Cosmic Dawn

by George Rhee

  Fig. 7.10A plot of dark energy density ΩΛ versus matter density ΩM. The shaded areas represent areas of the plot that are consistent with certain observations. The blue ellipse shows the constraints from the Type Ia supernova data. The values of ΩΛ and ΩM must lie within the blue area to agree with the supernova data. The values must also lie within the green shaded area to agree with the observations of the clustering of galaxies. The values must lie within the orange area to agree with the WMAP cosmic background observations. The values must lie within the grey ellipse to agree with all three sets of data. The fact that the grey ellipse lies above the red line means the expansion of the universe is accelerating and the universe will expand forever. The fact that the black line goes through the grey ellipse means the geometry of the universe is flat (seet also Fig. 7.8)

  The WMAP measurements have enabled us to determine the parameters of the universe to high accuracy but they do much more. To understand how galaxies formed we must know what the universe looked like before they formed. If we compare the density of our galaxy with the density of the universe we find that our galaxy is today about 200,000 times denser than the mean density of the universe. The measurements of very small temperature fluctuations in the background radiation temperature. These variations imply density variations of one part in 100,000. That is to say the density of the universe varied little from one place to the next when the universe was very young. You might measure the density to be 1.0000 times the mean density in one place and 1.00001 times the mean density in another. This is very different from the universe today where the density varies a lot from one place to the next. As we have mentioned, the density in our galaxy is many times larger than the mean density of the universe today.

  The cosmic background measurements made it possible to extract physical parameters from the galaxy surveys. The brightness variations gave us a benchmark for starting the computer simulations that explain the appearance of the universe today. This sets the stage for Part III of this book which is about the attempt to see and measure the formation of galaxies during the time period between the galaxy maps and the cosmic background observations. This is the new frontier of cosmology.

  Further Reading

  The Music of the Big Bang: The cosmic Microwave Background and the New Cosmology. A. Balbi, Berlin, Springer-Verlag, 2010.

  The 4% Universe. R. Panek, Boston, Houghton Mifflin Harcourt, 2011.

  Finding the Big Bang. P. Peebles, L. Page and R. Partridge, Cambridge, Cambridge University Press, 2009.

  Observational Cosmology. S. Serjeant, Cambridge, Cambridge University Press, 2010.

  Part 3

  The Search for the Cosmic Dawn

  George RheeAstronomers' UniverseCosmic Dawn2013The Search for the First Stars and Galaxies10.1007/978-1-4614-7813-3_8© Springer Science+Business Media, LLC 2013

  8. The Search for Light in the Dark Ages

  George Rhee1

  (1)Department of Physics & Astronomy, University of Nevada, Las Vegas, Nevada, USA

  Abstract

  One of the characters in the 1950s British comedy radio series, The Goon Show, once remarked that “Everybody’s got to be somewhere.” The answer to the question of where we are in the universe and how we got there has changed dramatically over the centuries. It is a question that all cultures try to answer in one way or another. We begin with a Native American myth and then discuss Greek thought and the idea of rational inquiry. The development of theories of planetary motion are discussed leading to the work of Isaac Newton. The implications of Newton’s theory for the idea of an infinite universe are presented. The telescope enters the stage, and we discuss its use in changing our view of the solar system. The discovery of nebulae by telescopic observations leads us to the story of how the nature of galaxies was revealed. We end with the discovery of the expanding universe and the idea of the Big Bang.

  Following the recombination and the formation of the first atoms, the early universe was a nearly formless primordial soup of dark matter and gas: there were no galaxies, stars, or planets. This was truly the dark ages. Things began to change when the slightly denser regions began to contract under the relentless pull of gravity. It took a few hundred million years, but eventually these dense regions gave birth to first stars, and black holes so that the universe became filled with light. These events lie largely in the realm of theory today, and existing telescopes can barely probe this mysterious era. Over the next decade, we expect this to change.

  Committee for a Decadal Survey of Astronomy and Astrophysics

  From half a million years to five hundred million years after the Big Bang, the hydrogen in the universe was in the form of neutral atoms. At earlier times, the cosmic background radiation had sufficient energy to keep the hydrogen atoms ionized. At later times the ultraviolet radiation from young stars and quasars was sufficient to keep the hydrogen gas between the galaxies ionized. The period between recombination and reionization during which the hydrogen gas was in the form of neutral atoms is known as the dark ages. The light emitted at 21-cm during the dark ages will be redshifted to meter wavelengths. Radio telescopes have been built to detect the imprint of neutral hydrogen at these early times. The majority of light in the universe during this period was in the form of cosmic background radiation left over from the Big Bang. During this period this radiation travels (almost) unimpeded through space. We say almost because the neutral hydrogen will leave a small imprint on the light by either adding or removing light at specific wavelengths. The physical processes that control the temperature of the hydrogen gas vary with time resulting in the fact that at some redshifts the gas could be detected in emission and at others in absorption.

