Cosmic inflation

Inflation was introduced to solve three major puzzles of the standard Big Bang model: the horizon problem, the flatness problem, and the magnetic monopole problem.¹ The horizon problem arises because regions of the cosmic microwave background (CMB) that are now far apart were never in causal contact, yet they have nearly identical temperatures. Inflation solves this by positing that these regions were in causal contact before inflation, then were stretched apart. The flatness problem concerns the observed near-flat geometry of the universe; inflation drives the geometry toward flatness regardless of initial conditions. The monopole problem is that grand unified theories predict an overabundance of magnetic monopoles, which inflation dilutes to undetectable levels.

Inflation is typically modeled by a scalar field, the inflaton, with a potential energy density that dominates the universe's energy budget.² During inflation, the field slowly rolls down its potential, causing the universe to expand exponentially. When the field reaches a steeper part of the potential, it oscillates and decays, reheating the universe with particles and radiation, marking the end of inflation. The exponential expansion stretches any initial curvature to near-zero, explaining the observed flatness. Quantum fluctuations in the inflaton field are stretched to cosmological scales, seeding the large-scale structure of the universe and leaving imprints in the CMB as nearly scale-invariant, Gaussian, adiabatic perturbations.

The strongest evidence for inflation comes from observations of the CMB, particularly by the Planck satellite. The CMB temperature fluctuations are consistent with the nearly scale-invariant spectrum predicted by inflation, with a spectral index n_s ≈ 0.965. The observed flatness of the universe (Ω_K ≈ 0.001) also matches inflation's prediction. Additionally, the absence of certain topological defects, such as cosmic strings, is consistent with inflation. However, the detection of primordial gravitational waves, a key prediction of many inflation models, has not been confirmed. The BICEP2 announcement in 2014 of B-mode polarization was later attributed to dust, not gravitational waves.

Many variants of inflation exist, including chaotic inflation, natural inflation, and hybrid inflation, each with different inflaton potentials.³ Some models embed inflation in string theory, such as brane inflation. Inflation is not universally accepted; alternatives include the ekpyrotic model and cyclic universe scenarios. The initial singularity problem remains: inflation does not explain what preceded it. Some theorists explore eternal inflation, where inflation never ends globally, producing a multiverse. This idea has implications for fine-tuning and the anthropic principle, but it is highly speculative and not directly testable.

Inflation was anticipated by earlier work: Alexei Starobinsky proposed a similar model in 1979, and Willem de Sitter's solution to Einstein's equations describes exponential expansion.⁴ The term 'inflation' was coined by Guth, but the first working model was developed by Andrei Linde and independently by Andreas Albrecht and Paul Steinhardt in 1982. Inflation also predicts that the universe is extremely close to flat, but not exactly flat; the current density parameter Ω is measured to be 1.00 ± 0.01. The inflaton field has not been identified, and its nature remains one of the biggest mysteries in physics. Some models predict a slight non-Gaussianity in the CMB, which future experiments may detect.

Inflaton

A hypothetical scalar field that drives cosmic inflation.

Horizon problem

The puzzle of why distant regions of the universe have the same temperature despite being causally disconnected.

Flatness problem

The puzzle of why the universe's spatial curvature is so close to zero.

B-mode polarization

A pattern in the polarization of the CMB that could be produced by primordial gravitational waves.