What's New
Site Support

9.2 Space
9.3 Motors & Generators
9.4 Ideas-Implementation
9.7 Astrophysics
9.8 Quanta to Quarks

9.7 Astrophysics Page 2
9.7 Astrophysics Page 3
9.7 Astrophysics Page 4


NOTE: This page is a continuation of the notes and worksheets for topic 9.7 Astrophysics.  Four separate pages were used for this topic because of the large volume of material in the topic.  This will keep download time within acceptable limits.







NOTE: Numbers appearing in parentheses at the end of sentences or paragraphs refer to the references provided in the Bibliography at the end of these notes.

The only direct way to determine the mass of a star is by studying its gravitational interaction with other objects.  Binary stars, two stars in orbit around a common centre of mass, provide an ideal opportunity to do just that.  At least half of all the “stars” in the sky are actually multiple star systems.  Analysis of the orbital parameters of such systems provides vital information about a variety of stellar characteristics, including mass.  (1) 

The methods used to analyse the orbital data depend on the geometry of the system, its distance from the observer and the relative masses and luminosities of each component.  As a consequence, binary star systems are classified according to the means by which they are detected.  (1) 

There are six main types of binary systems: visual, spectroscopic, eclipsing, astrometric, spectrum and optical (1).  The first four of these are required for study by the current syllabus.  Let us now describe these four different types of binary star systems.





When astronomers can actually see the two stars orbiting each other, the binary is called a visual binary.  The brighter star in the pair is called the primary star and is denoted by the letter A after its name, while the other star is called the secondary and is denoted by the letter B.  For example, the alpha star in the constellation of the Southern Cross is actually a visual binary consisting of Alpha Crucis A and Alpha Crucis B, both of which are easily discernable through a small telescope. 

Many years of careful observation are often necessary to ensure that the two stars are actually in orbit around each other.  If the stars do form a binary system, each star will follow an elliptical orbit around the centre of mass of the system.  The star with the larger mass will stay closer to the centre of mass of the system and will therefore have the smaller orbit.  This makes sense – just think of two children on a seesaw.  To balance, the child with the greater mass must sit closer to the fulcrum.




Some binary systems are detected due to a periodic shift in spectral lines.  These are called spectroscopic binaries and are detected using the Doppler Effect.  If two stars in a binary system have some component of their orbital motion along the observer’s line of sight, then as the stars move around their orbits, they periodically approach and recede from the observer.  Thus, the spectral lines of the two stars are alternately blueshifted and then redshifted.  (1 & 3) 

Through a telescope such a star system may appear to be one single star.  However, an examination of the spectra coming from “the star” will reveal two sets of spectral lines, one set from each star, each of which will display Doppler shifts.  The two overlapping spectra shift relative to each other due to the Doppler Effect.  When the spectrum of each star in the binary system is visible, the binaries are called double-line spectroscopic binaries.  If, however, one star in the pair is much more luminous than the other, the spectrum of the less luminous companion may be overwhelmed and only a single set of spectral lines will be seen.  In this case, the spectral lines shift back and forth due to the Doppler Effect and therefore still reveal the binary system.  This type of binary is called a single-line spectroscopic binary.  (1 & 3) 

Consider the diagram below, which I have adapted from Reference 1 p.204.


The diagram above shows the periodic shift in spectral features of a double-line spectroscopic binary.  For simplicity, only one spectral line from each star is shown.  Note that l0 is the non-Doppler shifted wavelength of the lines and wavelength increases to the right of that position.  The relative wavelengths of the spectra of stars 1 and 2 are shown at four different phases during orbit: (a) star 1 is moving towards Earth while star 2 is moving away; (b) both stars have velocities perpendicular to the line of sight; (c) star 1 is receding from Earth and star 2 is approaching Earth; and (d) again both stars have velocities perpendicular to the line of sight of the observer on Earth. 