  Observations at meter wavelengths are complicated by the emission from the Milky Way galaxy and also by radio and cell phone emissions on earth. This foreground contamination by our galaxy has to be removed from the data before the cosmic background radiation and associated small distortions can be detected.

  A large research effort is going into the design and construction of radio telescopes to detect this faint glow of light from the dark ages. This presents a huge technical challenge but the rewards will be commensurate with the effort. How then can we hope to observe the dark ages, the period between recombination and reionization?

  How to Detect Atomic Hydrogen

  The most common element in the universe, hydrogen, consists of one proton and one electron. For our present purposes we can think of the atom as a very small solar system with one planet (the electron) orbiting the Sun (proton). In the atom, the electron is only allowed to be in certain orbits or energy states as we call them. The electron can move from one orbit to the next be emitting or absorbing light of known color (wavelength) and energy. Like a planet that spins on its axis, the electron also has spin, but it can only have one of two spin values in the atom, spin up or spin down (Fig. 8.1). The electron can emit light by changing its spin from up to down and absorb light by changing its spin from down to up. The wavelength of light that is absorbed or emitted during a spin change is 21-cm corresponding to a frequency of 1,428 MHz. This wavelength is located in the radio part of the electromagnetic spectrum. This transition was predicted to be observable by the Dutch astrophysicist Henk van de Hulst. The 21-cm emission from our galaxy was first detected by Ewen and Purcell at Harvard in 1951. Neutral hydrogen gas thus reveals itself to us through the emission of radio waves. Light emitted by neutral hydrogen gas in the early universe will be redshifted to meter wavelengths by the time it reaches us.

  Fig. 8.1Transitions of the hydrogen atom. The 21-cm transition is between two states of slightly different energy where the spin of the electron (e) is aligned with that of the proton (p) and where the spins are opposed. A spin flip of the electron results in the emission of a photon with wavelength 21-cm. The transition between electron orbits (energy levels 1 and 2) results in the emission of a much more energetic photon in the ultraviolet part of the spectrum with wavelength 1,216 �
� or 1. 216 ×10 − 5 cm

  Our current calculations and WMAP observations suggest that reionization began at a redshift of about eleven and was over by redshift seven. For 21-cm these two redshifts correspond to observed wavelengths of about 2.5 and 1.7 m respectively. The Square Kilometer Array telescope in Australia is designed to detect neutral hydrogen at these redshifts.

  Observing the Spectrum of the Cosmic Dawn

  We have specified the redshift range of interest, and thus the wavelength range to which the 21-cm light will be redshifted. But what do we expect the signal to look like? To answer this question we must examine the physical picture of the universe during the dark ages. The universe at this time consists of a gas of mostly hydrogen and some helium. Wet also have the redshifted cosmic background radiation as well as dark matter assembling into clumps (or halos) of various masses.

  The hydrogen gas modifies the spectrum of the cosmic background radiation in ways that should be detectable with modern radio telescope technology. The specific frequencies at which radio emission peaks and troughs should occur are predicted by our theories. The detection of such features in the spectrum of the cosmic background radiation would place strong constraints on the nature of star and galaxy formation in the early universe.

  It is conceivable that there is no visible effect. This would mean that on average the hydrogen gas absorbs as many photons from the background radiation as it emits. The second possibility is that the gas is at a hotter temperature than the background radiation. In that case the gas would produce an excess of 21-cm radiation over what is already present from the cosmic background. The third option is that there are too many electrons in the low energy state (spin down). In that case the hydrogen gas would absorb a net amount of energy from the background radiation and we see an absorption feature in the spectrum. Which of these is the correct option? According to theoretical calculations, it depends on the wavelength that we use to carry out the observations.

  The effects are illustrated in Fig. 8.2. When the blue line is above zero we expect the gas to be seen in emission, when the line is below zero the gas absorbs more light than it emits and photons are removed from the background.

  Fig. 8.2The blue line shows the average spin-flip signal in a simple model of the dark ages and the reionization era. The horizontal axis gives the observed frequency. Because we are talking about the 21-cm line, the observed frequency is a measure of the redshift at which the 21-cm radiation was emitted. The brightness temperature plotted on the vertical axis is a measure of the excess or deficit of light measured relative to the cosmic background radiation intensity. We labeled four points towards the end of the dark ages. Stellar ignition marks the point at which we expect the first stars to appear. The second point around redshift twenty marks the period when the first black holes begin to heat the hydrogen gas. The hot bubble epoch marks the beginning of reionization and we see the end of reionization when all the hydrogen gas is ionized (Credit: Burns et al. Advances in Space Research, 49, 18 2012, reprinted with permission by Elsevier)

  When the universe is younger than 10 million years old we expect no visible effect on the background radiation because the gas has the same temperature as the cosmic background radiation. At later times the gas cools faster than the radiation as the universe expands. The result of this is that the gas is seen in absorption. Eventually the gas dilutes so much that it no longer produces detectable effects on the radiation. After the first stars form, they emit ultraviolet radiation which heats up the hydrogen gas, with the effect that the gas emits an excess of photons and can be seen in emission down to redshifts of about ten. By redshift six the universe is almost completely reionized. This puts an end to the absorption and emission of 21-cm radiation by hydrogen atoms since there are very few neutral atoms left.

