⚛️ high-energy theory

Dimensional Regularization of Bubble Diagrams in de Sitter Spacetime

This paper employs the Källén-Lehmann representation and dimensional regularization to analytically compute ultraviolet-divergent loop correlators, specifically 4-point and 2-point functions involving bubble diagrams of massive bulk propagators in de Sitter spacetime.

Original authors: Hongyu Zhang

Published 2026-01-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hongyu Zhang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Listening to the Baby Universe

Imagine the universe as a giant, expanding balloon. A long time ago, this balloon was inflating incredibly fast—a period scientists call "inflation." During this time, tiny quantum fluctuations (like tiny ripples on the surface of the balloon) were stretched out to become the seeds of galaxies and stars we see today.

Scientists want to understand what happened during that inflationary era. They do this by looking at "correlators." Think of a correlator as a recipe card that tells you how different parts of the universe are connected. If you know the recipe, you can figure out what ingredients (particles) were present and how they interacted.

The Problem: The "Bubble" Mess

In this paper, the author, Hongyu Zhang, is looking at specific recipes that involve a "bubble."

The Analogy: Imagine you are baking a cake (the universe). Sometimes, the recipe calls for a secret ingredient that appears, disappears, and reappears in a loop before the cake is finished. In physics, this is called a "loop" or a "bubble diagram."
The Issue: When scientists try to calculate the math for these bubbles, the numbers often blow up to infinity. It's like trying to measure the weight of a cake, but the scale keeps screaming "ERROR: INFINITY!" because the math gets too messy. This is called a "divergence."

To get a real answer, you have to use a technique called regularization. This is like putting a temporary cap on the scale to stop it from screaming, doing the math, and then carefully removing the cap to see what the true weight is.

The Solution: Changing the Dimensions

The author uses a specific type of regularization called Dimensional Regularization.

The Metaphor: Imagine you are trying to count the number of grains of sand on a beach. If you try to count them in 3D (height, width, depth), it's impossible because there are too many. But, if you could magically shrink the beach into a 2D line or a 1D string, the counting becomes manageable.
How it works here: The author temporarily changes the rules of the universe. Instead of calculating in our normal 3 dimensions of space, he calculates in a "fractional" dimension (like 2.999 dimensions). In this weird, slightly smaller world, the math stops blowing up to infinity. He solves the puzzle there, and then slowly expands the world back to 3 dimensions. The "infinity" appears as a specific, manageable error term that he can subtract out.

What Was Calculated?

The paper focuses on three specific types of "bubbles" (loops) involving different particles:

Scalar Particles (The Simple Balls): These are like simple, round marbles. The author calculated how these marbles interact in a loop.
Derivative Coupled Scalars (The Spinning Tops): These are marbles that are also spinning or moving in a specific way that changes the math. The author had to be extra careful with the "spinning" part to get the right answer.
Massive Vector Bosons (The Arrows): These are particles that have a direction (like an arrow). Calculating loops with arrows is much harder than with round marbles because the direction matters.

The "Bubble" vs. The "Contact"

A key discovery in the paper is about how these bubbles behave compared to direct interactions.

The Analogy: Imagine two people talking.
- Direct Interaction: They shout directly at each other.
- Bubble Loop: They shout, the sound bounces off a wall (the bubble), and then they hear it.
The Finding: The author found that when you calculate the "bouncing" (the loop), the math produces a specific kind of error (infinity). To fix this, you have to add a "counter-term" (a correction factor) to the recipe.
The Twist: The author discovered that for the "spinning tops" and "arrows," the correction factor depends on the Hubble parameter (the speed at which the universe is expanding). This means the "recipe" for the universe isn't just about the particles themselves; it's also about the shape and expansion of the space they are in. The particles are interacting with the curvature of spacetime itself.

The "Flat Space" Check

To make sure his math was right, the author did a "sanity check."

The Analogy: If you are trying to figure out how a boat floats in a stormy ocean (de Sitter space), you first check how it floats in a calm, flat swimming pool (flat spacetime).
The Result: When he turned off the expansion of the universe (made it flat), his complex, expanding-universe math matched the known, simple math for flat space perfectly. This proved his method works.

Summary

In short, this paper provides a new, clean way to solve messy math problems that arise when studying the early universe.

The Problem: Calculating particle loops in the early universe leads to infinite numbers.
The Tool: The author uses "Dimensional Regularization" (shrinking the dimensions temporarily) to tame the infinities.
The Result: He successfully calculated the exact "recipes" (correlators) for complex particle loops involving spinning and directional particles.
The Insight: He showed that these calculations require corrections that depend on the expansion of the universe, proving that the geometry of space plays a crucial role in how these quantum particles behave.

This work helps cosmologists better understand the "cosmic collider" signals—the unique fingerprints left behind by heavy particles that existed only for a split second during the birth of the universe.