Note that the spectroscopic detection of binaries is most likely if the period of the motion is short and the orbital speeds of the stars are high.  So, most spectroscopic binaries are close binary systems.  (3)






Some binaries have their orbital planes oriented approximately along the line of sight of the observer.  In this case, one star may periodically pass in front of the other, blocking light from the eclipsed star.  Such a system is called an eclipsing binary and is recognized by regular variations in the amount of light received at the telescope (1).  Light curves (plots of apparent visual magnitude versus time or orbital phase) can be recorded for eclipsing binaries using an appropriate photometer.  Such curves can reveal enormous amounts of information about the binary system.  For example, light curves can confirm the presence of two stars.  The ratio of surface temperatures can be determined from how much their combined light is diminished during eclipse.  The duration of the eclipse provides information on the relative radii of the stars and their orbits.  (3) 

Further to this, if the eclipsing binary also happens to be a double-line spectroscopic binary, an astronomer can calculate the actual mass and radius of each star in the system from the light curves and the velocity curves.  (3) 

Examine the light curve shown below.  It is typical of a binary system consisting of small hot star and a large cool star with the orbital plane edge-on to our line of sight.  Note that the smaller star is travelling from left to right across the front of the larger star.  This diagram was adapted from Reference 12 p.323.


Note: Try out this java applet, which allows you to model the light curves of eclipsing binaries using a computer simulation (Syllabus point 9.7.5, column 3, dot point 1). 


This simulation is also available at:






Sometimes one member of a binary is significantly brighter than the other and may be the only one of the pair visible through a telescope.  In such a case the existence of the unseen member may be deduced by observing the oscillatory motion of the visible star (1).  This reveals itself as a detectable “wobble” in the star’s proper motion (5).  Since Newton’s First Law requires that a constant velocity be maintained by a mass unless an external unbalanced force is acting upon it, such oscillatory motion requires that another mass be present.  Such a binary star system is called an astrometric binary.  (1)





As mentioned previously binary stars are very important in determining the masses of stars.  Binary stars orbit each other in accordance with Kepler’s Third Law – the Law of Periods.  For any binary system we can write:


where m1 = mass of star 1, m2 = mass of star 2, r = average distance of separation of the two stars, T = orbital period of the binary system and G = the Universal Gravitational Constant = 6.673 x 10-11 SI. 

This equation is known as Newton’s form of Kepler’s Third Law.  Clearly, for any binary system if we can determine the orbital period and the average distance of separation of the stars in the system, we can determine the total mass of the system (m1 + m2).

For visual binaries we can usually determine the orbital period, although in some cases observations over more than one lifetime may be needed.  If the distance from Earth to the binary can be calculated from parallax measurements or using spectroscopic parallax, then the angular separation between the stars in the binary can be translated into the physical distance between the stars.  Astronomers must also take into account the angle of tilt of the binary’s orbit to our line of sight.  Substituting into Kepler’s Third Law equation then allows us to calculate the total mass of the binary system.  (1 & 3) 

By plotting the individual orbits of the two stars in the system, the centre of mass of the system can be determined.  Then, comparing the relative sizes of the two orbits around the centre of mass enables the ratio of the masses m1 / m2 of the two stars to be determined.  Having both the sum of the masses and the ratio of the masses then allows the calculation of the individual masses of the two stars to be achieved.  (1 & 3)  See Ref. 1 pp.205-208 for a detailed analysis. 

For spectroscopic binaries the wavelength shift of each star’s spectral lines is measured and then this data is used to calculate the radial velocity of each star from the relevant Doppler shift formula.  Remember that radial velocity is velocity along our line of sight.  This radial velocity data is then used to plot a radial velocity curve (radial velocity versus time) for each star.  The ratio of the masses of the stars in the binary is obtained from this curve, along with the orbital period of the binary. 

Kepler’s Third Law and Newtonian mechanics are then used to relate the sum of the masses of the stars to the orbital period of the binary, the orbital speeds of the stars and the angle of tilt of the binary’s orbit to our line of sight.  If all of these details are known, then again the individual masses of the two stars can be determined.  Often, however, the angle of tilt is not known accurately and therefore stellar mass determinations from spectroscopic binaries are often uncertain.  (1 & 3)  See Ref. 1 pp.208-211 for a detailed analysis. 

There is one case in which the angle of tilt is known and that is when the spectroscopic binary is also an eclipsing binary.  The orbital plane of the binary must then lie edge-on to our line of sight.  Clearly, with this knowledge and with values for the orbital period of the binary and the orbital speeds of the component stars, the sum of the masses of the stars can be determined.  Then, since both the sum of the masses and the ratio of the masses are known, the individual masses of the two stars can be determined.  (1 & 3)  See Ref. 1 pp.211-214 for a detailed analysis. 

