Thursday 25 February 2016

Why we don't know why stars become red (super)giants

A while ago I started saying in talks at conferences “that we don’t know why stars become red giants”. I’ve had to stop saying this as people start asking questions and are disbelieving about that statement. What I really mean is that we know red giants exist in nature and we know stars evolve into red giants in our stellar evolution code but we have no simple explanation of why this happens.

First it's worth describing what is a red giant star (wikipedia article here). Red giants are post-main sequence stars, they have completed their hydrogen to helium fusion and have formed an almost pure helium core in the centre of the star. When this happens burning no longer happens at the centre of the star but in a shell around the core. The surrounding envelope expands greatly in radius, cools and increases in luminosity with the star becoming much brighter. Our own Sun is likely to become a red giant 100 times its current radius. However some red supergiants can be 1000s of times the Sun's radius.

So here is the key question, "why does a star with a helium core become a very large giant star?" When I mention this to other astrophysicists it is common that someone starts talking about “the mirror effect”. This is something that is in many textbooks but I have yet to find a derivation of the effect. It suggests that when there is a burning shell in a star whatever happens on one side of the shell, the reverse happens on the other side. Because the red giant helium core has no source of nuclear fusion it actually collapses and heats up, until the temperature is high enough for helium burning. But therefore the mirror effect implies that the envelope will expand and cool.

There are problems with this view, one it appears to be phenomenological. Just explaining what is seen without a rigorous definition. The key question we should ask about the mirror effect is: “what is being conserved?" It can't be energy and mass, it really can't be size or volume. So we are back to not understanding in a simple physical understanding why stars become red giants.

At some point in their own personal evolution every stellar theorist turns to this question. I include some useful references for interested reader below, however they are quite technical papers. The method most commonly used to understand why stars become red giants is to make models in a stellar evolution code and to artificially take out or change the physics until the star doesn’t become a red giant.

One interesting method has been to include “pseudo helium” in codes, something that looks like helium for a nuclear burning perspective but has the mean molecular weight and chemistry of hydrogen. This appears to show that one of the drivers to red giant-ness is the mean molecular weight in the core of a star.

The mean molecular weight is the average mass per particle. Nuclear fusion increases this as a star turns hydrogen into heavier elements. At the beginning hydrogen has a mass of 1 proton and 2 particles, a proton and an electron if ionized. However helium has the mass of 2 protons and 2 neutrons but over 3 particles, one helium nucleus and 2 electrons. Therefore the average mass of each particle is higher after nuclear fusion. If you take out this increase in mean molecular weight it doesn't stop a stellar model becoming a red giant but it does make it more difficult.

So in summary we know stars become red giants in nature and in our models but we can't give you a simple explanation and it isn't the mirror effect. It's probably something to do with the fact that stellar material becomes denser as nuclear material is processed from hydrogen to helium.

One final point to note is that one factor people sometimes use to check their models of stars to is the ratio of red supergiants to blue supergiant, i.e. cool to hot post-main sequence stars. It has always been thought to be sensitive to why stars become red giants but this is in a single star view of stellar evolution. We know today that 70% of stars have their evolution affected by a binary interaction. One thing is when a star becomes a red supergiant, it gets bigger, if another star is nearby to two will interact. This can reduce the expected number of red giants to one third of the expected number.

What this means that at least if we see a red giant at least it's most likely a single star when we see one! Also it means that stellar evolution is not the completed and finished subject that some people assume and there is still a number of mysteries.

Further reading:
Sugimoto & Fujimoto (2000)
Stancliffe et al. (2009)
Ball et al. (2012)