Title: Luminosity Bias (II): The Cosmic Web of the First Stars
Authors: Barkana, R.
Author’s institution: Raymond and Beverly Sackler School of Physics and Astronomy, Tel Aviv University, Tel Aviv 69978, Israel
In today’s paper, Rennan Barkana discusses how an until-recently unknown effect in early Universe cosmology should make it possible to see back farther into stars’ and galaxies’ histories than ever before. To understand how, we need to know a bit about the early Universe. What follows is a lightning introduction.
First, one piece of terminology: reionization. Reionization refers to the time when the Universe became ionized (electrons separated from protons, so no more neutral atoms, only free electrons and protons whizzing around) for the second time. Why second time? Well, the key fact here is that the Universe has been, since its beginning ~13.7 billion years ago, expanding. So, as you look backwards in time, the Universe was smaller, hotter, and denser than it is now. Hotter means more energy was available to allow the electrons in atoms to whiz fast enough to escape their protons. So, as you go back early enough in time, the Universe was all ionized. This was the case between 0 and 300,000 years after the Big Bang.
After that, the Universe became sufficiently cool that the protons and electrons could recombine, so it was no longer ionized. This remained so until enough stars and galaxies had formed to radiate energy sufficient to heat patches of the Universe and reionize them. This era is the epoch of reionization, and occurred about 500 million years after the Big Bang. It is still not completely understood: there are many complicated ways stars can affect the surrounding gas, and many different assumptions that go into modeling both this and even the formation of the stars themselves.
One major probe of reionization is the 21 cm line. This just refers to the energy of photons produced by a particular change in hydrogen atoms. (Remember, energy is proportional to 1/wavelength, so a statement of wavelength can be translated to a statement of energy). Protons and electrons have spin, and they can be spinning in the same direction or in opposite directions. When the direction goes from anti-aligned to aligned, a photon is emitted. Because this can only happen if the hydrogen is neutral, how many photons we see with wavelength 21 cm coming to us from a particular time in the Universe’s history probes how much hydrogen was neutral at that time. (Remember, looking out is looking back in time because light travels at a finite speed, so seeing photons coming to us from far away is seeing photons from a long time ago!) Because the line is from neutral hydrogen, it can also, in principle, probe the time before reionization, around 200 million years after the Big Bang.
However, the 21 cm signal from this earlier era, which is when the first galaxies formed, has been thought to be far beyond the sensitivity of current and even intermediate-future instruments. This is where Barkana’s paper, an invited review for a recent cosmology conference in Marseille, fits in.
Basically, in the rightmost (earliest in time) yellow region on the graphic at left, when the Universe was ionized and had not yet ever been neutral, regular matter (known as baryons), stuff such as composes you and me, was not able to fall under the influence of gravity because it was actually supported by the pressure of light itself (“radiation pressure”). Today, because the Universe is much larger than it was then, radiation pressure cannot do this, but when the Universe was a factor of 1000 smaller, and ionized, it could. The light providing the pressure was primoridal: not produced by any stars, it just was an initial component of the Universe, just as were matter and dark matter. The fact that the Universe was ionized is also key here: light interacts very strongly with free electrons (and the protons follow the electrons because opposite charges attract), but much less so with neutral atoms. So it was only in the first yellow region that light could support baryons against gravity: once the Universe cooled enough that it became neutral, the baryons began to fall under gravity. This epoch is called “cosmic recombination”, even though it is really the first time protons and electrons combine into neutral atoms.
On the other hand, there is another, mysterious form of matter called dark matter that experiences the force of gravity but does not interact with light. Dark matter was moving under the influence of gravity even when the Universe was ionized. Hence, by the time the baryons started to move too, the dark matter had a head start. The dark matter was moving faster than the baryons, and so there was a relative velocity between the two.
This relative velocity has been known for some time, but what was not known until 2010 was that it would actually have serious effects on how galaxies and stars form. Barkana summarizes three main effects that occur where the relative velocity is large. First, there will be fewer low-mass objects that form. Second, in each object that does form, there will be less gas. Third, for reasons I won’t detail here, the gas in each object will be less able to cool. These three effects mean, basically, that there will be fewer stars in regions of the Universe where the relative velocity is large.
What complicates the story is that the relative velocity varies spatially: in some regions of the Universe it is large, in others it is small. In the regions where it is small, it does not have much of an effect on star formation. But in the regions where it is large, as I said above, it will suppress star formation.
Hence, the relative velocity will actually enhance the contrast between regions where it is large and regions where it is small: the regions with small relative velocity will be brighter relative to the regions with large relative velocity. This increased contrast, and the fact that the contrast will modulate on very large, comparatively easily detected scales (because the relative velocity changes only on large scales), makes the 21 cm line easier to detect. That in turn means it should be detectable farther back in time than previously thought.