🔢 mathematics

Applications of representation theory and of explicit units to Leopoldt's conjecture

This paper establishes that Leopoldt's conjecture for certain intermediate fields implies the conjecture for the full Galois extension under specific group-theoretic hypotheses, and uses these results alongside explicit unit descriptions to prove the existence of an infinite family of totally real $S_3$ -extensions of $\mathbb{Q}$ where the conjecture holds for any prescribed finite set of primes.

Original authors: Fabio Ferri, Henri Johnston

Published 2026-03-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Fabio Ferri, Henri Johnston

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

Imagine you are a detective trying to solve a massive, centuries-old mystery called Leopoldt's Conjecture.

In the world of mathematics, specifically in a field called Number Theory, there are special numbers called "units" (think of them as the building blocks or the "atoms" of a number system). Mathematicians have a tool called a Regulator that measures how "spread out" these building blocks are.

The Mystery: Leopoldt's Conjecture asks a simple question: Is this spread-out measurement ever zero?

If the answer is No (the measurement is never zero), the conjecture is true. This means the number system is "healthy" and behaves nicely.
If the answer is Yes (the measurement is zero), the system is "broken" or "collapsed."

For simple number systems (like the rational numbers or simple extensions), we know the answer is "No, it's never zero." But for complex, twisted number systems (non-abelian extensions), we've been stuck. We didn't know if the measurement could ever be zero.

This paper by Fabio Ferri and Henri Johnston is like a new detective kit that helps us solve this mystery for a whole new class of complex number systems.

The Detective Kit: Two Main Tools

The authors use two powerful tools to crack the case:

1. The "Symmetry Map" (Representation Theory)

Imagine you have a giant, complex machine (a number field $L$ ) with many gears turning inside it. You want to know if the whole machine is working. Checking the whole machine is hard.

But, the authors realized that this machine is built from smaller, simpler machines (sub-fields). They discovered a Symmetry Map. This map says:

"If you check the smaller, simpler machines, you automatically know the status of the big machine."

The Analogy: Imagine a massive orchestra (the big number field). Instead of listening to the whole orchestra to see if they are in tune, you only need to listen to the individual sections (strings, brass, woodwinds). If every section is playing perfectly, the whole orchestra is perfect.
The Breakthrough: They proved that for many complex groups, if the conjecture holds for the "sub-sections" (intermediate fields), it must hold for the whole group. They didn't need to know the deep, messy details of the whole machine; they just needed to know how the parts fit together using symmetry.

2. The "Blueprint of Building Blocks" (Explicit Units)

Once they knew where to look (the sub-sections), they needed to actually check if those sections were healthy.

To do this, they used a technique developed by Buchmann and Sands. Think of this as having a precise blueprint of the building blocks (units) in these number systems.

The Analogy: Usually, checking if a building is stable requires guessing where the bricks are. But here, the authors had a blueprint that told them exactly where every brick was. They could look at the blueprint and say, "Ah, these bricks are arranged in a way that guarantees the building won't collapse."

The Big Wins: What Did They Prove?

Using these two tools, the authors achieved two major victories:

1. The "Infinite Family" of $S_3$ Extensions
They looked at a specific type of number system called an $S_3$ -extension (think of it as a system with a specific 6-way symmetry, like the rotations and flips of a triangle).

The Result: They found an infinite family of these systems. For any list of prime numbers you give them (like 2, 5, 7, 101), they can generate an infinite list of these number systems where Leopoldt's Conjecture is guaranteed to be true.
Why it matters: Before this, we didn't have a single confirmed example of a complex, non-abelian system where this was proven for odd primes. They broke the deadlock.

2. The "Infinite Family" of $D_8$ Extensions
They did the same thing for $D_8$ -extensions (systems with the symmetry of a square, involving 8 moves).

The Result: They found another infinite family where the conjecture holds for almost all small primes (up to a million, except for 3).

The "Aha!" Moment

The most exciting part of the paper is the connection between the two tools.

Tool 1 told them: "If you solve it for the small cubic fields, you solve it for the big $S_3$ fields."
Tool 2 told them: "Here is a blueprint that proves the small cubic fields are healthy."
Conclusion: Therefore, the big $S_3$ fields are healthy too!

