neutrons are unstable. How long do they last on their own?
Hint 2Try building up from helium-4 by adding a proton or another helium-4. What mass numbers do you reach?
Hint 3No stable nucleus has or nucleons.
Common mistake
Protons and neutrons formed from quarks a few minutes after the Big Bang.
Why it happens: Summaries often lump every stage of making matter into the same few minutes.
Quarks combined into protons and neutrons within about a millionth of a second. The few minutes window is when those protons and neutrons fused into nuclei. They are two separate events, at very different temperatures.
The evidence for the Big Bang
- Expansion. Galaxies recede at speeds proportional to their distance. Running the expansion backwards, everything was once much closer together and hotter.
- The cosmic microwave background. A hot, dense early Universe predicts radiation filling all space, with a black body spectrum stretched to a few kelvin. It was found in 1965, and satellites show it matches a black body almost exactly.
- The light elements. The predicted ratio of hydrogen to helium by mass, and the small amounts of deuterium and lithium, match what is observed.
CheckpointAnswer before reading on.
A student says the cosmic microwave background is the light from the Big Bang explosion itself, seen from outside. What is the best correction?
Hint 1Could light travel freely through the early plasma?
Hint 2What changed at about years?
Hint 3Why is the radiation at K now if it was released at about K?
CheckpointAnswer before reading on.
Explain how matter formed from radiation in the early Universe, and why this could only happen once the Universe had cooled.
Hint 1Which equation says energy can become mass?
Hint 2Very energetic photons can make a particle and its antiparticle. What happens to such pairs when the Universe is extremely hot?
Hint 3As the Universe expands it cools. How does that change whether the particles made can survive and combine?
Hubble and the expanding Universe
In 1916 Einstein's general relativity described gravity as the curving of space. Einstein himself assumed the Universe was static, but Alexander Friedmann in 1922 and Georges Lemaître in 1927 found that the equations allowed a Universe that expands or contracts. Lemaître went further and predicted that galaxies should recede at speeds proportional to their distances. The theory was ready; the evidence came from Edwin Hubble's measurements.
Hubble needed two numbers for each galaxy: how far away it is, and how fast it is moving.
Distance: Cepheid variables
Henrietta Leavitt studied Cepheid variables, stars whose brightness rises and falls regularly. She found that the longer the period, the more luminous the star. So measuring a Cepheid's period gives its luminosity , the power it actually emits.
Hubble found Cepheids in other galaxies and measured their periods and their apparent brightness , the power per square metre reaching Earth. Light spreads over a sphere, so
and knowing and gives the distance . In 1923 to 1924 this showed that the Andromeda "nebula" was far outside the Milky Way: galaxies were separate systems.
Velocity: red shift
Galaxies' spectra contain absorption lines of elements such as hydrogen and calcium. Vesto Slipher had found that in almost every galaxy these lines are shifted to longer wavelengths, a red shift. For speeds well below the shift gives the speed of recession:
Doppler shift of a spectral linenot on the formulae sheet, learn it
- Symbols
- observed wavelength minus laboratory wavelength; positive is a red shiftm
- wavelength of the line measured in a laboratorym
- speed of the source along the line of sight; positive is moving awaym s⁻¹
- speed of lightm s⁻¹
- Valid when
- The source's speed along the line of sight is much less than c, which is true for stars in our galaxy.
- Not valid when
- Speeds are a large fraction of c, or the shift of distant galaxies, which is mostly the expansion of space rather than motion through it.
- Rearranged
- for v: for \Delta\lambda:
- Where it turns up
- Finding how fast a star moves towards or away from us from the shift of its lines
- Finding a star's rotation speed from how much its lines are widened
- Where marks go missing
- Dividing by the observed wavelength instead of the laboratory one (a small error at low speeds, but the wrong method)
- Calling a shift to longer wavelengths a blue shift
The graph
In 1929 Hubble plotted velocity against distance and found a straight line through the origin. The further away a galaxy is, the faster it recedes, in proportion.
Hubble's lawnot on the formulae sheet, learn it
- Symbols
- recession velocity of a galaxy, found from its red shiftkm s⁻¹
- Hubble constant, about 70 km s⁻¹ Mpc⁻¹km s⁻¹ Mpc⁻¹
- distance to the galaxyMpc
- Valid when
- Galaxies far enough away, beyond about 10 Mpc, that the expansion of space dominates their own motion, and close enough that the red shift is small compared with 1.
- Not valid when
- Nearby galaxies such as Andromeda, which is approaching us because of gravity, or very distant galaxies where the simple Doppler relation for red shift breaks down.
- Rearranged
- for D: for t:
- Where it turns up
- Finding the distance to a galaxy from its red shift
- Finding the Hubble constant from the gradient of a velocity against distance graph
- Estimating the age of the Universe as 1/H₀
- Where marks go missing
- Mixing units, such as a velocity in m s⁻¹ with H₀ in km s⁻¹ Mpc⁻¹
- Forgetting to convert megaparsecs to kilometres before finding 1/H₀ in seconds
- Reading the law as evidence that we are at the centre of the expansion
Hubble's own value of was about km s⁻¹ Mpc⁻¹, because his Cepheid distances turned out to be too small. Modern measurements give about . Try fitting the data yourself.
