Here at the Young Astronomers we have looked in some detail at stars, the various types that exist as well as their spectra. This short series of posts will deal with what processes occur to power the stars, allows them to shine and how the original materials that made up the first stars got here in the first place.
This post will deal with the composition of stars, metallicity, stellar populations and primordial nucleosynthesis.
First lets look at the material a star is made from, for the sake of ease I will use our own sun – Sol – as an example of a typical star.
All stars are composed primarily of hydrogen and helium with smaller traces of all the other elements. As a star ages the proportion of hydrogen falls (slightly) as it is converted by nuclear reactions into the other elements.
Credit: Peter Clark
Credit: Peter Clark
All stars have a broadly similar composition though the exact balance of components varies from star to star. The ratio of the heavier elements to a stars hydrogen and helium content is a measurement termed metallicity.
Metallicity – Z
The next few lines can be enough to bring a chemist to their knees, so be warned. In astronomy the vast array of elements provided by nature or artificially synthesised in the various particle labs around the world are divided into just two groups, not the many used by chemists: – Metals and non-metals.
Making matters worse there are only two astronomical non-metals with all others (including many of the chemically non-metals) being classed as metals. Why is there such an obtuse system? Well let’s explore the issue.
Hydrogen and helium are the lightest two elements in the periodic table, and are the only two that were formed in any great quantities in the first era of nucleosynthesis (simply put element building) following the formation of the universe in the Big Bang – Primordial Nucleosynthesis (more on this later).
So astronomers make the divide between metals and non-metals not on a chemical basis but on one of initial origin. Non-metals were produced in Primordial Nucleosynthesis and metals were not. Though Helium was and is being produced by stars it is still classed as non-metal as a large quantity was originally produced within this era.
Lithium and beryllium were also produced in small quantities in Primordial Nucleosynthesis thought they aren’t generally considered non-metals though could perhaps be included depending on your definition.
Metallicity is a comparative measure of the metal to non-metal content of any particular star or nebula. It is calculated by comparing the intensity of various spectral lines to derive a ratio. Metallicity values are usually given relative to the Sol. So a star with a metallicity twice that of Sol has twice the relative proportions of heavy metals to hydrogen and helium than Sol.
As well as giving information about a star’s composition its metallicity is also an indirect measure of how old a particular star is. After every generation stars the interstellar medium (ISM) – the nebulous gas and dust from which stars form – is enriched with the dying remnants of stars throwing their atmospheres into space. This debris contains the elements that the star formed over its long life span. This enriches the ISM with metals so the next generation of stars have correspondingly larger metallicities.
Astronomers can use metallicity to divide stars into three groups termed Populations.
The three stellar populations are as follows:
- Population I stars are stars of similar or greater metallicity than the sun. In the current epoch they are the most common variety of stars present in the Universe
- Population II stars are the oldest stars currently detected and have very low metallicities. They are all red or orange stars (spectral class K and M) as the other heavier hotter stars born in the same time period have long since depleted their fuel reserves and burnt out.
- Population III stars were the first stars formed after the Big Bang. As such they would have virtually no metals in their structure and for reasons touched on later would have been many times the mass of the Sun. As all such stars would have burnt out within a few million years none have yet been detected as they would only be visible for a very short period of cosmic time and we currently do not have the technology capable of detecting them in the afterglow of the Big Bang observable to us today.
It is worth a note that the Populations are numbered in the reverse order suggested by common sense. Population III stars are essentially the 1st generation of stars with Populations II and I indicating later generations. It is also important to note that a population may contain more than one generation of stars and the line between each is somewhat ambiguous.
Now lets look at how the material to form the original stars was produced in the first place.
For a duration of about seventeen minutes, between three and twenty minutes post the Big Bang, the universe had the correct conditions (temperature, pressure and density) to serve as a nuclear fusion reactor; similar to the core of a star. These extreme conditions allowed the soup of sub atomic particles to fuse and in doing so form atomic nuclei (though not atoms as the conditions remained far to energetic for electrons to become associated with these nuclei for about 380,000 years).
Nucleosynthesis was initialized after the majority of sub-atomic particles had been formed following the Big Bang – that is after the slight asymmetry between matter and antimatter became evident, allowing ‘normal’ matter to come to dominate our Universe.
One of the most fascinating thoughts about this process is that it occurred everywhere in the observable universe at the same time. That includes the space where my laptop is sitting as I type this, as well as the space now occupied by your brain.
So what exactly happened during this time? To answer this we must first look at the initial conditions as the process begins.
The two basic building blocks of all atomic nuclei – the proton and the neutron (each composed of three quarks) – had already been produced by in large before the onset of the process. Secondary school chemistry would have you believe that the proton and neutron have the same mass, this I’m sorry to say isn’t entirely true. A neutron weighs in at 1.674927351×10−27 kg whilst a proton is slightly lighter with a mass of 1.672621777×10−27 kg. This tiny difference of 2.305574×10−30 kg can safely be ignored in almost all practical cases (including most if not all secondary school chemistry and physics exams😉 ) but becomes very important to our story.
As Einstein laid out with is mass-energy equivalence equation E=mc2 (arguably the most well known equation in physics), mass and energy are really two side of the same coin. Mass (under current understanding at least) is the most concentrated form of energy possible, indeed one gram of matter contains the energy released by the detonation of 21.4 kilotons of TNT. If we rearrange the formula we can see why the mass difference between the proton and neutron is so significant.
m=E/c2 – This may not look that much different but it reveals a great deal. For a fixed amount of energy in Joules, the equivalent mass is (tiny though it may be) mathematically calculated by dividing the quantity of energy by the speed of light squared which is about 9×1016ms−1. So for 1000J the equivalent mass is about 1.11×10-14 kg, demonstrating how such a tiny difference in mass allow for such drastic implications as we are going to look at now.