Note that in the case of astrometric binaries, it is usually possible to measure the orbital period of the binary and the average distance of separation of the two stars.  Thus, using Newton’s form of Kepler’s Third Law we can get an estimate of the total mass of the binary system.  It is usually not possible to determine the masses of the individual stars.


Example Questions

1.      An astrometric binary is observed to have a period of 44.5 years and an orbit with an average distance of separation between the component stars of 100 AU.  Determine the sum of the masses of the stars in this binary.  (Data: 1AU = 1.5 x 1011 m)  (Answer: 1.012 x 1033 kg)

2.      An eclipsing binary has a period of 44.5 years and an average distance of separation between the component stars of 3.9 AU.  Determine the combined mass of the binary system.  (Answer: 6.0 x 1028 kg)






A variable star is one that varies in brightness (5).  As at January 2004, the total number of designated variable stars was 38622 according to the “General Catalogue of Variable Stars”, 4th Edition, by P N Kholopov et al, Moscow (1985) and its published updates (also available on the Internet).  There are two broad categories of variable star: (a) Extrinsic Variables, which vary in brightness for some reason external to the star; and (b) Intrinsic Variables, which vary in brightness due to changes in the star itself.  Certain stars may vary in brightness due to both of these reasons. (5)



As mentioned above, Extrinsic Variables vary in brightness for some reason external to the star.  Such reasons include rotation, orbital motion or obscuration.  Examples of common types of extrinsic variables include: 

(a)    Rotating variables, in which either the star’s ellipsoidal shape or the presence of large cooler or hotter areas on the star’s surface change the star’s brightness as it rotates.  BY Draconis stars are examples of the latter.

(b)   Eclipsing Binaries, which as discussed previously, consist of orbiting stars that periodically eclipse each other.  Algol (Beta Persei) is an example of an eclipsing binary system about 100 light years from Earth.




Intrinsic Variables vary in brightness due to actual changes in the luminosity of the star itself not due to external processes such as rotation or eclipses.  The majority of variable stars fall into this broad category.  Intrinsic variables are further sub-divided into non-periodic or periodic variables.




These are intrinsic variables that show irregular variations in brightness.  Examples of such variables include: 

(a)    Novae – A nova is a star that undergoes a sudden, unpredictable increase in brightness typically of 11 to 12 orders of magnitude (5).  The sudden increase in brightness is accompanied by an explosive loss of a relatively small amount (about 10-5) of the mass of the star.  The outer atmosphere is blown away to form an expanding shell of gas around the star that can often still be seen many years after the explosion.  Nova Cygni, which flared up in 1975 by about 19 magnitudes is an example. (10)

(b)   Supernovae – A supernova is a violently exploding star, which may become over a billion times brighter than the Sun, and for many weeks may outshine the entire galaxy in which it lies (5).  The explosion completely destroys the star and throws most of its matter into space at high speeds, leaving behind either a tiny, very dense, collapsed remnant called a neutron star or in some cases a black hole.  SN1987A is a supernova observed in the Large Magellanic Cloud in 1987.  The Crab Nebula is the remnant of a supernova explosion in the constellation Taurus in 1054. (10)

(c)    R Coronae Borealis Stars – These stars are sometimes referred to as “reverse novae” since they decrease in brightness by as much as 10 magnitudes before returning to normal.  They are supergiants with carbon-rich atmospheres.  The sudden minima are due to the accumulation of clouds of carbon dust, which are then blown away and allow the star to return to its normal brightness.  The decreased brightness phase may last from months to years. (5 & 10)

(d)   Flare Stars – These stars are red dwarfs that exhibit intense outbursts of energy (flares) from small areas of their surface.  Examples include UV Ceti stars and certain BY Draconis stars. (5)

(e)    T-Tauri Stars – These are very young protostars.  They are less than 10 million years old, with masses similar to or less than that of the Sun and with diameters several times that of the Sun.  They are still contracting.  They exhibit irregular variability ranging from ultraviolet flares on a time scale of minutes to optical variations on time scales of days to years.  The prototype, T Tauri itself, lies within Hind’s Variable Nebula and varies irregularly between 8th and 13th magnitudes. (5 & 3)




These are intrinsic variables that show regular variations in brightness.  These variables are also known as pulsating variables.  A pulsating variable undergoes a periodic cycle of contraction and expansion as the star oscillates between a state where the inward acting gravitational force is dominant and another where the outward acting radiation pressure is dominant.  The periodic contraction and expansion produces the observed periodic variation in brightness.  As the star contracts, the increase in luminosity caused by the temperature increase exceeds the decrease caused by the contraction in radius, so the star becomes brighter.  As the star expands, the luminosity and therefore the brightness decrease. (9 & 10). 

The types of periodic variables can be categorized by the period and amplitude of their light curves.  The period is simply the amount of time between one occurrence of maximum (or minimum) brightness and the next.  The amplitude is the difference in magnitude between maximum and minimum brightness.  Examples of periodic variables include: 

(a)    Cepheid Variables – These are the best-known examples of pulsating variables and are named after Delta Cephei, the first star of this type to be discovered.  Cepheids are very luminous yellow giant or supergiant stars.  Their brightness usually varies by about one magnitude with a period of between 1 and 135 days, while their radius typically varies by 10 to 20 percent.  We will say a little more about these very important stars later.  They are extremely useful as a means of distance measurement.  (5 & 10)

(b)   Mira Stars (also called Red Variables) – These are the most common pulsating stars.  They are long period pulsating red giants and supergiants.  They take their name from the star Mira (Omicron Ceti) which lies about 250 ly from Earth.  Mira stars have periods of about 80 to 1000 days and have an amplitude of 2.0 to 10 magnitudes. (5 & 10)

(c)    RR Lyrae – These are the second most common type of variable.  They are old yellow giants.  Most have periods between 0.2 and 1.2 days and amplitudes of 0.2 to 2.0 magnitudes.  One of the two types of RR Lyrae star (the RRAB type) all have approximately the same absolute magnitude of +0.5, making them valuable distance indicators. (5 & 10)

(d)   RV Tauri – These are highly luminous, yellow supergiants.  The light curves have overall amplitudes of 3 to 4 magnitudes.  The period lies in the range from 30 to 150 days. (5)

NOTE: The above classification of variable stars is the one required by the current HSC Astrophysics Syllabus.  This classification scheme is based on the form of the light curve for the star, its amplitude and periodicity or lack of it.  This classification scheme has been superseded.  I do not know why the Syllabus Committee decided to go back to the old system.  The new classification scheme adopted by the “General Catalogue of Variable Stars”, 4th Edition, is based on the physical mechanisms that underlie the different forms of variation or the physical structure of the stars themselves.  This new classification scheme allocates variable stars to one of seven classes: Eruptive, Pulsating, Rotating, Cataclysmic (explosive nova-like), Close Binary Eclipsing Systems, Optically Variable X-Ray Sources and finally Unique Variables (for mistakes and variables no-one has figured out yet).





When trying to remember the classification scheme for variable stars, a simple flow chart like the following can be of great assistance.  This particular flowchart appears in Ref.12 on p.326.  A similar one is shown in Ref.9 on p.46.






The following figure shows where certain types of variable stars are found on the HR Diagram.  Cepheid variables and RR Lyrae variables are located in the Cepheid instability strip (between the dotted lines on the diagram), which occupies a region between the main sequence and the red giant branch.  A star passing through this region along its evolutionary track becomes unstable and pulsates. (3)


The diagram above is an amalgamation of similar diagrams from Ref. 3 p.530, Ref. 12 p.327 and Ref. 9 p.47.






In 1912, American astronomer Henrietta Leavitt reported her important discovery of the period-luminosity relationship for Cepheid variables.  Leavitt studied numerous Cepheids in the Small Magellanic Cloud, a small galaxy near our own Milky Way.  Leavitt found that the periods of these Cepheids were directly related to their average luminosities.  The longer the Cepheid’s period, the greater is its luminosity. (3 & 9) 

This period-luminosity relationship is very important in astronomy because it can be used to determine distances to objects in the universe (3).  The process is as explained below.

Cepheid variables vary in a regular and characteristic way, increasing in brightness rather more rapidly than they decrease (10).  Cepheid variables therefore have very distinctive light curves, which are plots of brightness versus time.  A typical light curve appears as below.