- Best fit
- Galaxies
- v = 55D
The numbers behind this graph
| Distance (Mpc) | Velocity (km/s) |
|---|---|
| 13 | 913 |
| 28 | 2357 |
| 35 | 2272 |
| 39 | 2734 |
| 41 | 3226 |
| 46 | 3293 |
| 64 | 4390 |
| 102 | 6913 |
| 104 | 7296 |
| 112 | 7547 |
| 124 | 8883 |
| 126 | 8901 |
| 146 | 9433 |
| 146 | 10149 |
| 154 | 10109 |
| 168 | 12802 |
| 182 | 12788 |
| 196 | 14091 |
| 223 | 16184 |
| 228 | 15641 |
| 234 | 15461 |
| 238 | 17343 |
| 250 | 18816 |
| 253 | 17827 |
| 260 | 18167 |
| 330 | 24953 |
| 376 | 26126 |
| 381 | 27916 |
- Trial H₀
- 55 km s⁻¹ Mpc⁻¹ = 1.78 × 10⁻¹⁸ s⁻¹
- Typical miss, root mean square
- 3212 km s⁻¹ (best possible 581 at H₀ = 71)
- Age estimate, 1 / H₀
- 17.8 billion years
Every slider is a normal range input, so the arrow keys move it one step and Home and End jump to the extremes.
A proportional law seen in every direction is exactly what uniform expansion produces. Imagine dots on a balloon being blown up: every distance doubles in the same time, so a dot twice as far away gains twice as much distance and recedes twice as fast. Every dot sees the same thing, so the law does not put us at the centre.
The galaxies are not flying through space. Space between them is stretching, and the light travelling through it is stretched too, which is the red shift. At very large distances Hubble's law gives recession speeds above . This does not break special relativity, which limits motion through space, not the expansion of space itself.
Worked example4 marks
Distance to a galaxy from its red shift
The hydrogen-alpha line, nm in the laboratory, is observed at nm in a galaxy's spectrum. Taking km s⁻¹ Mpc⁻¹, find the galaxy's recession velocity and distance, and estimate the age of the Universe as . Mpc km.
Fractional shift
Compare the shift with the rest wavelength.
Velocity
The fractional shift is v/c. Keep c in km s⁻¹ to match H₀.
Distance
Rearrange Hubble's law.
Age
If the rate never changed, every galaxy took the same time 1/H₀ to reach its distance.
Answer
The galaxy recedes at about km s⁻¹ and is about Mpc away. The Universe is roughly billion years old.
Is that answer sensible?
The accepted age is billion years, close to . With Hubble's own km s⁻¹ Mpc⁻¹ the same calculation gives only billion years, younger than the Earth, which was an early sign his distances were wrong.
CheckpointAnswer before reading on.
The hydrogen-alpha line has a wavelength of nm in the laboratory. In the spectrum of a distant galaxy it is observed at nm.
Taking km s⁻¹ Mpc⁻¹, calculate the distance to the galaxy in megaparsecs.
Give it to 2 significant figures.
Hint 1The fractional shift in wavelength equals for speeds much less than .
Hint 2. Multiply by in km s⁻¹.
Hint 3Then .
CheckpointAnswer before reading on.
A survey finds km s⁻¹ Mpc⁻¹. Estimate the age of the Universe as , in billions of years.
Mpc km and year s.
Give it to 3 significant figures.
Hint 1 has units of km s⁻¹ per Mpc. Convert the Mpc to km so the distance units cancel.
Hint 2 s⁻¹.
Hint 3Invert to get seconds, then divide by the seconds in a year.
CheckpointAnswer before reading on.
Almost every distant galaxy is red shifted, and the more distant it is the faster it recedes. A student concludes that the Milky Way must be at the centre of the Universe. What is the best response?
Hint 1Picture dots drawn on a balloon being blown up. What does any one dot see?
Hint 2If all distances are scaled up by the same factor, which galaxies move away fastest from any chosen galaxy?
Hint 3Is there anything special about our viewpoint in Hubble's law?
Describetypically 2 to 4 marks
- Demands
- State the features and characteristics. What happens, not why.
- Shape
- Specific detail: quantities, directions, order of events.
- Loses marks
- Explaining the cause. It is not wrong, but it earns nothing here and eats the clock.
Exam question
Harder · about 8 min
5 marks
Describe the evidence that led Hubble to conclude that the Universe is expanding. Include how he measured the two quantities he plotted.
Hint 1Hubble plotted velocity against distance. Where did each come from?
Hint 2Distances came from a type of variable star studied by Henrietta Leavitt.
Hint 3Velocities came from the spectra of galaxies. What did the lines show, and what did the graph look like?
Where students lose marks on this one
Saying Hubble measured distances from red shift.
Why it happens: Today red shift is used to find distance, using Hubble's law.
