Planckian Axions in String Theory

TL;DR

The paper investigates whether axion field spaces in Calabi–Yau compactifications can accommodate super-Planckian field displacements, a key requirement for large-field inflation. By formulating the fundamental domain ${\cal M}_{\Gamma}$ and its diameter ${\cal D}$ in terms of the field-space metric $K$ and the integer matrix ${\cal Q}$ defining the identifications, the authors derive a compact, general expression for the diameter and show that random-matrix universality yields a parametric enhancement: ${\cal D}$ typically scales as $\mathcal{D} \sim N f_N$ and, in the $P=N$ case where eigenvector delocalization occurs, as $\mathcal{D} \sim N^{3/2} f_N$. They validate these ideas in explicit Type IIB CY compactifications, notably the Denef–Douglas–Florea–Grassi–Kachru (DDFGK) setup with $h^{1,1}=51$, where the largest metric eigenvalue is $f_N \approx 0.013\,M_{\rm pl}$ and the diameter along the lightest direction is $\mathcal D_{\text{light}} \approx 1.13\,M_{\rm pl}$, matching semi-analytic Wishart-based expectations when accounting for constraint-induced rescalings. The work unifies decay-constant alignment (KNP) with eigenvector delocalization into a single framework, offering a path to large-field axion inflation within well-controlled flux vacua and highlighting caveats from saxion couplings and Planck-mass renormalization. Overall, the results demonstrate that super-Planckian axion diameters are generic in large-$N$ axion theories and can be realized in explicit string vacua, with potential phenomenological implications for primordial gravitational waves and UV-complete inflationary models.

Abstract

We argue that super-Planckian diameters of axion fundamental domains can naturally arise in Calabi-Yau compactifications of string theory. In a theory with $N$ axions $θ^i$, the fundamental domain is a polytope defined by the periodicities of the axions, via constraints of the form $-π<Q^{i}_{j} θ^j<π$. We compute the diameter of the fundamental domain in terms of the eigenvalues $f_1^2\le\...\le f_N^2$ of the metric on field space, and also, crucially, the largest eigenvalue of $(QQ^{\top})^{-1}$. At large $N$, $QQ^{\top}$ approaches a Wishart matrix, due to universality, and we show that the diameter is at least $N f_{N}$, exceeding the naive Pythagorean range by a factor $>\sqrt{N}$. This result is robust in the presence of $P>N$ constraints, while for $P=N$ the diameter is further enhanced by eigenvector delocalization to $N^{3/2}f_N$. We directly verify our results in explicit Calabi-Yau compactifications of type IIB string theory. In the classic example with $h^{1,1}=51$ where parametrically controlled moduli stabilization was demonstrated by Denef et al. in [1], the largest metric eigenvalue obeys $f_N \approx 0.013 M_{pl}$. The random matrix analysis then predicts, and we exhibit, axion diameters $>M_{pl}$ for the precise vacuum parameters found in [1]. Our results provide a framework for achieving large-field axion inflation in well-understood flux vacua.

Figures (4)

• Figure 1: The fundamental domain in the presence of $P \gg N$ constraints, for $N=2$. The square shown is the domain $|v_{1,2}|\le \pi$, and the lines are 100 hyperplanes defined by $|({\cal Q}Q^{-1}{\mathbf v})_{1,2}|=\pi$, where the elements of ${\cal Q}Q^{-1}$ are Gaussian distributed with standard deviation $\sqrt{2}$. The black circle illustrates the typical location of hyperplanes, while the dashed, red circle illustrates the analytic estimate (\ref{['qrectangulardiam']}) for the size of the fundamental domain.
• Figure 2: Left: Diameter versus the number of fields for a fixed number ${\mathcal{P}}=4N$ of non-vanishing entries in ${\mathbf Q}$. Right: Kinematic range vs. ${\mathcal{P}}/N$ for fixed $N=100$. Dashed lines illustrate numeric results, and the solid lines are the analytic results. From top to bottom, red: unit metric (\ref{['eqn:unitdiam']}); green: Wishart metric (\ref{['wisrange']}); gray: non-square $\mathcal{Q}$ matrix (\ref{['wisrange']}) with $P-N=3$; orange: heavy-tailed metric (\ref{['eqn:heavydiam']}); blue: $\sqrt{N}$ for comparison.
• Figure 3: Contour plot of a two-dimensional axion potential, along with the region of validity of the quadratic expansion and a set of randomly chosen inflationary trajectories. The axes are canonically normalized fields.
• Figure 4: Normalized probability distribution of the eigenvalue spectrum of $\mathbf Q^\top \mathbf Q$ along with the analytic Wishart eigenvalue spectrum.