From this light curve we can determine both the period of the Cepheid variable under study and its average apparent magnitude.  In this case the period is 8 days – the time between one occurrence of maximum brightness and the next.  The average apparent magnitude is (4.0 + 3.5)/2 which equals 3.75.  Using the period we can then determine the average absolute magnitude of the Cepheid variable from the appropriate period-luminosity curve (see the next graph).  Once the average absolute magnitude is known we can use the distance modulus formula to calculate the distance to the required Cepheid variable. 

Note that initially Leavitt and others thought that all Cepheids could be described by the same period-luminosity relationship.  This, however, was not the case.  We now know that there are two different types of Cepheid variable, each with its own period-luminosity relationship.  Type I or Classical Cepheids are the brighter, more massive (5-15 solar masses), metal-rich, younger, second generation stars, found exclusively in the disc population of galaxies, where they are often members of open clusters (5, 10 & 12).  Type II or W Virginis Cepheids are the dimmer, less massive (0.4 to 0.6 solar masses), metal-poor, older, red, first generation stars (5, 10 & 12).  Astronomers examine a star’s metal content from its spectrum in order to classify it as either a Type I or Type II Cepheid (3).  The period-luminosity relationship for both types of Cepheid variable is shown in the following graph.


This graph is an adaptation of the graph shown on p.49 of Ref.9.  Note that the relationship for Type I Cepheids has only been drawn up to a period of 50 days.  In reality, Type I Cepheids can have periods up to 135 days.  Remember too that the straight lines in the graph are simply showing general trends that are seen in the data.  The actual period-luminosity graphs recorded by astronomers consist of bands of stars plotted on and around these lines.  See Ref.1 Fig.14.4 on p.545 for example.

Now continuing with our example of how to determine the distance to a Cepheid, let us assume that the Cepheid in question has been identified as a Type I Cepheid.  So, using the known period of 8 days, we can read off the period-luminosity relationship that the absolute magnitude of our Cepheid is –2.5.

The final step is to use the distance modulus formula to calculate the distance to the Cepheid.


So, on re-arrangement we have:


which yields d = 177.8 parsecs, when m = 3.75 and M = -2.5 are substituted into the above equation.  Thus, the distance to our Cepheid variable is 177.8pc.

As mentioned previously, this method of using the period-luminosity relationship for determining distances to Cepheids is very important in astronomy.  It provides a relatively simple and accurate method of distance calculation.  An astronomer identifies a Cepheid as either Type I or Type II from its spectrum.  He/she then records the light curve for the Cepheid and from this determines the period and average apparent magnitude of the Cepheid.  Using the period, the astronomer then determines the absolute magnitude of the Cepheid from the period-luminosity relationship and finally calculates the distance to the Cepheid using the distance modulus formula. (3) 

Cepheid variables are extremely luminous objects.  They can be seen even at distances of millions of parsecs.  By applying the period-luminosity relationship in this way to Cepheids in other galaxies, astronomers have been able to calculate the distances to those galaxies with great accuracy.  Such measurements play an important part in determining the overall size and structure of the universe. (3)



Where in our galaxy does star formation occur?  The spiral arms of our Milky Way galaxy are laced with giant molecular clouds (GMC), immense clouds of interstellar gas (mainly hydrogen) and dust, so cold that their constituent atoms can form into molecules (3).  Particularly dense regions within these clouds form what are called “Dark Nebulae” and within these, stars are formed.  Star formation commences in these dark nebulae when a GMC is compressed, which can happen as the cloud passes through one of the spiral arms of the galaxy or as a result of a nearby supernova explosion (3).