  We have to look through the Milky Way galaxy in order to see what is happening at high redshifts since our solar system is located inside the Milky Way. Our galaxy is a strong emitter of radio waves at all wavelengths (Fig. 8.3). In fact the Milky Way galaxy is 1,000 times brighter than the radio signal we are trying to detect. To have any chance of detecting the signature of the high redshift hydrogen emission we have to very accurately model and subtract out the Milky Way emission. Techniques are being developed to subtract the foreground emission based on the fact that it varies smoothly with wavelength and has a different spatial distribution to the redshifted neutral hydrogen emission.

  Fig. 8.3Radio map of the sky at 100 MHz corresponding to redshift 13 for the 21-cm wavelength of hydrogen or about 350 million years after the Big Bang. The emission seen in this map comes mostly from the Milky Way galaxy. The plane of the edge on disk of our galaxy is seen as the horizontal feature in the center of this image. The emission from our galaxy has to be modeled and subtracted with very high precision if we can hope to see the redshifted 21-cm emission from the dark ages. The white lines represent the part of the sky that a radio telescope orbiting in space would look at credit (Credit: Burns et al., Advances in Space Research, 49, 18 2012; derived from a model by de Oliveira-Costa and collaborators, reprinted with permission by Elsevier)

  Images of the Dark Ages

  Throughout the dark ages the neutral hydrogen is following the dark matter and getting more clustered with time. We do not have observational evidence of this. Images of the neutral hydrogen gas over a range of redshifts would map the evolution of dark matter during the dark ages. The hydrogen gas distribution yields direct information on the dark matter density variations that were generated a very small fraction of a second after the Big Bang occurred.

  At redshifts smaller than about 25, the neutral hydrogen observations reveal something else about our universe. We believe that around this redshift the first stars and galaxies form and the universe starts to get reionized. The ultraviolet light emitted by the young stars and galaxies forms holes of ionized gas in the neutral hydrogen. The neutral hydrogen distribution starts to look like swiss cheese, at first Gruyère and later Emmental (i.e. the holes get bigger).

  The ‘holes’ in the neutral hydrogen are located around the first stars (Fig. 8.4). A hole is not an absence of gas but an absence of neutral gas (atoms); the gas is present in the form of free electrons and protons. The holes or bubbles near the end of reionization are believed to be about 100 million light years in size. Such bubbles would appear to us on the sky to be about the size of the full moon (as we see it from earth). After reionization is complete, neutral hydrogen is only present inside galaxies. The dark matter distribution can then be traced by the clustering of galaxies where each galaxy is treated as a point (see Chap. 5).

  Fig. 8.4Some key points about the reionization history of the universe are illustrated above. The age of the universe in years is shown on the top horizontal axis and the corresponding redshift is shown on the bottom ( is the present day). The white areas represent parts of the universe that are ionized and the gray region illustrates the part that is neutral. The small white circles represent the ionization regions around the first stars. These stars form in dark matter halos that have about 100,000 times the mass of the Sun. The 21-cm line observations will probe the era of the universe represented by the grey background (Credit: based on a figure by R. Barkana and A. Loeb)

  To recap; prior to a redshift of 25, the neutral hydrogen traces the dark matter distribution. Using this technique we can potentially map a much larger volume of space than the cosmic background radiation snapshot does. We can map the dark matter fluctuations down to much smaller sizes than those measured by the cosmic background. Between a redshift of 25 and a redshift of 6, ionized hydrogen bubbles of increasing size form until the bubbles merge and the neutral hydrogen between galaxies is gone. At redshifts less than six the neutral hydrogen maps the location of galaxies which are themselves tracers of the underlying dark matter distribution. The potential wealth of redshifted 21-cm observations has motivated radio astronomers to build meter-wave radio telescopes suited to high redsh
ift neutral hydrogen studies.

  Observing the Dark Ages with Radio Telescopes

  One goal is to collect enough light to get a spectrum of the redshifted 21-cm sky. Spectroscopy is in this sense easier than imaging since we do not need to map the sky. We expect the light to be coming from all directions (like the cosmic background radiation) so a large part of the sky can be studied at one go. One example of this approach is the Experiment to Detect the Global EOR Signature (EDGES). The EOR in the acronym stands for End of Reionization. This experiment is deployed at the Murchison Radio Astronomy Observatory in Western Australia to measure the radio spectrum between 100 MHz (3 m) and 200 MHz (1.5 m). The design is simple; an antenna, an amplifier, and a computer are connected to a solar energy source. The EDGES experiment has produced the best broad band radio spectrum of the sky and has ruled out instantaneous reionization.

  Proposals havet also been made to carry out similar measurements from space. One could place a satellite in orbit around the Moon and make measurements from the far side of the Moon. This method eliminates the most intense man-made foreground emission such as radio transmissions from radar, radio and TV stations.