Blurring the Boundaries
Aaah, 2006. The year of the great Pluto debacle! I have to admit to having been left unsatisfied by the IAU’s resolution on whether or not Pluto can be said to constitute a ‘planet’. In the end they designated it a ‘dwarf planet’, a designation that smells faintly of the sort of confused fudges beloved of politicians. Science is supposed to supply clarity, and precision on definitions is important.
Except sometimes, one has to be careful with definitions. They can cause more trouble than they solve.
The basic problem here, I think, is more to do with human beings than it is to do with what’s going on in the sky. The human animal is a categorical one – we like to drop objects and phenomena into neat, tidy boxes. It’s probably a side-effect of the sort of pattern-recognition that enabled our ancestors to spot predators lurking in the bushes, and tell them apart from the prey they were stalking. All very useful, from an evolutionary point of view, but not a behaviour that always maps so well onto an untidy, messy universe.
In particular, I’m thinking about this in a brown dwarfs context.
Pluto has some of the characteristics of a planet and some of the characteristics of a more conventional Kuiper Belt object. In an analogous manner, brown dwarfs can have some of the characteristics of giant planets and also some of the characteristics of stars. But they also have a whole set of properties all of their own.
The textbook, lying-to-children description of a brown dwarf goes something like this: an object with a mass range between 13 Jupiter masses to 7.5% of a solar mass. That way it’s just big enough for deuterium fusion to occur for a time, but not big enough for stable, sustained hydrogen fusion to happen.
So far, this is all nice and tidy and neat. And better yet, there’s even a very limited subset of objects for which this description is actually true! (Wonders will never cease…) This subset, in case you’re wondering, is for brown dwarfs with solar metallicity.
However, there is a ticking time-bomb lurking in the neat, neurotically-tidy definition: metallicity. You see, metallicity has a huge impact on stellar structure. Change the balance of elements inside a star and you can alter its opacity (its ‘transparency’ to light), you can alter the distribution of pressures inside it and in extremis, you can even alter the process of fusion itself. (The CNO cycle, for instance, uses carbon, nitrogen and oxygen as catalysts for the fusion of hydrogen – take the catalysts away, and no CNO cycle.)
Now, people have wondered if all this has implications for the fusion cutoffs. And it turns out, yes it does. Quite dramatic ones, in fact. In the limiting case of zero metallicity (no heavy elements of any kind), the hydrogen-burning threshold creeps up to 1/10th of a solar mass. For supersolar metallicities, however, the hydrogen burning limit can fall to as low as 4% of a solar mass.
So, a zero metallicity brown dwarf could weigh in at just a shade under 0.1 Suns – heavier than many solar-metallicity stars! – whereas a supersolar-metallicity star may weigh in at just more than 4% of a solar mass, meaning that an actual star would be lighter than many brown dwarfs.
Feeling confused? It’s only to be expected. This is the sort of muddle that happens when human ideas about tidiness and category-neatness collide with a complex and messy universe.
Also, a similar variance can happen for the deuterium burning limit as well. Change the metallicity and you can move that up and down as well. So we swiftly arrive at a situation where some things of the same mass are stars and some things are brown dwarfs. But there’s worse – you see, depending on how you look at it, some of the brown dwarfs could also actually be planets.
Unlike planets or stars, brown dwarfs are supported by something called electron degeneracy pressure. This is a consequence of quantum mechanics, and its origins are probably beyond our discussion here. Electron degeneracy starts to put in an appearance at about 4 Jupiter masses – which means you have some nominal ‘planets’ whose internal behaviour is much more like brown dwarfs!
And then there’s the good old spectrum issue. To make the point, here’s Jupiter in the near-infrared, overplotted with a late-type, ‘cold’ T dwarf:
They’re hardly identical, and Jupiter’s shorter wavelengths are dominated by reflected Sunlight – but you can see that there is some similarity. Even now, 4.5 billion years after its formation, Jupiter still radiates more heat than it receives from the Sun. So in this spectroscopic sense, one of our familiar, local planets is behaving like a brown dwarf!
To summarise, if there’s one thing I want you to take away from this post, it’s to treat nice, tidy theoretical definitions with a degree of healthy skepticism. It’s not necessarily that they’re wrong, as such, but rather that any theoretical model is of necessity simpler than the object it describes. Inevitably there will be gaps and errors, and some types of object will blur the boundaries. This is life, and this is why the Universe is worth studying – it’s full of these fascinating little odd pockets!
 Although I’m sure we’ve covered it before, a terminology note for new readers: a star’s metallicity is the proportion of its mass that isn’t composed of hydrogen or helium. The reason for this distinction is that H and most of the He come down to us from the Big Bang, but everything else was made inside stars and novae. (Very technically, the Big Bang also made some lithium, but lithium is pretty rare, so for most purposes we can ignore it.)
 See here for more on the CNO cycle.
 In detail, it’s due to the Pauli exclusion principle. Inside a brown dwarf, the pressures are far too high for conventional Coulomb pressure – electron shell on electron shell repulsion – to hold matter up against its own weight. But there is no thermal pressure – heat from fusion – to counter the weight, so the core collapses until the electrons are ‘squeezed’ away from their atoms. Adding an electron to this volume requires raising the other electrons’ energy levels to ‘make room’, and there’s only so much room to be had. Beyond this point, the electron-degenerate gas effectively starts ‘pushing’ back out and excludes new material from being added.