Dust, Distortions, and Shadows in the Universe’s Oldest Light
Fifty years ago, two radio astronomers working at Bell Labs in Holmdel, New Jersey, stumbled across a persistent unknown source of noise when they began taking measurements with a new horn antenna. The extremely sensitive apparatus was intended to receive radio waves from communications satellites but instead had received a signal from nearly the dawn of time: the cosmic microwave background (CMB) radiation, the thermal afterglow of the big bang. Arno Penzias and Robert Wilson later received the Nobel Prize for their discovery. Further characterization of the perfect blackbody spectrum of the CMB by NASA’s COBE satellite led to other Nobels in 2006.
The quest to pry every last secret from this radiation is a story that continues to this day, one that I am very humbled to take part in.
A window into the universe’s birth
The CMB has ancient origins; its photons are the oldest light ever seen in the universe. They were produced in the primordial brew of the big bang. For hundreds of thousands of years, they scattered frequently in a dense fog of electrons, protons, and helium nuclei that filled the universe. As the universe expanded, the fog cooled to progressively lower temperatures. Eventually, about 380,000 years after the big bang, the temperature was low enough for electrons and ions to combine. They formed neutral hydrogen and helium atoms, which no longer scattered the photons. The fog cleared.
Since that time, the photons have traveled, largely unimpeded. Their wavelengths stretched with the expansion of the universe, and they eventually arrived at our telescopes in the microwave band of the electromagnetic spectrum.
These CMB photons provide a window onto the conditions of the early universe, and thus a powerful tool with which to determine its fundamental properties, including its age, composition, geometry, and perhaps even its origin.
The prevailing theory describing the early universe, inflation, posits that the universe underwent a period of extremely violent expansion at its very beginning, growing in size by some 26 orders of magnitude in only 10-33 seconds—a truly audacious idea. Crucially, the theory makes specific predictions for CMB photons. It predicts the almost perfect uniformity of the photons’ temperature observed across the sky—2.726 Kelvin—as well as the ways in which temperature should slightly deviate from this uniformity. These temperature deviations, which are on the order of 1 part in 100,000, correspond to the differences in the density of the universe from place to place at the time the CMB photons were emitted. Most audaciously, inflation states that these small differences originated in quantum fluctuations that were stretched to macroscopic sizes during the initial violent expansion.
However, alternative theories, such as a cyclic or “bouncing” universe, might reasonably predict the properties of these temperature deviations as well. A key distinction of inflation is the prediction of a particular pattern in the CMB photon’s polarization, the direction of the light’s electric and magnetic fields. Inflationary expansion is thought to have been so violent that it disturbed the fabric of space-time itself, producing gravitational waves. These waves later manifest as “swirly” patterns in the polarization of the CMB photons (technically known as B-modes). The cyclic universe model does not predict this pattern, setting the stage for a powerful experimental test of our ideas about the origin of the universe.
Promising polarization signal bites the dust
In March 2014, the team behind the BICEP2 experiment dramatically announced a measurement of this highly sought B-mode polarization signal, an announcement met with international excitement. The pattern seemed to match theoretical expectations, but—very importantly—had been seen clearly at only one frequency, 150 gigahertz.
Dust grains in the Milky Way are known to emit thermal radiation around this frequency as they are heated by starlight. The grains are oriented by the galactic magnetic field, which leads to polarization in the emitted thermal radiation that could mimic a B-mode signal. But the strength of the polarized signal from dust was mostly unknown until the past two years, when high-frequency data from the Planck satellite began to shed light on its properties.
Using Planck data that had been publicly released, as well as other existing galactic surveys, I worked with a group at Princeton University (which included Raphael Flauger and David Spergel) in the months following the BICEP2 announcement, trying to understand whether the observed BICEP2 B-mode signal could be entirely explained by dust. The answer—unfortunately—turned out to be “yes,” although the uncertainties were large. Our reinterpretation was subsequently confirmed by the official joint analysis of the BICEP2 and Planck data released in February 2015. It showed that no statistically significant evidence for primordial B-modes remained after correcting for the dust.
Despite this disappointment, the path forward is clear. We need to understand the dust contaminating our measurements of the sky at microwave frequencies. A number of experiments have been planned or are already underway with this goal in mind, including the Advanced Atacama Cosmology Telescope (AdvACT), the Simons Array (with POLARBEAR-2 detectors), and BICEP3. By exploiting the different frequency spectra of the CMB and the dust (recall the perfect blackbody nature of the CMB), cosmologists might be able to effectively separate the two signals.
Cosmic trash or treasures?
But the dust in our galaxy is not the only thing obscuring our understanding of the universe’s birth. As CMB photons travel to our telescopes, they sometimes encounter distortions along the way that can warp our perspective. Fortunately, there is a silver lining. The signals produced by these encounters contain a wealth of information about the cosmos.
For example, the path of a CMB photon can be bent by gravitational fields encountered during its journey; the effect is called gravitational lensing. The twisting of light due to this lensing effect induces spurious B-modes in the CMB polarization, which must be accounted for (just like the dust B-modes) when searching for the primordial B-modes due to inflation. However, one can also use this gravitational lensing signal to reconstruct the gravitational fields themselves, and hence the distribution of matter (including both atomic matter and dark matter). This powerful technique has recently come of age and is now yielding precise constraints on the large-scale structure of the universe. Gravitational lensing maps from upcoming experiments such as AdvACT will unveil the precise distribution of dark matter at high resolution over more than half of the sky.
In addition to lensing distortions, CMB photons sometimes encounter large clouds of hot, ionized gas as they travel through the universe. The clouds cast “shadows” in our observed CMB maps because CMB photons scatter off of free electrons in the gas, a process known as the Sunyaev-Zel’dovich (SZ) effect. Because the SZ effect leaves the photons’ polarization essentially unchanged, it does not contaminate B-mode searches in the way that dust and gravitational lensing do. Instead, SZ shadows in CMB temperature maps are helpful because they can be used to find the most massive structures in the universe: galaxy clusters, where most of the hot, ionized gas causing the shadows is located. These rare structures are very sensitive probes of the amount of dark matter in the universe, for instance, and the properties of the quantum fluctuations generated during inflation.
Much of my own work has focused on new ways to pry secrets from the CMB using SZ shadows or the combination of gravitational lensing and SZ effects.
Fifty years after its discovery, the CMB continues to yield unexpected surprises. It may soon confirm our best ideas about what happened during the first moments after the universe was born. However, achieving this goal will require understanding the dust, distortions, and shadows present in CMB maps. It is an exciting challenge.
