⚛️ high-energy theory

Non-minimal unimodular inflation revisited: Swampland diagnostics and CMB constraints

This paper analyzes non-minimal unimodular inflation in the Einstein frame, deriving its exact analytical observables and demonstrating that while the model can reproduce acceptable CMB parameters at isolated field values, it fails to satisfy both scalar weak gravity conjectures and cannot simultaneously fulfill the refined de Sitter conjecture while supporting a complete 50–60 e-fold inflationary history.

Original authors: Saeed Noori Gashti, Mohammad Reza Alipour, Mohammad Ali S. Afshar, Behnam Pourhassan, żzzet Sakallı, Phongpichit Channuie

Published 2026-10-21

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Saeed Noori Gashti, Mohammad Reza Alipour, Mohammad Ali S. Afshar, Behnam Pourhassan, żzzet Sakallı, Phongpichit Channuie

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

In the earliest moments of the universe, a fraction of a second after the Big Bang, space itself is thought to have undergone a period of explosive expansion known as inflation. This rapid stretching smoothed out the cosmos, setting the stage for the formation of stars and galaxies. To understand how this happened, physicists build mathematical models of the energy fields that drove this expansion. These models must do two difficult things at once: they must match the precise measurements of the cosmic microwave background—the faint afterglow of the Big Bang that we can detect today—and they must be compatible with the deep, theoretical rules that govern how gravity and quantum mechanics work together. If a model fits the data but breaks the fundamental rules of quantum gravity, it is said to belong to the "swampland," a collection of theories that look consistent on the surface but cannot exist in a complete universe.

A team of researchers recently revisited a specific model of inflation called non-minimal unimodular inflation to see if it could survive this dual test. This model proposes that the energy driving the expansion comes from a field that interacts with gravity in a slightly unusual way, creating a specific shape for the energy landscape. The scientists wanted to know if this model could produce the right kind of universe we see today while also passing strict theoretical checks designed to ensure it fits within the framework of quantum gravity. They found that while the model can mimic the observed universe at a single snapshot in time, it fails to tell a complete story of how the universe evolved, and it breaks the fundamental rules of quantum gravity in the very regions where it is most needed.

The researchers began by translating the model into a standard language that physicists use to describe the universe, known as the Einstein frame. This step was crucial because the original version of the model used a different mathematical perspective that made it difficult to compare with other theories. By converting the model into this standard form, they could calculate exactly how the energy field would behave as the universe expanded. They derived precise formulas to predict two key numbers that astronomers measure: the tilt of the spectrum of density fluctuations, which tells us how clumpy the early universe was, and the ratio of gravitational waves to light waves, which hints at the energy scale of inflation.

When they ran the numbers, a structural problem emerged. The model works well when a specific parameter, which controls the strength of the interaction between the field and gravity, is very small. In this small range, the model predicts values for the cosmic tilt that match the latest observations from the Atacama Cosmology Telescope and the Planck satellite. However, for the universe to stop inflating and transition into the hot, dense state that followed the Big Bang, the model requires this same parameter to be large—specifically, larger than one-half. This creates a contradiction: the model needs the parameter to be tiny to match the data, but it needs the parameter to be large to finish the job of inflation. The researchers showed analytically that these two requirements cannot be met at the same time. The model cannot smoothly roll down its energy hill and then stop; it either never stops, or it stops in a way that produces a universe that does not look like ours.

Beyond this mechanical failure, the team tested the model against three theoretical criteria known as swampland diagnostics. These are rules derived from string theory and quantum gravity that act as a filter for valid theories. One of these rules, the further-refined de Sitter conjecture, involves adjustable parameters that allow for some flexibility. The researchers found that the model could satisfy this rule, but only if those adjustable parameters were set to extreme values that make the rule very weak. The other two rules, the scalar weak gravity conjecture and its stronger version, do not have such adjustable knobs. These rules are strict mathematical inequalities that the model must satisfy at every point in its evolution.

