Cosmic Gravitational Waves May Break the CMB's Adiabatic Rule

Imagine a relic hum so faint that no single detector can hear it: a sea of gravitational waves stretched across the universe by inflation. If that cosmological gravitational-wave background exists, its ripples would carry a memory of the first moments of cosmic time. But what kind of memory? For the cosmic microwave background, the answer is adiabatic: the initial fluctuations are consistent with a single common clock, with no independent entropy perturbations separating species. A new preprint asks whether gravitons—the quanta of gravitational waves in this quantum description—in a background generated by quantum fluctuations of spacetime follow that same tidy rule. The authors conclude they do not.

The CMB's tidy start does not automatically transfer

The CMB gives cosmologists a benchmark. Photons were tightly coupled to baryons before recombination, and their initial fluctuations are adiabatic. That means no independent entropy perturbations separated one species from another. For a cosmological gravitational-wave background, however, there is no guarantee. Gravitational waves interact extremely weakly, so their background can be non-thermal, with a spectrum and initial fluctuations that may depend on frequency. If the background comes from the decay of the inflaton—the field thought to drive inflation—then adiabatic initial conditions can be inherited. But if it comes from quantum fluctuations of the metric itself, the logic changes.

Too faint to bend Einstein's equations

The paper, posted to arXiv in July 2023 by Lorenzo Valbusa Dall'Armi and colleagues, focuses on that second case: gravitational waves generated by quantum fluctuations of spacetime during inflation. The authors first note a practical obstacle. The energy density of the cosmological gravitational-wave background is constrained by Big Bang nucleosynthesis and Planck data to be a tiny fraction of the radiation density—well below one percent. Because it is so small, gravitational radiation barely affects Einstein's equations. The usual route—reading the graviton overdensity from the scalar metric perturbations that govern ordinary matter—does not work.

So the team computed the gravitational-wave energy-momentum tensor directly from the gravitational strain, the microscopic ripple in spacetime geometry. They used a short-wave approximation, valid when the waves are much smaller than the scales over which the background changes, and averaged over those small scales. As a consistency check, they compared their result with an independent method based on perturbing the Einstein tensor. Both approaches converge when the waves are well inside the horizon. That let them connect the microscopic strain to a macroscopic energy density and, from there, to the initial perturbation of the graviton distribution.

Two polarizations behave like extra fields

The central result is that adiabatic initial conditions fail for this source. In standard single-field inflation, a single clock—the inflaton—sets the evolution, and the separate-universe argument makes adiabatic modes unavoidable for species produced by inflaton decay. But gravitational waves are tensor perturbations, with two independent polarizations. During inflation, those polarizations behave like two extra fluctuating fields. They are not simply passengers carried by the inflaton's density perturbation. As a result, even in single-field inflation, the cosmological gravitational-wave background can start with non-adiabatic initial conditions.

The initial anisotropy the authors derive contains a monopole from scalar and tensor perturbations, plus a quadrupole sourced only by large-scale tensor perturbations. A monopole is a uniform shift from patch to patch; a quadrupole varies with direction in a characteristic pattern. That quadrupole has no analogue in the CMB. The paper notes it is smaller than the monopole, but it could become relevant for future detectors with high sensitivity.

An order-of-magnitude boost

The practical consequence is large. Compared with the standard adiabatic case, these initial conditions enhance the total angular power spectrum—a measure of how the signal varies across angular scales on the sky—of the cosmological gravitational-wave background by about an order of magnitude. In the limiting case of a scale-invariant background, where the tensor tilt is zero, the paper reports an enhancement factor near 10. The exact size depends on the tensor tilt, which describes how the background's energy is distributed across frequencies.

The enhancement is not just a number. It changes how different contributions to the angular power spectrum combine. In the adiabatic case, the initial condition and the early integrated Sachs-Wolfe effect have the same sign and add coherently at small angular scales, creating a bump. With the new single-field initial conditions, they have opposite signs, so the spectrum stays flatter as angular scale decreases. The authors also find that the angular power spectrum becomes more sensitive to extra relativistic degrees of freedom—additional light particles—and to the equation of state of the universe at early times. In other words, the same measurement could test both inflation and hidden radiation.

The new initial conditions also alter the correlation between the gravitational-wave background and the CMB. For adiabatic initial conditions, the large-scale correlation is about 0.98. For single-field inflation, it drops to roughly 0.8 to 0.9. Measuring that correlation, if a cosmological background is ever mapped, could help distinguish between initial-condition scenarios.

A calculation, not a detection

This work is theoretical. It does not report a detection of the cosmological gravitational-wave background. Pulsar timing arrays have reported a stochastic gravitational-wave background, but its origin—astrophysical or cosmological—is not settled. The paper's calculations assume that the cosmological background comes from quantum fluctuations of the metric during inflation. If the background is instead produced by inflaton decay, adiabatic initial conditions may still apply. The enhancement depends on the tensor tilt and on assumptions about the early universe, and its observability will depend on the sensitivity of future detectors and their ability to separate signal from noise.

Future space-based and ground-based observatories such as LISA, the Einstein Telescope, and Cosmic Explorer might be able to detect the cosmological background. If they succeed, the angular pattern of the background could become a new probe of the early universe. The authors also suggest that their approach could apply to other decoupled particles that fluctuate during inflation, widening its reach beyond gravitons.

For now, the result is a warning about assumptions. The cosmic microwave background taught cosmologists to expect adiabatic beginnings. The gravitational-wave background may not have listened. If future detectors can read its faint anisotropies, they might find a different initial chapter of the universe—one written not only by the inflaton, but by the two polarizations of spacetime itself.