Cosmic Microwave Background

Every direction you look, the universe is glowing. Not with starlight or galaxy light, but with a faint, nearly uniform bath of microwave radiation that fills all of space. This is the cosmic microwave background (CMB), the thermal afterglow of the Big Bang itself, released 380,000 years after the universe began when the primordial plasma cooled enough for electrons to combine with protons and form neutral hydrogen. At that moment, photons that had been trapped in the opaque plasma were suddenly free to travel unimpeded through space. They have been traveling ever since, redshifted by the expansion of the universe from their original temperature of roughly 3,000 Kelvin to the 2.725 Kelvin we measure today. The CMB is the oldest light we can ever observe, a snapshot of the universe when it was 0.003% of its current age, and the single most informative dataset in all of cosmology.

Prediction and Discovery

George Gamow, Ralph Alpher, and Robert Herman predicted the existence of a cosmic thermal background in the late 1940s as a consequence of their hot Big Bang nucleosynthesis model. If the early universe was hot and dense enough to fuse hydrogen into helium (which the observed helium abundance of roughly 25% by mass requires), then it must have been filled with thermal radiation that would persist as a cooled remnant. Alpher and Herman estimated a present-day temperature of roughly 5 Kelvin, remarkably close to the actual value given the crude cosmological parameters available to them.

The prediction was largely forgotten by the astronomical community for over a decade. In 1964, Robert Dicke and Jim Peebles at Princeton independently re-derived the prediction and began building a detector to search for it. Before they could complete their instrument, Arno Penzias and Robert Wilson at Bell Labs in Holmdel, New Jersey, accidentally discovered the signal.

Penzias and Wilson were using a 6-meter horn antenna (originally built for satellite communication) to map radio emission from the Milky Way. They found a persistent, isotropic noise signal at a wavelength of 7.35 centimeters that they could not eliminate. They cleaned pigeon droppings from the antenna, checked for urban radio interference, and tested every component. The signal remained: a uniform 3.5 Kelvin thermal background coming from every direction in the sky.

A phone call connected Penzias and Wilson to Dicke's group at Princeton. The two papers were published back-to-back in the Astrophysical Journal in 1965: Dicke et al. providing the theoretical interpretation, Penzias and Wilson reporting the measurement. The discovery confirmed the Big Bang model and effectively ended the competing Steady State theory, which predicted no such background. Penzias and Wilson received the 1978 Nobel Prize in Physics.

COBE: Seeing the Seeds of Structure

The CMB is extraordinarily uniform: its temperature is the same in every direction to one part in 100,000. This uniformity is itself a profound cosmological constraint, implying that the observable universe was in thermal equilibrium at early times (a condition explained by inflationary theory, which posits a brief period of exponential expansion in the first fraction of a second).

But the CMB is not perfectly uniform. Tiny temperature fluctuations, at the level of roughly 30 microkelvin (one part in 100,000), encode the density variations in the primordial plasma that would eventually grow, under gravity, into galaxies, galaxy clusters, and the large-scale structure of the cosmic web.

NASA's Cosmic Background Explorer (COBE), launched in 1989, made two transformative measurements. The FIRAS instrument measured the CMB spectrum with extraordinary precision, confirming that it is a nearly perfect blackbody with a temperature of 2.725 +/- 0.002 Kelvin. The agreement between the measured spectrum and a theoretical blackbody curve is so precise that it is the most perfect blackbody ever observed in nature, a result that definitively confirmed the thermal origin of the CMB.

The DMR instrument detected the temperature anisotropies for the first time, producing the first map of CMB fluctuations. The map was crude (7-degree angular resolution), but the detection of primordial fluctuations was epochal. George Smoot, presenting the results in 1992, described them as "like looking at the face of God." Stephen Hawking called it "the greatest discovery of the century, if not of all time." John Mather and George Smoot shared the 2006 Nobel Prize in Physics for the COBE results.

WMAP: Precision Cosmology

NASA's Wilkinson Microwave Anisotropy Probe (WMAP), launched in 2001 and operational through 2010, mapped the CMB with 45 times better resolution and far greater sensitivity than COBE. The WMAP maps revealed the detailed pattern of acoustic oscillations in the CMB: a series of peaks in the angular power spectrum that encode the physics of the primordial plasma with extraordinary precision.

The physics is elegant. Before recombination, the universe was a plasma of protons, electrons, and photons. Photon pressure and gravity competed: gravity pulled matter into overdense regions, while photon pressure resisted compression. The result was acoustic oscillations, sound waves propagating through the plasma at roughly 57% of the speed of light. When recombination occurred and photons decoupled from matter, the oscillation pattern was frozen into the CMB as a characteristic pattern of hot and cold spots at specific angular scales.