That relies on the law Hubble discovered. He needed an independent distance, which the Cepheids gave him.
Written for this site.
Stars turn mass into energy
A star forms when a cloud of gas collapses under its own gravity. As it falls inwards the gas loses gravitational potential energy and heats up, but no fusion happens yet. When the core reaches about million kelvin, protons move fast enough to overcome their electrostatic repulsion and fuse. The star is born, and it settles into a balance: the outward pressure of hot gas and radiation holds it up against gravity.
The overall reaction in a main sequence star is
The helium nucleus is more tightly bound than four separate protons, so it has less mass. That missing mass is the energy the star radiates.
Mass–energy equivalenceon the NESA formulae sheet
- Symbols
- energy equivalent to the mass, or released when that mass is convertedJ
- mass, or the mass lost by a systemkg
- speed of light in a vacuumm s⁻¹
- Valid when
- Any process. The energy a system releases equals its loss of rest mass times c², whether the process is nuclear, chemical or an annihilation.
- Not valid when
- m is taken as the total mass of the reactants rather than the mass that disappears, or masses in atomic mass units are used without converting to kilograms.
- Rearranged
- for m:
- Where it turns up
- The rate at which the Sun loses mass
- The photon energy from electron–positron annihilation
- The mass change in burning a fuel
- Where marks go missing
- Using the whole mass of a fuel rather than the tiny mass lost
- Forgetting to square c
- Believing mass is converted only in nuclear reactions
Worked example3 marks
Energy from turning hydrogen into helium
The atomic masses of hydrogen-1 and helium-4 are u and u. Calculate the energy released each time four hydrogen nuclei become one helium-4 nucleus, and the fraction of the mass converted.
Mass lost
Using atomic masses on both sides lets the electrons, and the positrons they annihilate with, take care of themselves.
Energy
1 u is equivalent to 931.5 MeV.
Fraction
Compare the mass lost with the mass that went in.
Answer
MeV is released per helium nucleus, from per cent of the mass.
Is that answer sensible?
Helium-4's binding energy is MeV. Two protons have to become neutrons on the way, which costs about MeV, leaving MeV. A little of it, about MeV, leaves with the neutrinos, which pass straight out of the star.
Worked example3 marks
How fast the Sun loses mass
The Sun radiates W and has a mass of kg. How much mass does it lose each second, and what fraction of its mass will it lose over its billion year main sequence life?
Mass per second
In one second it radiates 3.83 × 10²⁶ J, and that energy was once mass.
Over its life
Multiply by the number of seconds in 10 billion years.
Fraction
Compare with the Sun's total mass.
Answer
The Sun loses about kg every second, yet only about per cent of its mass over its whole main sequence life.
Is that answer sensible?
Only per cent of the fused hydrogen is converted, and only the core's hydrogen, about a tenth of the star, is ever fused: per cent. The two routes agree.
CheckpointAnswer before reading on.
The overall result of the proton–proton chain is that four hydrogen-1 atoms become one helium-4 atom. The atomic masses are u and u.
Calculate the energy released in MeV each time the chain runs.
Give it to 3 significant figures.
Hint 1Find the total mass before and after.
Hint 2Atomic masses include electrons. Using atoms throughout lets the positron annihilations be counted automatically.
Hint 3Multiply the mass lost in u by MeV.
CheckpointAnswer before reading on.
The Sun radiates energy at W. Calculate the mass it loses every second as a result.
Give it to 3 significant figures.
Hint 1The power is the energy radiated each second.
Hint 2That energy was once mass, by .
Hint 3 with J.
Common mistake
The Sun loses 4 million tonnes of hydrogen a second.
Why it happens: The mass lost and the fuel used are easy to confuse.
The kg per second is the mass that disappears. Since only per cent of the fuel's mass is converted, the Sun turns about kg of hydrogen into helium every second. The rest stays in the helium.
Analysetypically 4 to 7 marks
- Demands
- Break the situation into its parts and show how they relate and affect each other.
- Shape
- Identify the components, then the relationships, then what follows from them.
- Loses marks
- Summarising the stimulus rather than taking it apart.
Exam question
Harder · about 8 min
5 marks
Analyse how Einstein's equivalence of mass and energy accounts for the energy output of a main sequence star such as the Sun. Support your answer with a calculation. The atomic masses of hydrogen-1 and helium-4 are u and u.
Hint 1Which reaction powers a main sequence star, and what goes in and comes out?
Hint 2Compare the masses before and after, and convert the difference to energy.
Hint 3Link the energy per reaction to the star's luminosity, and say what this means for its mass over time.
Where students lose marks on this one
Writing that the hydrogen nuclei are destroyed and turned into energy.
Why it happens: Mass disappears, so it seems particles must.
All four nucleons survive in the helium-4 nucleus (two protons become neutrons). Each is slightly lighter because it is more tightly bound.
Written for this site.
What starlight tells us
Starlight is the only thing we receive from a star, so everything we know about stars comes from its spectrum. The physics of spectra is in the electromagnetic spectrum lesson; here is what matters for stars.
