One of the main challenges in modern cosmology is to understand how the very low-density matter between galaxies (known as the inter-galactic medium, or IGM) came to be hot and ionized today, reaching temperatures of up to 10 million degrees. It hasn’t always been this way – after the Big Bang the Universe expanded and cooled, eventually reaching temperatures low enough for much of the Hydrogen and Helium plasma within it to combine and form a neutral atoms in a process known as recombination around 378,000 years after the Big Bang. After this, the expansion and cooling of the Universe continued for hundreds of millions of years, leaving it in a dark and increasingly cold state – an era cosmologists refer to as the ‘Dark Ages’.
So, how did the gas in the Universe become hot and ionized?
The main part of the answer to this question is thought to be the process of cosmic reionization, which forms the focus of my work. I’m Ellis Owen, a 1st Year PhD Student here in the Theoretical Astrophysics group at MSSL, and my research looks at the way in which some of the early astrophysical objects and sources of light in the Universe began to heat it up and ionize the gas within it (please take a look at my earlier MSSL blog post if you’re interested to read more about how this process began around the first stars).
We know there are several ways to heat up the intergalactic medium with the tools present during the Epoch of Reionization (EoR) – initially these were the first (Population III) stars, characterised by their high masses, low metal content and their strong emission of ionizing ultraviolet light. As the metal content of the intergalactic medium increased (due to the death of Population III stars and their subsequent contamination of the intergalactic gas with heavier metal products from nuclear fusion) different populations of stars – Population II stars – could begin to form. These had higher metallicities (metal content) due to the increasingly metal-contaminated composition of the gas from which they formed, and resulted in Population II stars having lower stellar masses than their predecessors. This is because the presence of metals in star-forming gas clouds allows for their more efficient cooling and collapse, leading to more cloud fragmentation and hence more stars of lower masses forming from these clouds.
Like the Population III stars, the first galaxies of smaller Population II stars emitted radiation strongly in the ultraviolet part of the spectrum, ionizing the gas directly around them. However, as their stellar populations evolved, these early galaxies also came to include a more varied population of objects. This would be due to the death of subsequent generations of stars injecting metals into their surrounding gas, allowing for smaller (and more metal-abundant) stars to emerge, emitting radiation across broad parts of the spectrum. Remnants of the dead stars also developed into high-energy astrophysical environments, able to contribute X-ray, gamma-ray and higher-energy emission.
Within galaxies, stars could be born, evolve and die. Their death would culminate in a highly energetic supernova explosion, leaving some kind of remnant. This might include a white dwarf, neutron star or black hole depending on the nature of the dying star and its surrounding environment. In cases where a pair or group of stars evolve in close proximity, it would not be unusual for one of the (more massive) stars to die first then, later its remnant could accrete gas from the remaining partner star as it also comes to the end of it’s life and begins to expand as it runs out of nuclear fuel in its core. This swelling of the dying partner star eventually means that its outer gaseous layers overspill the star’s gravitational well (a process known as Roche-Lobe overflow), and accretes on to the more dense remnant.
The accretion of this overspill gas – and, indeed, accretion in general – is an important process with regard to the heating and ionization of the intergalactic medium – it’s not just a mechanism by which a dense stellar remnant devours surrounding matter to increase its mass. The accretion of gas onto a massive object heats it up and causes it to emit a thermal spectrum of radiation. Interestingly, this is one of the most efficient ways by which matter is converted into energy, with orders of 10% of the rest mass energy of the accreting gas being converted into radiation (this compares with just a few tenths of 1% of mass being converted into radiation by nuclear fusion processes within the cores of stars) – and this efficiency increases with the compactness of the accreting object.
The mass and size of the accreting object determines the ionizing properties of the radiation emitted. For instance, for binary star systems the emitted radiation falls strongly in the X-ray part of the spectrum, whereas for Active Galactic Nuclei (which consist of a Supermassive Black Hole in the centre of a galaxy, accreting matter) tend to emit radiation more strongly in the lower-energy ultraviolet to infrared parts of the spectrum.
While all radiation of ultraviolet frequencies (or higher) can directly ionize neutral Hydrogen, the higher energy radiation tends to be more penetrating (for instance, X-rays) and so can reach further into the surrounding neutral medium before depositing energy and causing ionization. So, for strong ultraviolet emitters, a Stromgren-like ionization region may evolve (like that described by the bubbles in my previous blog post) while strong X-ray emitting systems will tend to deposit energy over a much broader region, partially ionizing some fraction of the IGM further away from the radiation source and allowing the electrons (which carry any excess energy after ionization) to heat the remaining semi-neutral gas, causing further ionization (by collisions of the heated particles).
Clearly there are many ways these early galaxies and their various constituent objects are able to heat up and ionize the surrounding intergalactic medium. But to study exactly how they contributed to the progression of cosmic reionization, they need to be observed at very large distances.
While some of the World’s most advanced telescopes may look very impressive, there is only so far they can see. In order to study the early Universe, we need to look at light which was emitted by it. As light has a fixed speed, the further away we look into space, the further back in time we are seeing (this is why astronomers sometimes measure distances in terms of light years – the distance light can travel in one year). So, to look back to the end of the Dark Ages and the time at which the first galaxies began to appear, we need to look to incredibly far distances, at light emitted around 13 billion years ago when the Universe was only 5% of its current age.
Unfortunately, we can’t really do this with current telescopes. The distances involved are too large and the galaxies we want to study simply aren’t bright enough. But occasionally, if we are lucky, nature offers a helping hand and shines a light for us to see – an incredibly bright one, in fact. These are highly energetic explosions (thought to be the most energetic events in the Universe, often outshining entire galaxies) known as Gamma-ray bursts (GRBs). While they occur in the current Universe, they are also known to have arisen right back to the earlier stages of the Epoch of Reionization, offering occasional sparks of light that we can see, exploding out of the otherwise Dark Ages of the Universe.
The actual physical nature of GRBs is an area of active research, and those closer to us which can be seen more easily have been observed to originate within galaxies, perhaps resulting from merging Neutron Stars or huge Supernovae explosions – so observations of them up to high redshifts allows them to act as signposts for the existence of early galaxies that wouldn’t otherwise be detectable.
There have been a number of high redshift observations of GRBs, such as those with SWIFT, which occurred during reionization era. The number and distribution of these can pinpoint host galaxies during reionization. Considering that only the brightest of these GRBs are observed as such large distances, some estimate can be made as to the total number of GRBs occurring during reionization based on the fraction of such events above a certain luminosity at a distance where we would expect to see them all.
Unfortunately, just knowing how many host galaxies there are, and their distribution across space, isn’t quite enough to find out exactly how they contributed to reionization. A good understanding of the amount (and energy distribution) of the radiation that these first galaxies emitted is also needed – and this is something that is less well understood. Theoretical models can be built from information about the metallicity of the material from which the stars in the galaxies formed (which can be observed by looking at the absorption lines in the spectrum of distant gas, backlit by a GRB) while estimates of the ionization fraction of intergalactic gases can also be made with distant GRB observations. But, for the most part, this link between GRB observations and the ionising ultraviolet emission from their host galaxies during the epoch of reionization, still remains an open question.
To find out more, see: