Scalars at the Cosmological Collider: Full Shapes of Tree Diagrams and Bispectrum Searches using Planck Data

This paper provides a unified derivation of the full bispectrum shapes for single, double, and triple exchange processes involving massive scalars in the Cosmological Collider framework and reports a null result from Planck data, while identifying a 1.5$\sigma$ hint of non-Gaussianity in an extended scenario featuring a scalar chemical potential that allows for on-shell production of super-Hubble mass particles.

The Gist

Imagine the universe as a giant, cosmic concert hall. For decades, physicists have been trying to hear the music played by the very first moments of the Big Bang. This "music" is hidden in the Cosmic Microwave Background (CMB)—the faint afterglow of the universe's birth, which we can still detect today.

This paper is about a new way to listen to that music to find out if there are any "heavy instruments" (massive particles) playing in the band that we can't see with our current telescopes.

Here is the breakdown of the paper's story, using simple analogies:

1. The "Cosmological Collider"

Think of the early universe during inflation (a period of super-fast expansion) as the most powerful particle accelerator in existence.

• Earth's Colliders: We have machines like the Large Hadron Collider (LHC) that smash particles together to see what's inside. But they are limited by how much energy we can put into them.
• The Cosmic Collider: The early universe had energy levels trillions of times higher than anything we can build on Earth. If there were heavy, exotic particles floating around back then, they would have been created and then decayed, leaving a unique "fingerprint" on the universe's structure.

2. The "Fingerprint": The Bispectrum

When these heavy particles decay, they don't just leave a simple mark. They create a specific pattern of ripples in the density of the universe.

• The Analogy: Imagine dropping three stones into a pond. If the water is calm, the ripples spread out evenly. But if there's a hidden underwater rock (a heavy particle), the ripples will interfere with each other in a weird, wavy pattern.
• The Bispectrum: Physicists call this three-way interaction a "bispectrum." It's a mathematical way of measuring how three different points in the sky are connected. If the pattern has a specific oscillating (wavy) rhythm, it's a "smoking gun" that a heavy particle was there.

3. The Three Ways to Listen (The Diagrams)

The authors looked at three different ways these heavy particles could have interacted with the universe's expansion field (the inflaton). They used a fancy name for these interactions: Single, Double, and Triple Exchange.

• Single Exchange: Imagine two people passing a ball once.
• Double Exchange: They pass the ball back and forth once.
• Triple Exchange: They pass it back and forth twice.

The paper is special because, for the first time, they calculated the entire shape of the sound wave for all three scenarios, not just the part where the signal is loudest. Previously, scientists only looked at the "squeezed" part of the wave (where one ripple is tiny and two are big), but this paper mapped the whole ocean.

4. The "Chemical Potential" Boost

There was a problem: If the heavy particles were too heavy, their signal would be so faint (like a whisper in a hurricane) that we couldn't hear it. The universe naturally suppresses these heavy signals.

• The Solution: The authors looked at a mechanism called "Scalar Chemical Potential."
• The Analogy: Imagine trying to hear a quiet violin in a noisy room. Usually, you can't. But what if someone gave the violinist a megaphone? The "Chemical Potential" acts like that megaphone. It injects extra energy into the system, allowing those super-heavy particles to be created and leave a loud, clear signal even if they are much heavier than the universe's expansion rate.

5. The Search Results (Did we find anything?)

The team took their new, detailed "sheet music" (theoretical shapes) and compared it against the actual data from the Planck satellite, which mapped the CMB.

• The Bad News: They didn't find a definitive discovery. There is no "smoking gun" yet. The data didn't scream, "We found a new particle!"
• The Interesting Hint: However, the data didn't perfectly match the "no particle" story either.
  • For the Triple Exchange and Chemical Potential models, the data showed a slight preference for the wavy, oscillating signal.
  • Specifically, for the Chemical Potential model, they found a "local significance" of about 1.5 to 2.5 sigma.
  • What does that mean? In statistics, a "sigma" is a measure of confidence. 5 sigma is a confirmed discovery (like finding the Higgs boson). 1.5 sigma is like hearing a faint noise in the background that might be music, but it could also just be static. It's a "maybe," not a "yes."