The analysis revealed that the model fails these strict tests completely. At the very point where the energy field is most stable, the model violates the scalar weak gravity conjecture for any non-zero value of the interaction parameter. Even when the researchers looked at the extreme limits of the model, where the field values are very large, the model only satisfied the rules if the interaction parameter was greater than 0.1. This is a problem because the values of the parameter that make the model fit the observational data are much smaller, around 0.003. In the region where the model looks good to astronomers, it breaks the laws of quantum gravity. In the region where it obeys the laws of quantum gravity, it produces a universe that does not match what we observe. Specifically, near the origin of the field, the Strong Scalar Weak Gravity Conjecture requires the interaction parameter to be at least approximately 0.0643. However, the observational data requires the parameter to be much smaller to produce the correct cosmic tilt; at the threshold of 0.0643, the model would predict a cosmic tilt far below the range observed in the cosmic microwave background.

The researchers also examined the transition period after inflation, known as reheating, where the energy of the inflation field is transferred to the matter and radiation that fill the universe. They found that the standard way of ending inflation in this model is incompatible with the slow, steady expansion required to produce the observed universe. Because the model cannot naturally exit the inflationary phase in a way that lasts for the necessary 50 to 60 cycles of expansion, it cannot be considered a complete history of the early universe. The fact that it can match a single data point at a specific moment does not save it; a valid theory must describe the entire journey from the start of inflation to the end.

Ultimately, the study concludes that the minimal version of non-minimal unimodular inflation is not a viable description of our universe. It creates a tension between three distinct requirements: matching the observational data, completing a full history of inflation, and satisfying the theoretical constraints of quantum gravity. The model manages to satisfy the first requirement at a single point, but it fails the other two. The researchers emphasize that this does not rule out the broader idea of unimodular gravity or non-minimal inflation entirely. Instead, it identifies a specific flaw in the simplest version of the theory. To make the idea work, future models would need to modify the way the inflation ends, change the shape of the energy potential, or alter the hierarchy of the mathematical derivatives that govern the field's behavior. Until such modifications are made and tested, this particular model remains in the swampland, unable to bridge the gap between the data we see and the deep laws that govern reality.

Technical Summary: Non-minimal Unimodular Inflation Revisited

Problem Statement
Recent joint analyses of ACT-DR6, Planck 2018, and DESI BAO datasets have refined the allowed parameter space for inflationary models, shifting the preferred scalar spectral index ( $n_s$ ) upward to approximately $0.9743 \pm 0.0034$ and tightening the tensor-to-scalar ratio bound to $r < 0.038$ . This shift creates tension for minimal inflationary constructions, such as Starobinsky $R^2$ and Higgs-like models, which often predict lower $n_s$ values. The paper investigates a specific candidate, the non-minimal unimodular inflation model, to determine if it can simultaneously satisfy three distinct criteria:

Phenomenological Viability: Reproducing the observed CMB observables ( $n_s, r$ ) over a complete 50–60 e-fold inflationary history.
Dynamical Consistency: Possessing a self-consistent slow-roll phase that terminates via the standard condition ( $\epsilon_V = 1$ ).
Swampland Consistency: Satisfying three quantum-gravity-inspired diagnostics: the Further-Refined de Sitter Conjecture (FRDSSC), the Scalar Weak Gravity Conjecture (SWGC), and its Strong form (SSWGC).

Methodology
The authors analyze the model in the Einstein frame, where the scalar kinetic term is canonically normalized. The effective potential is given by:
$U(\phi) = U_0 \left(1 + \xi \frac{\phi^2}{M_P^2}\right)^{-2}$
where $\xi$ is the non-minimal coupling shape parameter and $U_0$ is the energy scale.

The methodology proceeds through four distinct analytical stages:

Canonical Field Mapping: The authors derive the exact transformation between the Jordan-frame field and the canonical Einstein-frame field ( $\phi \to \varphi$ ) to ensure all derivative-based calculations are frame-independent.
Analytic Derivation: Exact expressions for potential slow-roll parameters ( $\epsilon_V, \eta_V$ ), the scalar spectral index ( $n_s$ ), the tensor-to-scalar ratio ( $r$ ), the scalar running ( $\alpha_s$ ), and the e-fold number ( $N$ ) are derived analytically.
Structural Tension Analysis: The paper analytically examines the conditions required for a slow-roll phase near the origin versus the conditions required for the inflationary exit ( $\epsilon_V = 1$ ).
Swampland Diagnostics: The three conjectures are evaluated using the canonically normalized field.
- FRDSSC: Tested with tunable conjectural parameters $a, b, q$ .
- SWGC & SSWGC: Tested as fixed inequalities involving derivatives of the potential ( $d_n$ ) with no tunable parameters in the single-field realization.
Numerical Scans: A global numerical scan is performed over the parameter space $(\xi, x)$ to check for simultaneous overlap between observational envelopes (Planck 2018 and P-ACT-LB) and the regions satisfying the Swampland criteria.

