Probing Cosmic Inflation: New Constraints on Primordial Gravitational Waves from CMB Polarization Data

The CMB is the thermal afterglow of the Big Bang, a fossil light that provides a snapshot of the infant universe. This light is faintly polarized, primarily due to scattering off electrons during the era of recombination. The polarization patterns on the sky can be decomposed into two distinct components: curl-free E-modes, which are sourced by scalar density perturbations, and divergence-free B-modes. While gravitational lensing of E-modes can generate B-modes on small angular scales, a primordial B-mode signal on large angular scales (~$>1^\circ$) is a unique and sought-after signature of primordial gravitational waves from the inflationary epoch.

This analysis combines publicly available, full-sky data from the Planck satellite with new, high-sensitivity, small-sky observations from our ground-based observatory. The primary challenge in any B-mode search is the contamination from polarized astrophysical foregrounds, principally synchrotron radiation and thermal dust emission from within our own Milky Way. To isolate the faint cosmological signal, we employed a sophisticated component separation method. This technique utilizes the different frequency dependence of the CMB and foreground signals across multiple observation bands to meticulously model and subtract the galactic contamination, producing the cleanest CMB polarization maps to date for the observed sky region.

The amplitude of the primordial gravitational wave background is parameterized by the tensor-to-scalar ratio, 'r'. This value represents the power in primordial gravitational waves (tensor perturbations) relative to the power in density fluctuations (scalar perturbations) that seeded the large-scale structure of the universe. A definitive measurement of a non-zero 'r' would constitute a monumental discovery.

We performed a Bayesian likelihood analysis on the power spectrum of our cleaned B-mode maps. Our analysis yields no statistically significant detection of a primordial B-mode signal. This non-detection allows us to place a new upper limit on the tensor-to-scalar ratio of $r < 0.03$ at 95% confidence. This is the most stringent constraint ever placed on 'r' from CMB data.

This new, tighter constraint on 'r' has profound implications for theoretical models of inflation. A large class of the earliest and simplest inflationary models, such as those with a simple quadratic potential, predict values of 'r' that are now definitively ruled out by our data. Our result places significant pressure on the remaining models, strongly favoring those that predict a much lower amplitude of primordial gravitational waves. These "low-scale" models, while still viable, are inherently more complex and predict a signal that will be exceptionally challenging to detect, even for future experiments. While inflation remains the dominant paradigm, our findings have significantly pruned the tree of possible inflationary mechanisms.

We have presented the most stringent upper limit to date on the tensor-to-scalar ratio 'r' by analyzing state-of-the-art CMB polarization data and utilizing advanced foreground mitigation techniques. This result rules out a significant portion of the parameter space for canonical models of cosmic inflation. The search for primordial B-modes has now entered a new regime, requiring the unprecedented sensitivity of next-generation observatories. The upcoming CMB-S4 experiment will be critical in this quest, poised to either finally detect this faint echo from the beginning of time or force a fundamental rethinking of the physics that governed the infant universe.