6. Why This Matters

Even though they didn't find a new particle yet, this paper is a huge step forward for two reasons:

1. Better Maps: They created the most complete "maps" of what these signals should look like. Before, scientists were trying to find a needle in a haystack using a blurry picture. Now they have a high-definition photo of the needle.
2. The "Background" Noise: They realized that if you only look at the loudest part of the signal, you might miss the context. By calculating the entire shape (including the quiet background parts), they can distinguish between a real signal and random noise much better.

The Bottom Line

The authors built the best possible theoretical "search templates" for heavy particles from the early universe. They ran the search using the best data we have (Planck). They didn't find a confirmed new particle, but they found a faint, intriguing whisper in the data that suggests the universe might be hiding some heavy, exotic physics that we haven't fully understood yet.

It's like they tuned the radio to a new frequency and heard a static-filled song that might be a new band. They haven't identified the band yet, but now they know exactly what to listen for when the next, clearer radio (a future telescope) comes online.

Technical Summary

1. Problem Statement

The "Cosmological Collider" (CC) program aims to probe particle physics at energy scales far beyond terrestrial colliders (up to $H \sim 10^{13}$ GeV) by analyzing primordial non-Gaussianity (NG) in the Cosmic Microwave Background (CMB). Specifically, massive particles produced during inflation can decay into inflaton quanta, inducing oscillatory, non-analytic signatures in the bispectrum (three-point correlation function).

Key Challenges Addressed:

• Incomplete Kinematic Coverage: Previous theoretical studies often focused only on the "squeezed limit" (where one momentum is much smaller than the others) to extract the oscillatory signal. However, current observational data (Planck) contains a vast amount of statistical power in the non-squeezed (e.g., equilateral) regions. Naively extrapolating squeezed-limit formulas to the full kinematic space is misleading because off-shell, non-oscillatory background contributions dominate the non-squeezed region.
• Missing Topologies: While "Single Exchange" (SE) diagrams have been extensively studied, "Double Exchange" (DE) and "Triple Exchange" (TE) diagrams, as well as mechanisms involving "Scalar Chemical Potentials" (SCP) that allow for heavy particles ($M \gg H$) to be produced on-shell, lacked full kinematic shape calculations.
• Search Limitations: Without accurate full-shape templates, observational searches using Planck data have been unable to rigorously constrain or detect these specific CC signals.

2. Methodology

The authors employ a unified computational framework to derive the full bispectrum shapes for various tree-level diagrams across the entire kinematic space.

Computational Techniques:

• Cosmological Bootstrap: An analytical method solving differential equations for correlators. While effective for SE, it suffers from numerical instability near the "folded limit" ($k_1 + k_2 \approx k_3$) due to the cancellation of divergences. It was not yet available for TE.
• Coupled Mode Function (CMF): A numerical approach that treats quadratic mixing between the inflaton and heavy scalars non-perturbatively. This reduces exchange diagrams to contact diagrams, which are then numerically integrated over Euclidean de Sitter time.
  • Strategy: The authors use CMF for SE, DE, and TE to ensure stability across the full kinematic space. For the SCP model (which involves complex scalars and extra degrees of freedom), they utilize existing analytical Bootstrap results adapted for the specific couplings.

Models Investigated:

1. Single Exchange (SE): One massive scalar propagator.
2. Double Exchange (DE): Two massive scalar propagators.
3. Triple Exchange (TE): Three massive scalar propagators.
4. Scalar Chemical Potential (SCP): A mechanism where the rolling inflaton field injects energy ($\omega$) into the interaction vertices, allowing the production of particles with $M \gg H$ without exponential Boltzmann suppression.

Observational Search:

• The authors generated shape functions on a grid in momentum space ($k_1, k_2, k_3$).
• They utilized the CMB-BEST pipeline to search for these shapes in the Planck 2018 CMB data.
• Normalization: A novel normalization scheme was adopted where the maximum absolute value of the shape function over the physical momentum region is set to 1. This avoids artificial fluctuations in $f_{NL}$ constraints caused by the oscillatory nature of the shapes.