Three kinds of spectrum
- Continuous (black body). A hot, dense object, such as a star's photosphere, emits at all wavelengths. The curve's shape depends only on temperature, and its peak is at .
- Emission. A hot, thin gas emits only at particular wavelengths, when electrons drop between energy levels. Each element has its own set of bright lines.
- Absorption. Light with a continuous spectrum passing through cooler gas loses the wavelengths that gas's atoms can absorb. The atoms re-emit in random directions, so those wavelengths appear as dark lines, at exactly the positions of that element's emission lines.
A star shows the third kind: a continuous spectrum from the photosphere, crossed by dark lines from the cooler gas above it.
Wien's displacement lawon the NESA formulae sheet
- Symbols
- wavelength at which a black body's emission is most intensem
- Wien's constant, 2.898 × 10⁻³ m Km K
- absolute temperature of the bodyK
- Valid when
- A body that emits close to a black body spectrum, such as a star's photosphere, a hot filament or a cavity with a small hole. T must be in kelvin.
- Not valid when
- The source emits a line spectrum, such as a gas discharge tube or a laser, or the temperature is left in degrees Celsius.
- Rearranged
- for T:
- Where it turns up
- Estimating the surface temperature of a star from its peak wavelength
- Explaining why hotter bodies glow blue and cooler ones red
- Finding the peak wavelength of radiation from the human body or the Earth
- Where marks go missing
- Using the temperature in degrees Celsius
- Leaving the wavelength in metres when nanometres were asked for, or the reverse
- Reading the peak as the colour the body appears, when the eye sees a mixture of wavelengths
CheckpointAnswer before reading on.
Compare how emission and absorption line spectra are produced with how a continuous black body spectrum is produced.
Hint 1Line spectra come from electrons in atoms. What determines which wavelengths appear?
Hint 2An absorption spectrum needs two things: a source and something in front of it.
Hint 3A black body spectrum comes from a hot dense object. Which feature of it depends on temperature?
CheckpointAnswer before reading on.
The Sun's spectrum shows dark lines at the wavelengths of sodium. A student says this proves the Sun contains no sodium, since sodium's wavelengths are missing. What is the best correction?
Hint 1Which atoms could absorb exactly those wavelengths?
Hint 2Where is the absorbing gas relative to the hot photosphere?
Hint 3What happens to the energy the atoms absorbed?
Classifying stars by their spectra
Stars are classified in the Harvard system by spectral class, in order of surface temperature: O, B, A, F, G, K, M, with each class split from (hottest) to . The Sun is G2.
| Class | Temperature (K) | Colour | Strongest lines |
|---|---|---|---|
| O | above | blue | Ionised helium; few lines overall |
| B | to | blue-white | Neutral helium, some hydrogen |
| A | to | white | Hydrogen, at its strongest |
| F | to | yellow-white | Hydrogen weaker; ionised metals |
| G | to | yellow | Ionised calcium, many metal lines |
| K | to | orange | Neutral metals |
| M | below | red | Molecules such as titanium oxide |
Three features of a stellar spectrum are used.
The peak wavelength and colour give temperature. A star's photosphere is close to a black body, so Wien's law gives its temperature from where its spectrum peaks, and the colour we see tells us which side of the visible band that peak is on.
Which lines appear also gives temperature. Stars have broadly similar compositions; their spectra differ mainly because of temperature. In the hottest stars, collisions strip electrons from most atoms. An ionised atom has few bound electrons left to jump between levels, so O stars show few lines. In cool stars atoms are not ionised, and molecules can even survive, adding bands of lines. Hydrogen lines are strongest at about K, in A stars, where many hydrogen atoms have their electron in the second level ready to absorb visible light.
The width of the lines gives size. Your notes put this well. In a small, compact star the atmosphere is close to the core, gravity at the surface is strong, and the gas is dense. Frequent collisions disturb the atoms' energy levels, so each line is spread over a range of wavelengths. A giant star's surface is far from its centre, so gravity there is weak and the atmosphere is thin, giving narrow lines. Two stars of the same class can be told apart this way, and a luminosity class is added: V for main sequence stars, III for giants, I for supergiants. The Sun is G2V.
The spectrum also gives the star's composition, by matching lines to laboratory spectra, and its motion towards or away from us, from the Doppler shift of the lines.
CheckpointAnswer before reading on.
The continuous spectrum of a star peaks at nm. Calculate its surface temperature in kelvin.
Give it to 3 significant figures.
Hint 1Use Wien's law, .
Hint 2Convert the wavelength to metres.
Hint 3
CheckpointAnswer before reading on.
The spectra of O-type stars, with surface temperatures above K, show far fewer absorption lines than the spectra of G-type stars like the Sun. Why?
Hint 1Absorption lines need electrons bound in atoms to jump between levels.
Hint 2What happens to atoms when they collide at very high temperatures?
Hint 3An atom with no electrons cannot make lines at all.
CheckpointAnswer before reading on.
Two stars have the same spectral class, K2. Star X shows narrow, sharp absorption lines; star Y shows broad lines. What does this suggest?
