⚛️ high-energy theory

The growth of $SO(9)$ super-representations in Type II string theory

This paper investigates the growth of massive $SO(9)$ super-representations in Type II and Type I string theories by deriving empirical formulas, computing multiplicities via refined partition functions over finite fields, and establishing accurate asymptotic approximations using Rademacher sums that reveal contributions from both even and odd terms.

Original authors: Alessandro Georgoudis, Joseph A. Minahan, Gustav Ström, Athanasios Zoumis

Published 2026-09-04

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Alessandro Georgoudis, Joseph A. Minahan, Gustav Ström, Athanasios Zoumis

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 vast landscape of theoretical physics, string theory offers a framework where the fundamental particles of nature are not point-like dots, but tiny, vibrating loops of energy. The way these loops vibrate determines what kind of particle they appear to be, from the massless photon that carries light to the massive particles that make up the core of atoms. As these strings vibrate more vigorously, they gain mass, creating an endless tower of heavier and heavier states. Physicists have long known that the number of these possible states grows incredibly fast as the mass increases, a phenomenon that hints at a fundamental temperature limit for the universe known as the Hagedorn temperature. However, while the total number of states is well understood, the specific types of particles that appear at each level of mass have remained a complex puzzle. These particles are organized into groups based on the symmetries of the universe, specifically a geometric structure called SO(9), which dictates how these massive states relate to one another. Understanding exactly how many of each type of particle exists at a given mass is crucial for testing the consistency of string theory and for exploring the deep mathematical structures that underpin reality.

A team of researchers has now taken a significant step forward in solving this puzzle by mapping out the growth of these massive particle groups with unprecedented precision. Instead of trying to guess the pattern from a few low-mass examples, the team developed a method to calculate the exact number of these particle groups, known as super-representations, up to extremely high levels of mass. They pushed their calculations far beyond previous limits, reaching a point where over 1.5 million distinct types of particles were found to exist at a single level of mass. To achieve this, they treated the problem like a massive counting exercise, using advanced computational techniques to break down a complex mathematical function that describes the string's vibrations. By performing these calculations on a specialized system of numbers that avoids rounding errors, they were able to count the multiplicities—the number of times each specific particle type appears—for individual groups up to level 500 and for all groups up to level 226. This massive dataset provided a clear picture of the landscape of string states, revealing that while the total number of states grows predictably, the distribution of specific types of particles follows a more intricate set of rules.

The researchers discovered that these particle groups do not appear randomly. They found a clear, empirical rule that dictates which types of particles can exist at any given mass level. For the vast majority of cases, a particle group appears if it satisfies a specific set of conditions related to its internal structure, filling out a triangular pattern of possibilities. There is, however, one notable exception: a specific type of particle that appears at one mass level but vanishes at the very next, creating a small gap in the otherwise solid wall of possibilities. This rule allowed the team to predict exactly which particle groups should exist before they even performed the heavy calculations, streamlining the process of counting the millions of states. The sheer scale of their findings is staggering; at level 226, the number of distinct particle groups with non-zero counts is so large that visualizing them requires three-dimensional graphs, where the height of the data points represents the number of times each group appears.

With this massive dataset in hand, the team turned their attention to understanding the long-term behavior of these numbers. For decades, physicists have had a formula that predicts how the total number of string states grows as mass increases, but applying a similar formula to the specific groups of particles has been much harder. The researchers found that the simple formulas used for the total count do not work well for individual particle groups, especially at lower mass levels. To fix this, they employed a sophisticated mathematical technique known as a Rademacher sum, which is essentially a way of adding up contributions from different "peaks" in the mathematical landscape of the string's vibrations. By including not just the main peak but also smaller, secondary peaks, they were able to construct a new, highly accurate approximation. This new formula works remarkably well, matching the actual counts of particles with high precision even at relatively low mass levels where previous formulas failed. They found that for most particle groups, the approximation requires contributions from both even and odd mathematical terms, a detail that distinguishes the behavior of these specific groups from the overall string count.

The study also revealed interesting differences between how these massive groups behave and how ordinary, non-supersymmetric particle groups behave. When the researchers applied their methods to the simpler, ordinary representations, they found that the calculations were faster and the resulting formulas were cleaner, yet the underlying patterns of growth remained strikingly similar. This suggests that the complex machinery of supersymmetry, which pairs particles with different properties, does not fundamentally alter the way these massive states proliferate, but rather adds a layer of structure that can be peeled away to reveal the core mathematical engine. The team's work provides a powerful new tool for exploring the high-energy regime of string theory, offering a way to predict the abundance of specific particle types with a level of accuracy that was previously impossible. By combining direct computation with refined mathematical approximations, they have turned a chaotic-looking explosion of possibilities into a structured, predictable landscape, bringing us closer to understanding the full spectrum of the string theory universe.