A nebula is an interstellar cloud of gas and dust.  A dark nebula is a nebula with a relatively dense concentration of microscopic dust grains, which scatter and absorb light very efficiently.  The Horsehead Nebula in the constellation of Orion is a very good example of a dark nebula.  Dark nebulae partially or completely block our view of any stars that lie behind them.  A typical dark nebula has a low temperature of 10 K to 100 K, which is low enough for hydrogen to form molecules, and contains from 104 to 109 particles (atoms, molecules & dust grains) per cubic centimetre.  (3)

These characteristics make dark nebulae the only parts of the interstellar medium suitable for star formation.  The relatively high density enhances gravitational attraction within the medium and the low temperature ensures a low (outwards pushing) pressure within the medium.  (3)

Within a dark nebula, the densest regions of gas and dust may begin to contract under their own gravity.  This contraction may also be triggered by the gravity from a passing star or the shock wave from a nearby supernova explosion.  The density at the centre of a contracting region increases more quickly than that at the outer edges.  Thus, the gravitational attraction at the centre increases more quickly than that at the edges, resulting in the formation of a central core of material surrounded by a more slowly contracting envelope.  (3 & 9)

As the core contracts, the gravitational potential energy of its constituent particles is transformed into kinetic energy, heating the core.  This heat causes the gas in the core to glow.  Convection currents carry the heat outwards, creating an outward-directed pressure that opposes the gravitational contraction of the core.  As the core temperature increases, the outward pressure increases until eventually the gravitational collapse of the core is almost balanced by the outward pressure.  At this stage the core is called a protostar.  Protostars with masses between 0.08 solar masses and 100 solar masses will eventually evolve into main sequence stars. (1 & 3)

Check out the following four star-forming regions:

http://hubblesite.org/gallery/album/nebula/pr2006001a/large_web/ - The Orion Nebula (M42), 1500 light years away in the constellation of Orion, is the nearest star-forming region to the Sun (13).  It is certainly one of the best examples of such an area in space (13).

http://antwrp.gsfc.nasa.gov/apod/ap000111.html - The Rosette Nebula is an impressive star-forming area at a distance of 5500 light years in the constellation Monoceros (The Unicorn).  (13)

http://antwrp.gsfc.nasa.gov/apod/ap990511.html - Barnard 68, a dark Bok globule at a distance of 410 light years in the constellation Ophiuchus (The Serpent Holder), is on the point of collapsing into a new star.  (13)

http://www.daviddarling.info/encyclopedia/T/Trifid_Nebula.html - The Trifid Nebula (M20), a star forming region a few thousand light years away in the constellation Sagittarius.  (13)

It is worth noting that even though young protostars are quite luminous, they cannot be detected using visible light.  This is because the dust in the protostar’s immediate surroundings, called its cocoon nebula, absorbs much of the visible light emitted by the protostar.  However, as the dust in the cocoon nebula re-radiates this absorbed heat at infrared wavelengths, the presence of the protostar can be detected using infrared telescopes.  Many protostars have been detected in this way.  (3)

During the birth process stars both gain and lose mass.  Mass loss occurs in a couple of different ways.  Protostars less than 3 solar masses go through a T-Tauri phase in which they eject up to a solar mass of material over about 107 years and exhibit irregular variation in luminosity on a time scale of a few days.  Protostars more massive than 3 solar masses do not vary in luminosity like T-Tauri stars but do lose mass due to the huge radiation pressure at their surfaces that blows gas into space.  Many protostars also lose mass by ejecting gas along two oppositely directed jets in a process called bipolar outflow.  The huge stellar winds produced by these mass ejection processes often blow away the remnants of the surrounding cloud and allow the protostar to be seen in visible light.  (3)

Protostars gain mass at the same time as they are losing it.  As the envelope of material around the protostar contracts, it spins faster and faster and flattens into a disk with the protostar at the centre.  Particles orbiting the protostar within the disk collide with each other, causing them to lose energy and spiral inward onto the protostar, adding to its mass.  This process is called accretion.  (3)

While these processes of mass loss and gain are continuing, so too is the very slow contraction of the protostar itself.  There is still insufficient energy flowing outwards from the protostar to completely balance the inward pull of gravity.  As the radius of the protostar decreases, its luminosity decreases and its internal temperature increases.  Eventually, the internal temperature of the protostar reaches around 107 K, sufficient for thermonuclear reactions to begin converting hydrogen to helium.  These reactions eventually produce sufficient heat and internal pressure to stop the star’s contraction.  The outward radiation pressure and gas pressure has balanced the inward force of gravity.  Hydrostatic equilibrium has been achieved.  Also, the rate at which energy is produced in the core has balanced the rate at which energy is transported to the surface of the star and radiated away into space.  Thermal equilibrium has been established.  The protostar is now said to be a zero-age main sequence star and the processes of stellar formation are complete. (3)