Hint 1Line width is affected by the pressure of the gas in the star's atmosphere.
Hint 2In dense gas, collisions disturb the energy levels and broaden the lines.
Hint 3Where is surface gravity lower: on a giant or on a dwarf?
The Hertzsprung–Russell diagram
Around 1910 Ejnar Hertzsprung and Henry Norris Russell independently plotted stars' luminosities against their temperatures. The stars did not scatter everywhere. They fell into a few distinct groups, and those groups turn out to be stages in the lives of stars.
The conventions are unusual, so learn them:
- Temperature, or spectral class, is on the horizontal axis, and it increases to the left. Hot blue O stars are on the left, cool red M stars on the right. The scale is logarithmic.
- Luminosity is on the vertical axis, relative to the Sun, on a logarithmic scale. Sometimes absolute magnitude is used instead; smaller magnitudes are brighter, and magnitudes is a factor of in brightness.
Luminosity is the total power a star emits. It is not how bright the star looks, which depends on distance as well.
Size from position
A star is close to a black body, so each square metre of its surface radiates . Multiplying by the surface area gives the luminosity:
Luminosity of a starnot on the formulae sheet, learn it
- Symbols
- luminosity, the total power the star radiatesW
- radius of the starm
- Stefan–Boltzmann constant, 5.67 × 10⁻⁸ W m⁻² K⁻⁴W m⁻² K⁻⁴
- surface temperatureK
- Valid when
- A star whose surface radiates close to a black body spectrum, so its surface temperature can be read from its peak wavelength.
- Not valid when
- Sources that are not black bodies, such as a gas cloud giving a line spectrum.
- Rearranged
- for R:
- Where it turns up
- Explaining why a cool star high on the HR diagram must be a giant
- Explaining why a hot star low on the HR diagram must be a white dwarf
- Comparing two stars' radii from their positions on the HR diagram
- Where marks go missing
- Squaring the temperature instead of raising it to the fourth power
- Forgetting to take the square root when solving for the radius
- Assuming a hotter star must always be more luminous
This is what makes the diagram so informative. Two stars at the same temperature radiate the same power per square metre, so if one is a thousand times more luminous it must have a thousand times the surface area. Cool but luminous means enormous; hot but faint means tiny. Lines of constant radius run diagonally across the diagram.
Explore the diagram. Drag the star around and watch its radius, class and stage change.
- Main sequence
- Giants
- Supergiants
- White dwarfs
- Region
- Main sequence
- Evolutionary stage
- Fusing hydrogen into helium in its core, the longest stage of its life.
- Spectral class and colour
- G, yellow. Peak wavelength b / T = 500 nm
- Luminosity
- 1 L☉ = 3.8 × 10²⁶ W
- Radius, from L ∝ R²T⁴
- 1.0 R☉
- Mass and main sequence lifetime
- about 1.0 M☉, about 9.7 billion years
Every slider is a normal range input, so the arrow keys move it one step and Home and End jump to the extremes.
The regions
- Main sequence. A diagonal band from hot, luminous and massive (upper left) to cool, faint and light (lower right). About per cent of stars are here, fusing hydrogen into helium in their cores. A star's position on the main sequence is set by its mass: more massive stars are hotter and much more luminous.
- Red giants. Upper right: cool (about to K) but to times as luminous as the Sun, so to times its radius. They have finished core hydrogen fusion. Their narrow spectral lines confirm their low surface gravity.
- Supergiants. Across the top, from blue to red, up to a million times the Sun's luminosity. Massive stars after the main sequence, fusing heavier elements.
- White dwarfs. Lower left: hot (about to K) but very faint, so about the size of the Earth. They are the exposed cores of dead low mass stars, with no fusion, slowly cooling.
Worked example5 marks
Reading a star from the HR diagram
A star appears on an HR diagram at a surface temperature of K and a luminosity of . A main sequence star at K has a luminosity of about .
Determine the star's spectral class, colour, peak wavelength, radius relative to the Sun ( K), and evolutionary stage.
Class and colour
Read off the temperature scale: K stars are 3700 to 5200 K.
K is class K, near K7. The star is orange.
Peak wavelength
Wien's law, with T in kelvin.
This is at the red edge of the visible band, so the star emits more red than blue light, consistent with its colour.
Radius
Divide L = 4πR²σT⁴ by the same equation for the Sun so the constants cancel.
Compare with the main sequence
A star at the same temperature radiates the same power per square metre, so the luminosity ratio is the area ratio.
It is times as luminous as a main sequence star of the same temperature, so it has times the surface area and times the radius. It lies far above the main sequence.
Stage
Cool, luminous and large is the red giant region.
The star is a red giant: it has used up the hydrogen in its core, and hydrogen now fuses in a shell around a contracting helium core, which has pushed its outer layers out.
Answer
Class K, orange, nm, about : a red giant that has left the main sequence.
Is that answer sensible?
Place it on the interactive diagram above at K and : it falls in the giant region, between the and solar radius lines. Its spectrum would show narrow lines, luminosity class III.
CheckpointAnswer before reading on.