Technical Summary: The Growth of SO(9) Super-Representations in Type II String Theory

Problem Statement
The paper addresses the problem of determining the multiplicities of massive SO(9) super-representations in Type II and Type I string theories as a function of the string level $n$ . While the asymptotic growth of the total number of string states is well-established (following the Hagedorn behavior), the specific growth of individual SO(9) representations is less understood. Previous work by Curtright and Thorn provided an asymptotic formula for these multiplicities, but it is only valid at extremely high levels ( $n \gtrsim 22$ ) and fails to capture the behavior at lower, physically relevant levels. Furthermore, extracting exact multiplicities for high levels is computationally prohibitive due to the sheer number of representations and the rapid growth of their multiplicities.

Methodology
The authors employ a multi-pronged approach combining empirical pattern recognition, high-precision numerical computation, and analytic asymptotic techniques:

Empirical Rules and Representation Counting: By analyzing data up to level 10, the authors derive an empirical rule (Table 1 and Eq. 3.6) that determines exactly which SO(9) representations $[\ell, m, r, s]$ appear at a given level $n > 6$ . This allows for the precise counting of the number of representations with non-zero multiplicities.
Refined Partition Function and Finite Field Arithmetic: To compute multiplicities at high levels (up to $n=501$ for specific representations and $n=226$ for all representations), the authors utilize a "refined" partition function $Z_L(\omega; \tau)$ . This function is expanded in terms of SO(9) characters. To handle the massive integer values of the multiplicities, the coefficients are computed over finite fields using the Mersenne prime $M = 2^{61}-1$ and additional 61-bit primes. The Chinese Remainder Theorem is then used to reconstruct the exact integer multiplicities.
Weyl Character Formula and Triangular Systems: The extraction of specific multiplicities $a_{k,n}$ from the partition function is performed using the Weyl character formula. By multiplying the partition function by the Weyl denominator, the problem is reduced to a triangular system of linear equations, allowing for the recursive determination of multiplicities.
Asymptotic Analysis via Rademacher Sums: The authors derive improved asymptotic formulae by integrating the refined partition function around the peaks of the modular transformation (saddles). Unlike the total string partition function, which receives contributions only from odd Rademacher terms, the multiplicities of individual representations receive contributions from both even and odd terms. The authors compute the leading saddle (first Rademacher term) and sub-leading saddles (second, third, and fourth Rademacher terms) to construct an infinite series of asymptotic corrections.

Key Contributions and Results

High-Level Multiplicity Data: The paper presents the first computation of SO(9) super-representation multiplicities up to level $n=226$ for all representations (covering over 1.5 million representations) and up to $n=501$ for the six lowest-dimensional representations.
Empirical Existence Rules: A complete set of conditions (Table 1) is provided to determine the presence or absence of any representation $[\ell, m, r, s]$ at level $n$ , revealing a structured "triangular lattice" of allowed representations.
Improved Asymptotic Formulae: The authors derive a general asymptotic expression for the multiplicity $\rho_k(n)$ of any representation $k$ . The leading order matches the Curtright-Thorn result but includes a computable constant and a power-law prefactor. Crucially, they show that the convergence to this asymptotic limit is slow and requires higher-order corrections.
Rademacher Sum Corrections: The paper demonstrates that the asymptotic approximation can be made arbitrarily accurate by including higher-order Rademacher terms.
- Leading Term: The first term (leading saddle) provides a baseline but deviates significantly at low levels.
- Sub-leading Terms: The inclusion of the second, third, and fourth Rademacher terms significantly improves accuracy. Notably, the authors find that for spinor representations (where the last Dynkin index $s$ is odd), the second Rademacher term is suppressed or vanishes due to symmetry, explaining why the leading saddle is a better approximation for spinors than for non-spinors.
- Accuracy: By summing the first few Rademacher terms, the authors can predict multiplicities with high precision (within a few units) for levels as low as $n \sim 30-60$ , far below the regime where the simple Curtright-Thorn formula becomes valid.
Ordinary Representations: The techniques are also applied to ordinary (non-super) SO(9) representations. Due to the absence of the universal long-multiplet factor $Z_Q(\omega)$ in the denominator, computations are faster, and the asymptotic analysis is more tractable, yielding similar structural results.

Significance
The paper claims that its primary significance lies in providing a bridge between exact combinatorial counting and asymptotic analysis in string theory. By establishing that the multiplicities of specific representations can be approximated with high accuracy using Rademacher sums, the need for brute-force computation at extremely high levels is mitigated for many practical purposes. The work validates the Curtright-Thorn asymptotic behavior while quantifying the rate of convergence and the specific corrections required.

The authors note that while the asymptotic results converge to the true multiplicities, the rate of convergence is slower than that of the total string degeneracy, particularly for representations with large dimensions. The paper concludes by suggesting that these results could potentially be applied to the counting of super-representations in $\mathcal{N}=4$ Super Yang-Mills theory in the planar limit, given the known connections between SO(9) super-multiplets and the spectrum of free $\mathcal{N}=4$ SYM. Additionally, the paper highlights the open theoretical challenge of deriving the empirical existence rules (Table 1) from first principles.

More like this