Supernovae - "Standard Candles" ?

At a time when telescopes were able to reach only into the relatively near parts of the Universe, distances to stars and galaxies were calculated either trigonometrically using parallax or with a technique pioneered by Henrietta Leavitt using Cepheid variables and a wealth of data was gathered regarding observable objects. One observer, Edwin Hubble, using the Mount Wilson telescope in the 1920’s and taking measurements of the spectra of galaxies, discovered that the redshift, that is the apparent displacement, due to the Doppler effect, of the wavelength of spectral lines emitted by these galaxies, was proportional to their distance. From this he deduced that the further away a galaxy is, the faster it is travelling away from us.

As telescopes, both optical and radio, improved, it became possible to observe objects with greater and greater redshifts, therefore travelling at faster velocities and, by corollary, further away from us. The space telescope bearing Hubble’s name has enabled more and more distant objects to be observed with speeds approaching 90% of the speed of light. The distances to these galaxies have been calculated or, more correctly, estimated using the "Hubble" relationship and, when expressed in light years give a direct indication of the age and size of the known Universe.

Of course, the distance vs. redshift relationship depends much upon that established earlier for the nearer objects. Even then, due to the inherent measurement uncertainties, these yielded significant variability in the "Hubble" constant. The traditional methods of parallax and Cepheid variables cannot be relied upon for measuring such vast distances, now approaching 12 to 15 billion light years. The angles are simply too small and Cepheid variables too faint and, therefore, another basis for distance measurement needed to be established.

One such method involves the use of phenomena known as supernovae. Literally, "bright new stars" these have been observable with the naked eye and, although extremely rare in our own galaxy (the last one was 300 years ago), are thought to occur, on average, at a rate of one per second throughout the observed Universe. Bright they certainly are, with luminosity exceeding that of the entire remainder of their home galaxies for the few days of their duration. The optical output, however, is thought to represent only about 0.01% of the total energy released, thus making them the most violent events in the Universe. The brightness is not questioned but the newness is probably incorrect. There are a number of causes but most involve the death of a star rather than its birth.

In fact, supernovae can now be graded into several types according to their cause and it is one of these that has proven to be important in the desire to measure distances to far galaxies. Known as type Ia, these supernovae are recognised by their spectra which are distinctive in lacking any lines for hydrogen but their important attribute, which makes distance measuring possible, is the apparent consistency of their luminosity. The presence of an object of known luminosity in a distant galaxy renders that distance readily calculable by the inverse square law of propagation of light. By measuring the intensity of the light received we can directly infer how far it has travelled and therefore the time it has taken to reach us.

Of course, the assertion that a particular type of supernova always produces a fixed amount of luminosity is key to this method of measurement and it owes its origins to the inspired work of Subrahmanyan Chandrasekhar who calculated an important property of certain types of stars known as white dwarfs. The "average" sized star, when its nuclear fuel is spent, is overcome by its own gravity and shrinks down to a size, possibly similar to that of the Earth. The matter of which the star is formed, increases in density to the order of magnitude of a million times that of water but Chandrasekhar calculated that there is a limit to the mass of a star that could form a white dwarf and that this is approximately 1.4 times the mass of our sun. Beyond this limit, the collapsing star will form a neutron star or even a black hole, from which no light or matter will escape.

Some white dwarfs, however, exist as one partner of a binary - two stars in relatively close proximity and in orbit about each other. The high density of the white dwarf and the associated gravitational pull results in the accretion of material from its neighbour and, although starting out below the Chandrasekhar limit of 1.4 solar masses, the point in time eventually comes when this limit is reached. At this point a runaway reaction is triggered, resulting in the explosion that we see as a type Ia supernova. As the mass of the star is known, so is the amount of energy released and hence the luminosity. The peculiar process of the explosion also results in the unique absence of hydrogen which provides the identifier. We therefore have a type of "standard candle" with which we can make measurements of distances to the galaxies that contain them.

Studies, in the late 1990s, of large numbers of galaxies, already establised as being distant, have resulted in the observation of type Ia supernovae and thence the calculation of distances. So far, using this method, it has been estimated that the most distant objects in the observed Universe are around 12 billion years old. This, however, is in conflict with estimates for the age of the Universe, of 15 to 18 billion years, drawn from other methods and has given rise to fresh consideration of ideas originally proposed by Albert Einstein in 1915.

Einstein’s Universe, as described by his theories of gravity and general relativity, was static. There was no provision for the expansion that we now believe, and Einstein accepted, to be present. This apparent anomoly lead Einstein to introduce the "cosmological constant" which inferred the presence of forces which repel between all bodies in the Universe. The new data from the supernovae lent credance to this part of Einstein’s theory and much of the puzzle, which had eluded cosmologists for decades, seemed to be falling into place. Not only was the size and age of the Universe now known, but it was also predicted that the expansion is increasing and that, with all parts of the Universe accelerating away from each other, there would be no violent end, or "big crunch". Instead the Universe would gradually thin out, with galaxies and clusters of galaxies becoming more and more isolated, into a kind of dismal, uninteresting perpetuity.

Just as the apparent properties of type Ia supernovae were allowing cosmologists to make parts of the jigsaw fit together, new data regarding these "standard candles" has upset the whole picture. Within the last few years, more detailed studies of supernovae have shown that there is a significant difference in the time taken for luminosities to "peak" when comparing nearby type Ia’s with those in more distant galaxies. Respectively, these average 20 and 17.5 days and this fact has given rise to theories regarding possible differences in the processes which form the supernovae. Maybe the process required to cause a type Ia supernova was different when the Universe was young, as demonstrated by the distant variants, to what it is in the more recent events.

It is believed that the fusion process in the type Ia supernova produces radioactive nickel from a combination of oxygen and carbon and that this begins when the white dwarf draws gas from its neighbour (possibly a red giant) until reaching the 1.4 solar mass limit, when it "ignites". The luminosity is attributed to the radioactive nickel as it decays. However, it is possible that there is considerable variability in the "burning front" of this process as it penetrates through to the star’s surface and that this, in turn, affects the luminosity. The theory holds that, if the "burning front" moves slowly, the luminosity can be reduced by a factor of 10 with corresponding increases in a more rapid system.

Maybe the more distant supernovae have different rates of fusion and do not bear the same standard brightness as previously thought. Alternatively, it may be that some explode prematurely before reaching the 1.4 solar mass limit. Another possible reason to doubt the reliability of data from type Ia studies is the inadequacies of the methodology. Studies of nearer supernovae have largely been done with photographic plates whereas, for those more distant, electronic detectors have been used. The tendency for photographic emulsion to become overexposed due to the intensity of light near the centre of a galaxy, may have lead to some unwitting selection bias in the studied supernovae and, if there is a chance that those on the edge of a galaxy behave differently to those at its centre, more reason for doubt could be present.

One reasonable conclusion is that the type Ia supernova is not the reliable "standard candle" that we thought and that theories of an accelerating universe may be premature. Fortunately there is other emerging evidence regarding the age, size and possible fate of the Universe and cosmologists look, among other places, towards the study of the background microwave radiation and the quest for knowledge of the "dark matter" for new clues. Meanwhile the study of type Ia supernovae continues as an important topic in its own right, regardless of any notion of cosmological "standard candles".