Note that the dark nebulae in which star formation occurs typically contain tens or hundreds of solar masses of gas and dust, enough to form many stars.  Thus, young stars tend to form in groups or clusters.  Star clusters typically include stars with a range of different masses, all of which began to form out of the parent nebula at roughly the same time.  (3)





Curves that depict the life histories of stars on the H-R Diagram are called evolutionary tracks (1).  The diagram below shows the evolutionary tracks of five protostars of different masses.  Each track shows us how the protostar’s appearance changes in terms of its luminosity and temperature because of changes in its interior.  Where on the main sequence a given track ends depends on the mass of the protostar.  Protostars eject a lot of mass into space as they form, so the mass shown for each evolutionary track is the mass of the final main sequence star.  Note that for a star like the Sun, the birth process takes about 50 million years.  (This diagram was produced by Prof. Dale Gary of the New Jersey Institute of Technology.  The original link was: http://physics.njit.edu/~dgary/202/Lecture18.html.  This link no longer works.)


Note that the plot of the main sequence using only stars of zero-age is called the zero-age main sequence (ZAMS) plot (12).  It forms the complete diagonal main sequence shape shown on most generic H-R Diagrams (12).  Stars on ZAMS have just ended their protostar stage (3). 

The theory of how protostars evolve helps explain why the main sequence has both an upper and lower mass limit.  A protostar of less than 0.08 solar masses can never develop sufficient pressure and temperature to start hydrogen fusion in its core.  Such a protostar becomes a hydrogen-rich brown dwarf – a failed star.  A protostar more massive than 100 solar masses rapidly becomes very luminous, resulting in tremendous internal pressures.  These pressures overwhelm gravity, expel the outer layers into space and disrupt the protostar.  (3)






By definition, a main sequence star is one that produces energy by the fusion of hydrogen nuclei (protons) to helium nuclei in its core.   This fusion reaction produces energy by the conversion of some of the hydrogen nuclei mass into energy according to Einstein’s equation, E = mc2.  Note that astronomers are notorious for referring to fusion reactions as “burning”.  So they speak of “hydrogen burning” instead of hydrogen fusion and “helium burning” instead of helium fusion, and so on.

Two different fusion mechanisms are responsible for the helium production and consequent release of energy in main sequence stars.  Both mechanisms can occur simultaneously in a main sequence star.  However, for stars whose core temperatures are below 16 million K the proton-proton chain reaction is the main mechanism, while for stars whose core temperatures are above this, the carbon-nitrogen-oxygen (or CNO) cycle predominates (3).  Let us now have a brief look at these two mechanisms.


The Proton-Proton Chain Reaction

This reaction predominates in stars like our Sun.  Originally proposed by the American physicist Charles Critchfield, this reaction has three branches (3).  Since the primary branch PP I, accounts for the production of 85% of the Sun’s energy, we will consider only this branch of the reaction in detail.  PP I consists of three steps:




For those not familiar with nuclear equations, refer to the Key above.  In step 1, two protons combine to form a deuterium nucleus (an isotope of hydrogen), a positron and a neutrino.  In step 2, another proton combines with the deuterium nucleus to form a nucleus of light helium and a gamma ray photon, which carries energy away from the reaction.  In step 3, two light helium nuclei combine to produce a nucleus of ordinary helium and two protons.  (10)

Thus, the overall reaction is to convert four protons into a nucleus of helium with the release of some energy.

In the PP II and PP III branches of the reaction the light helium produced in step 2 above suffers different fates.  Details of these branches can be found in Refs. 1 & 3.




The Carbon-Nitrogen-Oxygen (CNO) Cycle

This reaction mechanism predominates in stars whose core temperatures are above 16 million K.  Hans Bethe and Carl von Weizsacker discovered it independently (3).  In the CNO cycle the carbon-12 nucleus acts as a catalyst and the following six-step reaction takes place (10).


Overall, in this reaction four protons are converted into a helium nucleus, two positrons, two neutrinos and high-energy gamma ray photons (10).  Again, for those not familiar with nuclear reactions the following Key is provided.


For the CNO cycle to proceed, there must be carbon-12 nuclei present.  Obviously, as the carbon-12 is returned at the end of the cycle, it is not actually used in by the reaction.

