The first data-driven bounds on the quantum decoherence of inflationary gravitational waves

Abstract

The (large-scale) structures we observe in the Universe are classical, but within the inflationary scenario they do originate from quantum fluctuations. This leads to the question: ''How did this quantum-to-classical transition occur?''. A potential explanation is quantum decoherence due to interactions between different fields present during inflation. The tensor modes (i.e. primordial gravitational waves) can interact with a scalar sector, causing their quantum decoherence to occur and inducing a change in the gravitational wave (GW) background. The power spectrum of these GWs can be constrained using the upper bounds found by Planck, BICEP/Keck Array, LIGO-Virgo-KAGRA, Big Bang Nucleosynthesis, and the Pulsar Timing Array detections. These impose constraints on the interaction between the fields. We find that the observational upper bounds mainly constrain scenarios with a strong interaction, especially if the interaction is also strongly time dependent. Furthermore, we find which observationally allowed scenarios have not completed decoherence by the end of inflation, thus possibly leaving quantum signatures in the GW background. Lastly, we show that, interestingly enough, there are decoherence scenarios corresponding to the signal observed by PTA experiments. This highlights the importance of the quantum decoherence effect on GWs.

Figures (9)

• Figure 1: The 1D and 2D 68% and 95% CL interval constraints on $p, \beta^2\sigma_\gamma$, using PL+BK18+LVK. We set $r_{\ast}=0.00461$ to the value of Starobinsky model of inflation STAROBINSKY198099. Additionally, we set $n_T=0, H_{\ast}l_E = 10^{-3}$, and $\Delta N_{\ast} = 50$. In the left panel, we apply a smoothing, namely smooth_scale_1D = 0.4, and smooth_scale_2D = 0.4. In the right panel, we apply no such smoothing to highlight that the smoothing does not affect the conclusions drawn in this work.
• Figure 2: The 1D and 2D 68% and 95% CL interval constraints on $r_{0.05}, p, \beta^2\sigma_\gamma$, using PL+BK18+LVK. Additionally, we set $n_T=0, H_{\ast}l_E = 10^{-3}$, and $\Delta N_{\ast} = 50$.
• Figure 3: The 1D and 2D 68% and 95% CL interval constraints on $r_{0.05}, p, \beta^2\sigma_\gamma$, using PL+BK18+LVK. We also enforce the decoherence criterion on CMB scales, shown as the black solid line. Additionally, the dashed black line indicates the values of $p$ and $\log(\beta^2\sigma_\gamma)$ where $\delta_{\boldsymbol{k}_{\rm LVK}} =1$, meaning that above this line the scales LVK probes have fully decohered (this is not implemented in this work). Furthermore, we set $n_T=0, H_{\ast}l_E = 10^{-3}$, and $\Delta N_{\ast} = 50$.
• Figure 4: The 1D and 2D 68% and 95% CL interval constraints on $r_{0.05}, p, \beta^2\sigma_\gamma, n_T$, using PL+BK18+LVK. Additionally, we set $H_{\ast}l_E = 10^{-3}$, and $\Delta N_{\ast} = 50$.
• Figure 5: The 1D and 2D 68% and 95% CL interval constraints on $r_{0.05}, p, \beta^2\sigma_\gamma, H_\ast l_E, \Delta N_\ast$. The results are shown using PL+BK18+LVK (grey), PL+BK18+LVK+NANO (red), and PL+BK18+LVK+NANO+BBN (blue). Additionally, we set $n_T=0$. We have negated any smoothing to highlight the peaked scenarios on intermediate scales, allowed by PL+BK18+LVK.