A star has a surface temperature of K and a luminosity of . Where is it on the HR diagram, and what stage of its life is it in?
Hint 1Hot stars are on the left of the HR diagram. Is high or low?
Hint 2A main sequence star at K is about a thousand times more luminous than the Sun.
Hint 3Hot but faint means tiny.
CheckpointAnswer before reading on.
A star on the HR diagram has a surface temperature of K and a luminosity of . The Sun's surface temperature is K.
Using , calculate the star's radius in solar radii.
Give it to 2 significant figures.
Hint 1Write the equation for the star and for the Sun, and divide one by the other so cancels.
Hint 2
Hint 3
CheckpointAnswer before reading on.
A student says Sirius B, a white dwarf at K, must be more luminous than Betelgeuse, a red supergiant at K, because it is much hotter. What is wrong?
Hint 1Luminosity is power from each square metre times the surface area.
Hint 2. Which star has the larger radius, and by how much?
Hint 3Betelgeuse is about times the radius of the Sun; Sirius B is about .
CheckpointAnswer before reading on.
An M-type star has a surface temperature of K, so its spectrum peaks at nm, in the infrared. Why does it look red rather than invisible?
Hint 1A black body emits over a wide range of wavelengths, not just at the peak.
Hint 2Which part of the visible band is closest to the peak?
Hint 3The eye sees the mixture of visible wavelengths reaching it.
Common mistake
Hotter stars are always more luminous.
Why it happens: It is true along the main sequence, which is where most stars are.
Luminosity depends on surface area as well as temperature. A white dwarf at K is a hundred thousand times fainter than a red supergiant at K, because the supergiant is about times wider. Only compare temperature and luminosity directly for stars on the main sequence.
Through a marker’s eyes
3 marks
A star has a surface temperature of 25 000 K and a luminosity of 0.03 L☉. Use the HR diagram to explain what type of star it is. (3 marks)
The attempt
1 out of 3
It is a white dwarf because it is in the bottom left of the HR diagram. It is hot and faint.
What the marker sees
The classification is right and the position is right, which earns one mark. But the answer never explains why hot and faint means a white dwarf. The marks are for linking temperature and luminosity to size, and size to the stage of the star's life.
The same answer, fixed
3 out of 3
At K the star is on the left of the HR diagram, and at it is far below the main sequence, where a star of that temperature would be thousands of times more luminous than the Sun.
Each square metre of a hot surface radiates strongly (), so for the total luminosity to be so low the star must have a tiny surface area: , about the size of the Earth.
A hot, tiny star in the lower left is a white dwarf: the exposed core of a low mass star that has finished fusion and is slowly cooling.
How stars evolve
A star spends most of its life at one spot on the main sequence, set by its mass. It does not travel along the main sequence. It leaves when the hydrogen in its core runs out, and what happens next depends on its mass.
How long a star lasts
More massive stars have more fuel, but their stronger gravity squeezes their cores to higher temperatures and densities, and fusion rates rise very steeply with temperature. Luminosity rises much faster than mass, roughly as , so heavier stars burn out sooner. A useful estimate is fuel over rate of use:
A star lives only about million years; a red dwarf, over a trillion.
CheckpointAnswer before reading on.
A main sequence star has ten times as much hydrogen as the Sun. A student concludes it will live ten times longer. What is the best correction?
Hint 1What does a larger mass do to the pressure and temperature in the core?
Hint 2How sensitive is the fusion rate to temperature, especially for the CNO cycle?
Hint 3A star is several thousand times more luminous than the Sun.
Stars like the Sun (below about 8 solar masses)
- Main sequence. Hydrogen fuses to helium in the core, mostly by the proton–proton chain, for about billion years in the Sun's case.
- Red giant. The core hydrogen runs out. With no fusion to hold it up, the helium core contracts and heats up. Hydrogen now fuses in a shell around it, faster than before, and the extra energy pushes the outer layers out. The surface expands and cools. The star moves up and to the right on the HR diagram.
- Helium burning. When the core reaches about K, helium fuses to carbon and oxygen by the triple alpha process. In stars below about , like the Sun, this starts suddenly in a helium flash, because their cores are so compressed that they cannot expand to regulate the reaction. The star settles, a little smaller and hotter, and moves back down and to the left.
- Second giant phase. When the core helium runs out, a carbon–oxygen core contracts, helium and hydrogen burn in shells around it, and the star swells again.
- Planetary nebula and white dwarf. The star is not massive enough to fuse carbon. Pulses gently push its outer layers into space as a glowing planetary nebula. The exposed core is a white dwarf, hot but Earth-sized, in the lower left. It does not collapse further because electrons cannot be squeezed any closer (electron degeneracy pressure). With no fusion, it cools and fades, moving down and to the right.
Massive stars (above about 8 solar masses)
- Main sequence, mostly by the CNO cycle, for a few million to tens of millions of years.
- Supergiant. After hydrogen, the core burns helium, then carbon, neon, oxygen and silicon, each stage hotter and shorter than the last. The star builds up layers like an onion, with heavier elements closer to the centre, and swings back and forth across the top of the HR diagram between blue and red supergiant.