Throughout the lifetime of the main sequence star, the helium produced by the proton-proton chain reaction and CNO cycle accumulates in the centre of the star, since it is denser than the hydrogen.  Also as the main sequence star ages, changes occur in its luminosity, surface temperature and radius (3).  Hydrogen burning decreases the total number of atomic nuclei in the star’s core (four hydrogen nuclei are used up to make each single helium nucleus).  The resulting decrease in internal pressure causes the core to contract slightly under the weight of the star’s outer layers.  In turn, this contraction increases the core’s density and temperature, which effectively raises the pressure in the core to a level higher than it was previously.

The increased core pressure pushes outwards on the star’s outer layers, causing the star’s radius to increase slightly.  Also, the increased density and temperature in the core cause hydrogen nuclei in the core to collide more frequently, causing the rate of hydrogen burning to increase.  Hence the star’s luminosity increases.  Since the star’s surface temperature depends on the star’s luminosity and radius, it changes as well.  Thus, as the star ages, its core shrinks and its outer layers expand and shine more brightly.  As an example, over the last 4.6 x 109 years, our Sun has become 40% more luminous, grown in radius by 6% and increased its surface temperature by 300K.  (3)

As a main sequence star ages and evolves, the increase in energy outflow from its core also heats the material immediately surrounding the core.  As a result, hydrogen burning can begin in this surrounding material.  This is called shell hydrogen burning since it is happening in the shell surrounding the core.  By tapping this fresh supply of hydrogen, a star manages to last a few extra million years on the main sequence.  (3)

A star’s lifetime on the main sequence depends critically on its mass.  The more massive the star, the shorter its main sequence lifetime.  This is because the more massive the main sequence, the more luminous it is.  In order to emit energy so rapidly, these massive stars deplete their hydrogen stocks very much more quickly than less massive stars.  High mass O and B stars completely exhaust their hydrogen supplies in only a few million years, whereas low mass M stars take billions of years to use all their hydrogen.  A star of around one solar mass will spend roughly 1010 years on the main sequence.  So, our Sun, which has been on the main sequence for about 4.6 billion years, should have about another 5 billion years left to enjoy the main sequence status.  (3)


Go to the previous page of the Astrophysics Topic


Go to the next page of the Astrophysics Topic





1.      Carroll, B.W., & Ostlie, D.A. (1996).  "An Introduction To Modern Astrophysics", New York, Addison-Wesley Publishing Company Inc.

2.      Gondhalekar P. (2001). "The Grip of Gravity", Cambridge, Cambridge University Press

3.      Kaufmann, W.J. III, & Freedman, R.A. (1999).  "Universe", (5th Edition), New York, W.H. Freeman & Company

4.      "Sidereus Nuncius or the Sidereal Messenger". Translated with introduction, conclusion, and notes by Albert Van Helden. Chicago: University of Chicago Press, 1989

5.      Ridpath, I. (Ed.) (1997).  "Oxford Dictionary of Astronomy", Oxford, Oxford University Press

6.      Hollow, R.  "Why Build Big Telescopes?", paper presented at Science Teachers Workshop 2002

7.      http://www.eso.org/projects/aot/introduction.html and http://www.ls.eso.org/lasilla/sciops/ntt/telescope/esontt.html

8.      Eisberg, R. & Resnick, R. (1974).  "Quantum Physics of Atoms, Molecules, Solids, Nuclei and Particles", Canada, Wiley

9.      Playoust, D.F., & Shanny, G.R. (1991).  "An Introduction to Stellar Astronomy", Queensland, The Jacaranda Press

10. Bhathal R., (1993).  "Astronomy for the Higher School Certificate", Kenthurst, Kangaroo Press Pty Ltd

11.  Dawes, G., Northfield, P. & Wallace, K. (2003).  "Astronomy 2004 Australia – A Practical Guide to the Night Sky", Australia, Quasar Publishing

12.  Andriessen, M., Pentland, P., Gaut, R. & McKay, B. (2001). "Physics 2 HSC Course", Australia, Wiley

13. Schilling, G. (2004). "Evolving Cosmos", Cambridge, Cambridge University Press




Last updated:

© Robert Emery 2002 - view the Terms of Use of this site.