Technical Summary: Dimensional Regularization of Bubble Diagrams in de Sitter Spacetime

Problem Statement
Inflationary cosmology relies heavily on the study of correlation functions (inflation correlators) to probe heavy particles and their interactions at the inflation scale. While tree-level correlators are well-understood, loop-level processes are crucial for many particle models where leading "cosmological collider" signals appear only at the one-loop level (e.g., models with Standard Model particles in the symmetric phase or heavy neutrinos). However, computing these loop correlators in de Sitter (dS) spacetime presents significant challenges, primarily due to ultraviolet (UV) divergences arising from loop momentum integrals. Existing regularization schemes in dS spacetime have largely focused on massless scalar fields or specific conformal cases. There is a lack of a systematic method to regularize correlators involving bubble loops with internal lines of arbitrary mass and spin (specifically spin-0 and spin-1), particularly when derivative couplings are involved.

Methodology
This work employs dimensional regularization combined with the Källén-Lehmann spectral representation to analytically compute one-loop bubble diagrams in de Sitter spacetime. The core methodology involves the following steps:

Seed Integral Formulation: The authors utilize a "loop seed integral" ( $J_{\sigma\sigma}^{p_1 p_2}$ ) which encapsulates the dependence on external momenta through specific ratios ( $r_1, r_2$ ). This integral is derived from the Källén-Lehmann representation, expressing the bubble loop as an integral over a spectral function ( $\rho$ ) of tree-level seed integrals ( $I_{e\nu'}^{p_1 p_2}$ ).
Spectral Decomposition: The tree-level seed integrals are decomposed into non-local signals (NL), local signals (L), and background (BG) terms. The full loop result is obtained by integrating these components against the spectral function.
Handling Divergences: The authors identify that while individual terms in the spectral integration may converge, the summation over the spectral index (often denoted as $n$ $n$ ) diverges as the spatial dimension $d \to 3$ $d \to 3$ (corresponding to spacetime dimension $D \to 4$ $D \to 4$ ).
- Instead of standard cutoff regularization, the authors extract the divergent part of the summation by analyzing the asymptotic expansion of the summands with respect to the summation index $n$ .
- The divergent series are regularized by subtracting these asymptotic expansions and introducing Riemann zeta functions ( $\zeta(s)$ ) to analytically continue the sums to arbitrary dimensions.
- The divergences manifest as poles in $(3-d)$ (or $1/(3-d)$ ). These poles are subtracted using the Minimal Subtraction (MS) scheme.
Application to Specific Models: The method is applied to three distinct cases:
- $\sigma\sigma$ Model: Directly coupled massive scalars.
- $\nabla\sigma\nabla\sigma$ Model: Derivatively coupled massive scalars.
- $AA$ Model: Massive vector bosons (spin-1) coupled to scalars.
- For derivative and vector couplings, the authors address the fact that the Källén-Lehmann representation cannot be applied directly to the time-ordered propagators in the loop. They derive necessary correction terms arising from the difference between the spectral representation and the actual time-ordered bubble, specifically handling contact terms involving delta functions and derivatives.

Key Contributions and Results

Analytic Expressions for Loop Correlators: The paper provides explicit analytic expressions for 1-loop 4-point and 2-point correlators with bubble exchanges of massive scalars and massive vectors in dS spacetime. These expressions are valid for arbitrary masses and include the finite parts after renormalization.
Regularization of Derivative and Vector Loops: A significant contribution is the successful application of dimensional regularization to models with derivative couplings ( $\nabla\sigma\nabla\sigma$ ) and massive vector loops ($AA$). The authors demonstrate that naive application of the spectral representation fails for these cases due to contact terms and derive the necessary corrections to obtain the correct UV divergences.
Counter-term Determination: The work calculates the specific coefficients of the local counter-terms required to cancel the UV divergences.
- For the $\nabla\sigma\nabla\sigma$ and $AA$ models, the counter-term coefficients are found to depend on the Hubble parameter $H$ (via the mass parameter $e_\nu$ ). This implies that the renormalization of these models in dS spacetime requires interactions between the scalar fields and the spacetime curvature (e.g., terms like $R(\nabla\phi)^2$ or $R^2\phi^4$ ).
- The authors verify that in the flat space limit ( $H \to 0$ ), the coefficients of the divergences match those calculated in flat spacetime, ensuring consistency.
Spectral Functions: The paper derives explicit spectral functions for the $\nabla\sigma\nabla\sigma$ and $AA$ models using the embedding formalism, which are essential for the spectral decomposition method.

Significance and Claims
The paper claims to provide a robust and systematic framework for handling UV divergences in one-loop inflation correlators with arbitrary mass and spin. By utilizing dimensional regularization via the Källén-Lehmann representation, the authors avoid the difficulties associated with cutoff regularization in curved spacetime (such as the difficulty of taking asymptotic expansions of cutoffs).

The work highlights that:

Dimensional regularization in dS spacetime can be effectively performed by isolating divergent summations and mapping them to Riemann zeta functions.
The renormalization of derivative-coupled and vector-loop models in dS spacetime necessitates curvature-dependent counter-terms, a feature distinct from flat space renormalization.
The derived expressions serve as a foundation for analyzing cosmological collider signals in models where tree-level signals are absent or subdominant.

The authors note that while the method is demonstrated for bubble diagrams, it is expected to be applicable to more complex topologies, though the full expressions for such correlators remain to be derived. The work is presented as a technical advancement in the analytical toolkit for inflationary cosmology, specifically for the "cosmological collider" program.