3. Key Contributions

• First Full-Kinematic Shapes: The paper provides the first complete calculation of bispectrum shapes for Triple Exchange (TE) and Scalar Chemical Potential (SCP) scenarios across the entire kinematic space, not just the squeezed limit.
• Unified Treatment: It offers a unified framework comparing SE, DE, and TE, demonstrating how the prominence of oscillatory features changes with the number of exchanges.
• Mechanism for $M \gg H$: It extends the search for CC signals to the regime of very heavy particles ($M \gg H$) by incorporating the Scalar Chemical Potential mechanism, which counteracts the exponential suppression typical of standard CC scenarios.
• Template Construction: The authors explicitly construct templates that include both the oscillatory "signal" and the smooth "background," proving that neglecting the background leads to misleading observational constraints.

4. Results

A. Multiple Exchange Models (SE, DE, TE):

• Search Outcome: No significant evidence for NG was found in the SE or DE models. The constraints on the coupling constants were weaker than theoretical bounds from perturbativity and power spectrum corrections.
• Triple Exchange (TE) Anomaly:
  • The TE model showed a local significance of $1.25\sigma$ at $\tilde{\nu} \approx 1.9$ (where $\tilde{\nu} = \sqrt{M^2/H^2 - 9/4}$).
  • Physical Insight: At $\tilde{\nu} \approx 1.8$, the smooth background contribution to the bispectrum vanishes numerically, leaving the shape dominated purely by the oscillatory signal. This makes the TE shape distinct from the equilateral shape, leading to a higher Signal-to-Noise Ratio (SNR) in the data, though not statistically significant enough to claim a discovery.
  • Constraints: Unlike SE and DE, the TE model is not constrained by power spectrum corrections in the same way (due to partial wave unitarity bounds being weaker), making it a promising target for future CC searches.

B. Scalar Chemical Potential (SCP) Model:

• Search Outcome: A search was conducted over a 2D parameter space of chemical potential ($\omega$) and mass ($\tilde{\nu}$).
• Global Significance: The full model showed a global significance of $1.5\sigma$ (local significance up to $1.75\sigma$) for the parameter region where $\omega - \tilde{\nu} \approx 4$.
• Component Analysis:
  • When separating the interaction into "time derivative" and "covariant derivative" contributions, the covariant derivative component showed a higher local significance of $2.5\sigma$ (global $1.5\sigma$) at $\omega - \tilde{\nu} \approx 3.5$.
  • The significance is driven primarily by the non-squeezed (equilateral) region of the bispectrum, not the squeezed limit. This confirms the necessity of full-kinematic templates.
• Interpretation: While the $1.5\sigma$ evidence is intriguing, it is not yet a discovery. However, it suggests that the data slightly prefers an oscillatory signature over a purely smooth equilateral shape.

5. Significance and Future Directions

• Paradigm Shift in CC Searches: The paper establishes that accurate CC searches require full-kinematic templates including background contributions. Relying solely on squeezed-limit approximations misses the bulk of the statistical power available in CMB data.
• Validation of TE and SCP: The results highlight that Triple Exchange and Chemical Potential mechanisms are theoretically viable and observationally distinct, with TE potentially offering the strongest signals due to the suppression of background noise at specific mass values.
• Future Work:
  • The authors suggest that the observed $1.5\sigma$ hints warrant further investigation with future CMB experiments (e.g., CMB-S4) which will have higher sensitivity.
  • They propose exploring "Primordial Feature" models and more model-agnostic searches for oscillatory features, as the current data hints at the presence of such signals.
  • The work underscores the potential for the Cosmological Collider to probe particles with masses $M \gg H$ if energy injection mechanisms like the Scalar Chemical Potential are active.

In summary, this paper bridges the gap between theoretical CC calculations and observational data analysis by providing robust, full-shape templates. It finds no definitive discovery but identifies a mild ($1.5\sigma$) preference for oscillatory signatures in Planck data, specifically favoring models like Triple Exchange and Scalar Chemical Potentials that produce distinct shapes in the non-squeezed regime.