The angular scale of the first acoustic peak (roughly 1 degree) depends on the geometry of the universe (flat, open, or closed). WMAP confirmed that the universe is spatially flat to within 0.4%, consistent with the prediction of inflationary theory. The relative heights of successive peaks constrain the baryon density (ordinary matter: 4.6%), the dark matter density (23%), and the dark energy density (72%). The overall pattern constrains the age of the universe (13.77 billion years), the Hubble constant (70.0 km/s/Mpc from WMAP), and the spectral index of primordial fluctuations (slightly less than 1, consistent with the simplest inflationary models).

WMAP established the era of "precision cosmology": the determination of cosmological parameters to percent-level accuracy from a single dataset. The six-parameter Lambda-CDM model (which describes a universe dominated by dark energy and cold dark matter, with spatial flatness and nearly scale-invariant primordial fluctuations) fits the WMAP data with remarkable fidelity.

Planck: The Definitive Map

ESA's Planck satellite (2009-2013) produced the definitive CMB maps, with angular resolution of 5 arcminutes (12 times better than WMAP) across nine frequency bands spanning 30 to 857 GHz. The multi-frequency coverage enabled precise separation of the CMB signal from foreground contamination (galactic dust emission, synchrotron radiation, free-free emission, and the Sunyaev-Zel'dovich effect from galaxy clusters).

Planck's results refined the cosmological parameters to sub-percent precision. The universe is 13.799 +/- 0.021 billion years old. Ordinary matter constitutes 4.9% of the total energy density, dark matter 26.8%, and dark energy 68.3%. The Hubble constant from Planck is 67.4 +/- 0.5 km/s/Mpc, a value that is in significant tension with local measurements (73-74 km/s/Mpc from Cepheid-calibrated Type Ia supernovae), a discrepancy known as the Hubble tension that may indicate new physics beyond the standard model.

Planck detected the gravitational lensing of the CMB by intervening large-scale structure, providing an independent measurement of the matter distribution between us and the surface of last scattering. It also measured the CMB polarization with sufficient precision to constrain the optical depth to reionization (the fraction of CMB photons that have been re-scattered by ionized hydrogen in the intergalactic medium), providing information about when the first stars and galaxies turned on.

Polarization and the Search for Inflation

The CMB is polarized: the microwave photons have preferred orientations of their electric field vectors, imprinted by the scattering process at recombination. This polarization contains information beyond what the temperature fluctuations alone can provide.

CMB polarization is decomposed into two patterns: E-modes (curl-free, produced by density fluctuations and detected by multiple experiments) and B-modes (divergence-free). B-modes have two sources: gravitational lensing of E-modes by large-scale structure (detected by Planck and other experiments), and primordial gravitational waves generated during cosmic inflation.

The detection of primordial B-modes would be direct evidence for inflation and would constrain the energy scale at which inflation occurred. The signal is expected to be extremely faint (the tensor-to-scalar ratio r, which characterizes the amplitude, is constrained to be less than approximately 0.036 by Planck and BICEP/Keck data).

The BICEP/Keck series of experiments at the South Pole target this signal. In 2014, BICEP2 announced a detection of B-modes that was initially interpreted as evidence for inflation, but subsequent analysis (incorporating Planck dust maps) showed that the signal was likely dominated by polarized emission from galactic dust. The current upper limit from BICEP3/Keck is the tightest constraint on r, but the primordial signal has not been detected.

Future experiments targeting primordial B-modes include the Simons Observatory (ground-based, Chile), CMB-S4 (a comprehensive ground-based CMB experiment combining multiple observatories), and JAXA's LiteBIRD satellite. A definitive detection of primordial B-modes would be one of the most important discoveries in physics, providing direct observational evidence for quantum gravity effects during the first 10^-36 seconds of the universe.

What the CMB Tells Us

The CMB encodes more cosmological information per pixel than any other observable. From a single map of temperature and polarization fluctuations, cosmologists have extracted the age, geometry, composition, expansion rate, and statistical properties of the initial conditions of the universe. The consistency of the Lambda-CDM model with CMB data, combined with independent constraints from baryon acoustic oscillations, galaxy clustering, and Type Ia supernovae, represents one of the greatest achievements in the history of science: a quantitative, predictive model of the universe's origin and evolution that is confirmed by multiple independent lines of evidence.

The CMB is also a reminder of what we do not understand. The dark energy that dominates the energy budget has no accepted physical explanation. The dark matter that provides the gravitational scaffolding for structure formation has not been directly detected in a laboratory. The inflationary epoch that explains the CMB's uniformity and the origin of its fluctuations has not been confirmed by the detection of primordial gravitational waves. And the Hubble tension between CMB-derived and locally measured expansion rates may be pointing to physics that the standard model does not accommodate.

The oldest light in the universe is still teaching us.