Abstract
The conventional mechanism for primordial gravitational wave (GW) production—the amplification of vacuum metric fluctuations—uniquely links the tensor-to-scalar ratio ($r$) to the inflationary expansion rate ($H$). A detection of $r \sim 0.1$ is typically interpreted as evidence for an energy scale of $\sim 10^{14}$ GeV. This paper explores a covariant model where a "spectator field" $\sigma$ with a low sound speed ($c_s \ll 1$) acts as a source for GWs, potentially breaking this direct link. While the model finds that $r$ is not enhanced relative to vacuum results due to scalar perturbation constraints, the observable signal becomes highly sensitive to $c_s$, complicating the interpretation of future CMB data.
Introduction: The Standard Paradigm
In the standard single-field model of inflation, primordial gravitational waves arise from vacuum fluctuations in the spacetime metric. This leads to the famous Lyth relation, which provides a direct window into the energy scale of the early universe. If we observe a high value of $r$, we conclude that inflation occurred at a high energy scale. However, this conclusion is only robust if the vacuum is the dominant source of gravitational waves. This paper investigates whether an additional field, which does not drive inflation, could be responsible for the observed signal.
The Spectator Field Mechanism
The authors construct a rigorous covariant formulation of a spectator field $\sigma$ that interacts only gravitationally with the inflaton. Key technical components of this model include:
- A $P(X)$ Lagrangian: This allows the spectator field to have a sound speed $c_s$ significantly smaller than the speed of light ($c_s \ll 1$).
- Non-minimal Coupling: To prevent the field's energy density from diluting as the universe expands (like $a^{-3}$), the authors introduce a derivative coupling between the spectator field and the Einstein tensor ($G^{\mu\nu}$).
Fluctuations in this spectator field source both tensor (GW) and scalar (density) perturbations at second order through non-linear interactions.
The Role of Sound Speed
The "aha moment" of this research is the role of the sound speed $c_s$. A small sound speed acts as a powerful amplifier for non-linear production. The authors derived analytical expressions showing that the sourced power spectra scale with high inverse powers of $c_s$:
- Tensor Power Spectrum: $P_\lambda^{(s)} \propto H^4 / c_s^{6.2}$
- Scalar Power Spectrum: $P_\zeta^{(s)} \propto H^4 / c_s^{10}$
Because these effects scale so strongly with $c_s$, even a spectator field with negligible energy density can produce a significant signal if its sound speed is sufficiently low.
The Scalar Constraint
A critical realization of this work is that one cannot boost gravitational waves without also boosting density fluctuations. Because the scalar power spectrum is precisely measured (normalized to $\sim 2.5 \times 10^{-9}$ by experiments like Planck and BICEP2), it acts as a "speed limit" on the spectator field's contribution.
Since the scalar perturbations are enhanced more aggressively ($c_s^{-10}$) than the tensor perturbations ($c_s^{-6.2}$), the spectator field in this specific covariant realization cannot actually increase the total $r$ beyond the standard vacuum prediction.
Conclusion: A Broken Connection
Even though $r$ is not enhanced, the research proves that the direct $r \leftrightarrow H$ connection is broken. The value of $r$ becomes a degenerate function of both the expansion rate $H$ and the spectator field's sound speed $c_s$. This means a detection of $r \sim 0.1$ does not automatically guarantee a specific inflationary energy scale.
Furthermore, sourced perturbations are inherently non-Gaussian. Future analysis of the CMB bispectrum will be the definitive way to distinguish between vacuum-produced gravitational waves and those sourced by spectator fields, potentially ruling out or supporting this alternative mechanism.
Summary updated