Big Bang, Give Me a Twirl

Radio telescopes reveal surprisingly neat rotation in extremely young galaxies

“We are made of star stuff.” These timeless words from astrophysicist and space evangelist Carl Sagan are more than poetic rhetoric. The carbon, nitrogen, oxygen and silicon—so central to life on earth—and the iron in our blood are all produced when hydrogen fuses into heavier elements in the cores of massive stars. The nuclear fusion process is what powers these stars to light up our sky. As the stars age and die, they release these elements into the gas around them, which will eventually form planets, sometimes populated by sentient organisms that build telescopes to peer back out into where they came from.

A team led by Renske Smit, a postdoctoral fellow at Cambridge University, recently measured the star-forming gas in two galaxies that sent us light when the Universe was just one seventh of its current size. This means that any photon—a packet of light—that left a source at that time would, therefore, have its wavelength stretched out, or made redder on the electromagnetic spectrum, by this same factor. Astronomers say the galaxies are at “redshifts” of 6.8. In effect, we are seeing these galaxies as they were when the Universe was just 800 million years old.

Using the Atacama Large Millimeter Array (ALMA), a collection of radio telescopes in the high, dry Chilean desert, Smit looked for the light emitted by carbon atoms. When a carbon atom collides with another particle, the energy of that collision can kick one of the outer electrons of the carbon to a higher energy level, which is called collisional excitation. But electrons like to stay in the lowest energy states available, so it will soon jump back down. In the process, it releases a photon of wavelength 157.7 micrometers, which matches the energy difference between those two electron levels.

Quantum mechanics dictates that an atom will always emit a photon of exactly the same wavelength for a given electron transition. However, if the atom is moving away from us, this wavelength will get longer and redder; if it is moving towards us, the light gets shorter and bluer. This is known as the Doppler effect; most of us experience the aural version of it every time an ambulance drives past, its siren getting shriller as it approaches and then dropping as it drives away.

Since the Universe is expanding away from us, light coming from its most distant sources is red-shifted. Smit explains that the brightest emission in the most distant galaxies, which has optical wavelengths in a lab, becomes redshifted into the mid-infrared. In her PhD thesis at the Leiden observatory in the Netherlands, Smit used the infrared space telescope, Spitzer, to measure redshifts precisely for a very large sample of galaxies. The CII line, with a wavelength of 157.7 micrometers, is already in the infrared, but it gets further redshifted into longer radio waves. This is exactly what ALMA can detect.

“Getting time on ALMA is hard,” said Pascal Oesch, former postdoctoral fellow at Yale and now assistant professor at the University of Geneva, a co-author on the paper. The entire array of telescopes must be reconfigured every time a researcher wants to look at a different range of wavelengths. “You really have to know the redshifts and locations of the targets, and Renske constrained those tightly with her Spitzer observations,” Oesch said.

Now picture a disk of swirling gas, moving clockwise. Rotate the disk so you’re viewing it from the side. Gas to the right half of the disk will appear to be moving towards you, and on the left it’ll be moving away from you. If there were carbon atoms emitting photons at exactly 158 micrometers everywhere in the disk, you’d think the light from the right side of the disk actually had a wavelength shorter and bluer than that, and that from the left larger  and redder. Now think of two coins sitting on this disk, at different distances from the center. As the disk rotates, the coin farther from the center moves a longer total distance than the one closer in. In other words, the velocity of the disk is greater at larger radii. So the 157.7 micrometer line is redshifted increasingly more the further left you look from the center, and blueshifted more the further you look right. The line gets broadened into a bell shaped curve, the width of which tells you how fast the gas in the disk is rotating.

Current simulations of galaxy formation in the early Universe show these galaxies colliding with their neighbors often in what are called mergers. These mergers disrupt the formation of any disks, and tend to leave behind gas clumps with lots of star formation in the center. “We expected the carbon emission to be pretty concentrated,” Smit said. Therefore, she only expected to see lines from the center of the rotating disk. Instead, the line-emission was extended enough that she could measure the velocity variations across the galaxy. “The fact that we could actually see the rotation meant that CII emission was more spread out,” Smit said.

That was not the only surprise. The CII line emission is only generated  if the carbon undergoes a lot of collisions, usually with electrons coming from dust. “We see the carbon line but we don’t see any dust, and that is a puzzle we haven’t solved yet,” Smit said.

Oesch doesn’t think it’s that surprising to find so little dust in these galaxies: dust is released during a relatively late stage in a galaxies’ evolution, and since they are still so young, the stars simply may not have reached this phase of their lives. Further, he says, they may not have found dust because they were looking for a very specific kind. “You have to make assumptions of the temperature of the dust to predict how much emission you would see,” said Oesch. It is just another example of how carefully astronomers have to design their experiments to encompass all of the components of a galaxy that they’re interested in.

Astronomers are also very careful about making inferences from small samples. Smit is excited about the upcoming James Webb Space Telescope, which will detect hundreds or even thousands of galaxies at these high redshifts. James Webb will have an IFU, or  Integral Field Unit, a device that takes a spectrum on every pixel of the camera. “We’ll at least be able to get low resolution but large samples,” she said.

Smit already has time on ALMA to look at six more galaxies, which will help us build a picture of how common galaxy disks really are in the early Universe. She is also preparing to observe one of these galaxies at a much higher resolution. “We’ll be able to see how organized the disk is, or if its messier than we thought. It’ll tell us about the physics of at least one disk in much more detail,” Oesch said. She emphasizes that this really a new frontier of observation. “Maybe it’s not a single disk—maybe it’s multiple clumps merging! We really haven’t been able to do any of this before.”


  1. Smit et al. “Rotation in [C II]-emitting gas in two galaxies at a redshift of 6.8.” Nature, Volume 553, Issue 7687, pp. 178-181 (2018)
  2. Smit et al. “High-precision Photometric Redshifts from Spitzer/IRAC: Extreme [3.6] – [4.5] Colors Identify Galaxies in the Redshift Range z ~ 6.6 – 6.9”. The Astrophysical Journal, Volume 801, Issue 2, article id. 122, 12 pp. (2015)
  3. Interview with Dr Renske Smit, Kavli Institute of Cosmology, University of Cambridge. Interview on 02/01/18.
  4. Interview with Prof Pascal Oesch, Observatoire de Geneve, University of Geneva. Interview on 02/09/18.
  5. ESO website: Retrieved 02/10/18
  6. Retrieved 02/10/18