Just after the Big Bang the universe was an exceedingly hot soup. Particles popping into existence at random, before encountering their antimatter partner and annihilating each other in a flash of gamma rays. As stated above, as the universe expanded it cooled rapidly, as it did so the slight difference between probability of a ‘normal’ matter particle being generated and its antimatter opposite (in the favour of the ‘normal’ version) allowed our universe to become dominated by ‘normal’ matter. As the universe cooled these particles (the quarks) joined up to form the familiar protons and neutrons. I’m sure you are now wondering why I waffled on about their differing masses for three paragraphs – I’m getting there!
NGC225 Credit: Ken Crawford
As protons are ever so slightly less massive, they require a lot less energy to generate and so more spontaneously popped into existence from the primordial fireball compared to neutrons. This effect was so significant that the universe had seven times as many protons as it has neutrons at the start and end of this first phase of element building. This explains why the universe has an inordinate amount of hydrogen a whopping 75% by mass1 of all ‘normal’ matter – the simplest element containing just a single proton as its nucleus (for the chemists I am explicitly dealing with the lightest isotope protium here rather than the heavier deuterium and tritium which do indeed contain neutrons).
Adding to this ‘proton bias’, free neutrons (that is to say, neutrons that are not bound into atomic nuclei) are unstable and tend to decay to protons within about 15 minutes give or take a bit. Thankfully for the neutrons the density of the early universe was high enough to allow the majority to be incorporated into stable nuclear configurations within the first few minutes, thus avoiding a neutron deficient scenario where most had already decayed.
Despite helium-4 (the most common isotope of helium) being more stable than either a free proton or neutron, and thus should be relatively easy to form, the process encounters a snag. You can’t simply fuse two free protons and two free neutrons together at once to produce a helium nucleus, the process must first pass through an intermediary step of two deuterium atoms. Deuterium is a heavier isotope of hydrogen containing one proton and one neutron, though unlike helium is somewhat unstable and as such any deuterium that did not immediately collide with another deuterium nucleus was broken back down to its component proton and neutron. This in turn prevented any major nucleosynthesis to occur until after the universe had cooled below about 300 million Kelvin. This restriction for the commencing of the majority of the fusion reactions is termed the Deuterium Bottleneck.
Deuterium formation and breakdown. Red indicates a proton and grey a neutron. Credit: Peter Clark
Once the universe had cooled past this point the reactions kicked into overdrive (as deuterium nuclei are able to remain stable at these temperature) with hydrogen being converted to helium via deuterium at a rate not seen since. Though the fact that we can detect any deuterium at all is very telling. As no known process other than primordial nucleosynthesis could produce anywhere near the detected level of deuterium in the universe today (despite that proportion being quite tiny), meaning that virtually all the deuterium in existence was produced in the first twenty minutes of our universe.
He-4 Production Credit: Peter Clark
Primordial nucleosynthesis is therefore tightly constrained by the level of deuterium present within the universe. If it had continued much past the projected twenty minutes, most perhaps even all deuterium would now be tied up within Helium-4 nuclei. So the detected level of deuterium allows us to determine a great deal about this age of rapid element building.
The brief duration of the process also set up restrictions also set up restrictions on the possible final products. Without any ‘massive’ nuclei specifically a stable nucleus containing 5 or 8 nucleons (being protons or neutrons) rapid build up of any further elements is impossible. Such build up requires extremely rare circumstances that produce even heavier nuclei containing more nucleons, and can only occur in significant numbers over millions of years in the cores of high mass stars.
Two He-4 nuclei can collide and fuse to produce a highly unstable Beryllium-8 nuclei, this under normal circumstances would decay back to the original two He-4 nuclei. This process is exceedingly rapid with the half-life being slightly longer than 6.7×10-13 seconds.
Be-8 Formation and Decay Credit: Peter Clark
However, very occasionally a third He-4 nucleus can collide and fuse with the Be-8 nucleus before it decays. This produces a stable Carbon-12 nucleus which in turn can go on in a whole new series of fusion reactions in turn producing all the heavier elements.
Carbon-12 Production Credit: Peter Clark
This process is incredibly slow taking millions of years for any appreciable masses of carbon to be produced and so only a few very isolated atoms of carbon would have been produced in this epoch of the universe. The process eventually becomes significant allowing for the initiation of the CNO cycle in high mass stars.
Taken together, the current models suggest that beryllium would be the heaviest element produced in any (tiny) quantity outside of exceedingly rare freak events as part of primordial nucleosynthesis, with the remaining heavier elements requiring longer term build up within stars, supernovae and through the action of cosmic rays (cosmic ray spallation) long after this first burst of activity had ground to a halt.
New observations however have detected unusually high levels of boron isotopes in some very ancient red dwarfs. This cannot be explained though standard models as the stars are too old to have formed from sufficiently enriched material to contain such levels of boron (produced almost exclusively in Type Ic supernovae2) and such serve as an indication that our current understanding may be incomplete.
The next post in the series will deal with the internal structure of stars and the processes that allow a star like the Sun shine for several billion years.
1 Made even more impressive by the statement that just under 92% of all atoms in the Universe are hydrogen with helium filling up the majority of the remainder at just under 8%