- Iron core. Fusing iron absorbs energy rather than releasing it, so once the core is iron nothing holds it up. It collapses in under a second; electrons and protons are forced together into neutrons and neutrinos.
- Supernova. The collapse stops suddenly when the neutrons are packed as tightly as in a nucleus, the infalling material rebounds, and the star's outer layers are blasted into space. In the flood of neutrons, elements heavier than iron are built.
- Remnant. The core is left as a neutron star, about km across. If it is above about , not even neutrons can resist gravity and it becomes a black hole.
Ageing a cluster
The stars in a cluster formed from the same cloud at about the same time, but with different masses. So a cluster's HR diagram is a snapshot: the heaviest stars have already left the main sequence, while lighter ones are still on it. The point where the main sequence turns off towards the giants is the turn-off, and the stars at the turn-off are finishing their hydrogen right now. The cluster's age is their main sequence lifetime.
As a cluster ages, the turn-off moves down the main sequence. Slide the age and watch.
- Zero age main sequence
- Still on the main sequence
- Evolved off it
- Age
- 1.0 billion years
- Turn-off
- 8834 K, class A, 2.1 × 10¹ L☉
- Heaviest star left on the main sequence
- about 2.1 M☉
Every slider is a normal range input, so the arrow keys move it one step and Home and End jump to the extremes.
Worked example3 marks
Age of a cluster from its turn-off
A cluster's main sequence turns off at spectral class A0, where stars have about times the Sun's mass and times its luminosity. Estimate the cluster's age.
Identify what the turn-off tells you
Stars at the turn-off are just running out of core hydrogen now.
The cluster's age equals the main sequence lifetime of a star.
Estimate the lifetime
Fuel is proportional to mass and rate of use to luminosity.
Answer
The cluster is about million years old.
Is that answer sensible?
Set the cluster model above to about years: the turn-off sits near K, class A. An older cluster, turning off at the Sun's position, would be about billion years old.
CheckpointAnswer before reading on.
In a star cluster, the most massive stars still on the main sequence have times the Sun's mass and times its luminosity. The Sun's main sequence lifetime is years.
Taking lifetime to be proportional to fuel divided by the rate it is used, , estimate the age of the cluster in billions of years.
Give it to 3 significant figures.
Hint 1The stars at the turn-off are just finishing their hydrogen, so the cluster's age equals their lifetime.
Hint 2
Hint 3
Describetypically 2 to 4 marks
- Demands
- State the features and characteristics. What happens, not why.
- Shape
- Specific detail: quantities, directions, order of events.
- Loses marks
- Explaining the cause. It is not wrong, but it earns nothing here and eats the clock.
Exam question
Harder · about 10 min
6 marks
Describe the life of a star like the Sun after it forms, referring to its position on the Hertzsprung–Russell diagram and the nuclear reactions in its core at each stage.
Hint 1Start on the main sequence. What reaction runs, and what keeps the star stable?
Hint 2What happens when the core hydrogen runs out? Where does the star move on the diagram, and why?
Hint 3Which reaction starts in the core next, and what is finally left?
Where students lose marks on this one
Saying the Sun will become a supernova or a black hole.
Why it happens: Mixing the fates of low mass and high mass stars.
Only stars that start above about explode. The Sun will end as a white dwarf.
Saying stars move along the main sequence as they age.
Why it happens: The main sequence looks like a path.
Each star stays at nearly one point on the main sequence, set by its mass. It leaves the main sequence when its core hydrogen runs out.
Written for this site.
Nucleosynthesis in stars
Stars make every element heavier than lithium. Main sequence stars fuse hydrogen into helium by two routes; after the main sequence, helium and heavier nuclei fuse too.
The proton–proton chain
The main energy source in the Sun and all lighter stars. It builds helium from protons directly, in three steps.
The first two steps happen twice for each helium-4 made, so the net result is four protons becoming one helium-4 nucleus, two positrons, two neutrinos and gamma rays, releasing MeV. The first step is very slow, because a proton has to turn into a neutron (a beta plus decay, controlled by the weak force) at the moment two protons touch. A given proton in the Sun waits billions of years for it. That slowness is why the Sun lasts so long.
The CNO cycle
In stars above about , with cores hotter than about million K, a second route dominates. Carbon-12 captures protons one by one, turning into nitrogen and oxygen isotopes, until the last step splits off a helium-4 nucleus and returns the carbon-12. Explore each step.
The CNO cycle
1Carbon-12 captures a proton
A proton has to get close enough to a nucleus of charge for the strong force to take over. The repulsion is six times what it is between two protons, which is why the CNO cycle needs a hotter core than the p-p chain.
2Beta plus decay
Nitrogen-13 has too many protons for its size, so one proton turns into a neutron and a positron and a neutrino are emitted. The same happens later to oxygen-15. These two decays are how two of the four protons become the neutrons that helium-4 needs.