Key Contributions and Results

1. Structural Dynamical Tension
The analysis reveals a fundamental incompatibility within the minimal model:

Slow-Roll Regime: To maintain a slow-roll regime near the origin (required for the observed red tilt $n_s \approx 1 - 8\xi$ ), the parameter must satisfy $\xi \ll 1$ .
Exit Condition: To terminate inflation via the standard condition $\epsilon_V = 1$ , the parameter must satisfy $\xi > 1/2$ .
Conclusion: These conditions are mutually exclusive. For $0 < \xi \leq 1/2$ , $\epsilon_V$ asymptotically approaches a value less than 1, meaning inflation does not end via the standard mechanism. For $\xi > 1/2$ , the slow-roll condition $|\eta_V| < 1$ is violated well before the exit point is reached. Thus, the model cannot support a complete, self-consistent 50–60 e-fold history with a standard graceful exit.

2. Swampland Diagnostics

FRDSSC: This criterion can be satisfied, but only under specific conditions. It requires the conjectural parameter $a$ to be very close to unity ( $a \gtrsim 0.987$ for the P-ACT-LB central value), implying $b$ is at the percent level. This is a theoretical compatibility condition, not an empirical measurement.
SWGC: The Scalar Weak Gravity Conjecture is violated at the stationary point ( $x=0$ ) for every non-zero $\xi$ . In the large-field limit, the criterion requires $\xi \geq 0.1$ . However, the observationally relevant region requires $\xi \sim 10^{-3}$ . Numerical scans confirm that the observationally relevant ACT/P-ACT region does not satisfy the SWGC.
SSWGC: The Strong form is less restrictive than the SWGC at finite fields. A small subregion compatible with Planck 2018 data satisfies the SSWGC. However, this region disappears when the stricter P-ACT-LB observational envelope is applied.

3. Numerical Findings
Global scans of the $(\xi, x)$ parameter space demonstrate that the regions satisfying the scalar weak-gravity criteria (SWGC/SSWGC) are disjoint from the regions satisfying the observational constraints (Planck/P-ACT).

A specific benchmark point ( $\xi = 0.003, x^* = 5.1413$ ) successfully reproduces the ACT-era pivot values ( $n_s = 0.9743, r = 0.02818$ ).
However, at this exact point, both $\Delta_{SW}$ and $\Delta_{SSW}$ are negative, indicating a failure of the Swampland criteria.
Increasing $\xi$ to satisfy the Swampland criteria ( $\xi \geq 0.1$ ) drives the predicted $n_s$ far below observational bounds (e.g., $n_s \approx 0.2$ for $\xi=0.1$ ).

Significance and Claims
The paper concludes that the minimal non-minimal unimodular inflation model fails to simultaneously satisfy observational constraints, dynamical completeness, and Swampland consistency.

The model can reproduce acceptable observables at an isolated pivot scale, but this does not constitute a complete inflationary history (50–60 e-folds with a standard exit).
The failure to satisfy the SWGC and SSWGC in the observationally relevant region is a structural feature of the potential's shape and its canonical derivatives, not a result of parameter tuning.
The authors emphasize that satisfying the FRDSSC by tuning $a \to 1$ does not rescue the model, as the SWGC and SSWGC contain no analogous tunable parameters and remain violated.

The work serves as a "no-go" result for the minimal construction, suggesting that a viable extension would require modifying the mechanism governing the end of inflation, the functional form of the potential, or the hierarchy of its canonical derivatives. The authors explicitly state that these results do not rule out unimodular gravity as a framework generally, but rather identify a specific obstruction within the minimal potential and its canonical dynamics.

Technical Summary: Non-minimal Unimodular Inflation Revisited

More like this