3Two more proton captures
The nitrogen-14 capture is the slowest step of the whole cycle, so nitrogen-14 builds up. This is where most of the nitrogen in the Universe, including in the air you breathe, was made.
4The last step closes the cycle
The fourth proton makes a nucleus that splits into carbon-12 and helium-4. The carbon is back where it started, ready to go round again.
5Carbon is a catalyst
Carbon, nitrogen and oxygen are not used up: they are catalysts. They must already be in the star, so the CNO cycle only runs in stars that formed from gas enriched by earlier generations of stars. The very first stars, made of pure hydrogen and helium, could not use it until they had made their own carbon.
6The net result
Four protons in, one helium-4 nucleus, two positrons and two neutrinos out: exactly the same as the p-p chain, with the same MeV released. The difference is the rate. The CNO cycle speeds up with temperature far more steeply (about against for the p-p chain), so in hot, massive stars it burns hydrogen very fast.
In the Sun, with a core temperature of about million K, the p-p chain makes about per cent of the energy and the CNO cycle about per cent. In a star the CNO cycle already dominates.
CheckpointAnswer before reading on.
A student says that a star running the CNO cycle will eventually run out of carbon, since carbon-12 is used up in the first step. What is the best correction?
Hint 1Follow the cycle all the way round. What is produced in the last step?
Hint 2What goes in overall, and what comes out overall?
Hint 3What do we call something that takes part in a reaction but is returned at the end?
Comparetypically 3 to 5 marks
- Demands
- Show similarities and differences, addressing both things in every point.
- Shape
- Points of comparison, each covering both items on the same criterion.
- Loses marks
- Writing one paragraph on each item and leaving the reader to do the comparing.
Exam question
Harder · about 6 min
4 marks
Compare the proton–proton chain and the CNO cycle as sources of energy in main sequence stars.
Hint 1What goes in and comes out overall in each? Is the energy released the same?
Hint 2What does the CNO cycle need that the p-p chain does not?
Hint 3Which dominates in the Sun, and which in stars above about ? Why?
Where students lose marks on this one
Saying the CNO cycle releases more energy per helium nucleus.
Why it happens: It powers the more luminous stars.
Energy per helium nucleus is the same, because the start and end are the same. The CNO cycle runs faster, so more reactions happen each second.
Written for this site.
After the main sequence: the triple alpha process
Once a star's core hydrogen is gone, its core contracts and heats. At about K, helium nuclei move fast enough to overcome their charges and fuse. Two helium-4 nuclei make beryllium-8, which falls apart in about s, but in a dense hot core a third helium-4 sometimes arrives first:
This is how stars get past the mass and mass gap that stopped nucleosynthesis after the Big Bang. Some carbon-12 captures another helium-4 to make oxygen-16, so the core becomes carbon and oxygen.
Stars above about go on, each stage needing a hotter core: carbon burning makes neon and magnesium, then oxygen and silicon burning, and finally iron and nickel. Iron-56 has the highest binding energy per nucleon, so fusing it releases no energy, and the sequence ends.
CheckpointAnswer before reading on.
Which reaction is the main source of energy in the core of a red giant after the helium flash?
Hint 1The core hydrogen has run out by now.
Hint 2The core has contracted and reached about K.
Hint 3Three of which nucleus make carbon-12?
Beyond iron
Elements heavier than iron are made by adding neutrons to existing nuclei, which then beta decay into heavier elements. Slow neutron capture happens in giant stars near the end of their lives. Rapid neutron capture, which makes gold, platinum and uranium, needs an enormous flood of neutrons, found in supernovae and in collisions of neutron stars. Since a neutron star merger was observed making heavy elements in 2017, mergers are thought to be the main source.
Why the expansion rate is still argued about
Two ways of measuring disagree. Cepheids and supernovae in nearby galaxies give about km s⁻¹ Mpc⁻¹; working from the cosmic microwave background with the standard model gives about . The gap is larger than either method's stated uncertainty, and nobody yet knows whether it points to a measurement problem or to new physics. This is known as the Hubble tension.
In 1998 surveys of distant Type Ia supernovae found they were fainter than expected, meaning the expansion has been speeding up for the last few billion years. The cause is called dark energy. Some recent analyses have questioned whether the acceleration is still going on, and the question is being tested by current surveys.
Magnitudes
Astronomers often use magnitudes instead of luminosity. The scale is backwards and logarithmic: a difference of magnitudes is a factor of exactly in brightness, so magnitude is a factor of . Apparent magnitude is how bright a star looks from Earth. Absolute magnitude is how bright it would look from a standard distance of parsecs, so it measures luminosity. The Sun's absolute magnitude is ; the faintest stars visible to the naked eye have apparent magnitude about .
Why the main sequence is a mass sequence
A star's mass sets the pressure its core must provide, and so its core temperature. Because fusion rates depend so steeply on temperature, a small increase in mass gives a large increase in power. Luminosity rises about as , so a star of is several thousand times as luminous as the Sun. Below about the core never gets hot enough for hydrogen fusion at all, giving a brown dwarf; above about the radiation pressure is so great that the star blows off its outer layers.