Summary in Plain English

Imagine you are trying to prove that a thousand different types of bridges are safe to drive on.

Old way: You try to drive a truck over every single bridge. It takes forever, and you can't do it for the weird, twisted ones.
This paper's way:
1. They realized that every weird bridge is made of standard, simple beams.
2. They proved a rule: "If the standard beams are safe, the whole weird bridge is safe."
3. They then used a special blueprint to prove that a specific set of standard beams is indeed safe.
4. Boom: They instantly proved that thousands of weird, complex bridges are safe, without ever driving a truck over them.

This paper opens the door to proving this famous conjecture for many more complex number systems than ever before, using the power of symmetry and precise blueprints.

Technical Summary: Applications of Representation Theory and Explicit Units to Leopoldt's Conjecture

Authors: Fabio Ferri and Henri Johnston
Source: arXiv:2301.05700v2 (March 2026)

1. Problem Statement

The paper addresses Leopoldt's Conjecture, a central open problem in algebraic number theory. For a number field $K$ and a prime $p$ , the conjecture asserts that the $p$ -adic regulator $R_p(K)$ is non-zero, or equivalently, that the Leopoldt defect $\delta(K, p)$ (the dimension of the kernel of the $p$ -adic regulator map) is zero.

While the conjecture is proven for abelian extensions of $\mathbb{Q}$ and imaginary quadratic fields (Ax-Brauer theorem), it remains unproven for non-abelian totally real extensions of $\mathbb{Q}$ for odd primes $p$ . Prior to this work, no infinite family of such extensions was known to satisfy the conjecture for any odd prime. The authors aim to:

Establish relations between the Leopoldt defects of a Galois extension $L/K$ and its intermediate subfields using representation theory.
Construct infinite families of non-abelian totally real extensions where the conjecture holds for specific sets of primes.

2. Methodology

The authors employ a dual approach combining representation theory of finite groups with explicit unit constructions.

A. Representation Theory and Idempotent Relations

The core theoretical framework relies on the fact that the Leopoldt kernel $L(L, p)$ is a module over the group algebra $\mathbb{Q}_p[G]$ , where $G = \text{Gal}(L/K)$ .

Idempotent Relations: The authors utilize generalized useful idempotent relations in $\mathbb{Q}[G]$ . These are equalities of the form $1 = \sum a_H e_H$ (where $e_H$ are idempotents associated with subgroups $H$ ).
Deduction of Conjectures: They prove that if such a relation exists, the validity of Leopoldt's conjecture for $L$ is equivalent to its validity for the fixed fields $L_H$ of the subgroups $H$ appearing in the relation.
Brauer Relations: They connect idempotent relations to Brauer relations (relations among induced trivial characters). This allows them to derive linear relations between the Leopoldt defects $\delta(L_H, p)$ of various subfields.
Group Characterization: They apply recent results (Biasse et al.) to identify which finite groups admit these relations, specifically focusing on groups containing non-cyclic subgroups of order $\ell_1 \ell_2$ or specific subgroups like $SL_2(\mathbb{F}_\ell)$ .

B. Explicit Units and Congruence Conditions

To prove the conjecture for specific fields, the authors use techniques developed by Buchmann and Sands:

Explicit Units: They identify families of number fields with explicitly known independent units (e.g., roots of specific polynomials).
$p$ -adic Independence: They verify that these units are linearly independent modulo $p$ (or higher powers of $p$ ) using congruence conditions.
Perturbation/Continuity: Using a theorem by Buchmann and Sands, they show that if the conjecture holds for a specific field $K_{t_0}$ defined by a parameter $t_0$ , and the units satisfy certain congruence properties, then the conjecture holds for an infinite family of fields $K_{t'}$ where $t' \equiv t_0 \pmod{p^m}$ .

3. Key Contributions and Results

Theoretical Advances

Theorem 1.5 & 1.8: Established that Leopoldt's conjecture for a Galois extension $L$ holds if and only if it holds for specific intermediate fields $L_H$ , provided the Galois group $G$ admits a generalized useful idempotent relation. This reduces the problem to smaller, often abelian, subfields.
Theorem 1.7 & 1.9: Proved that for non-abelian extensions over $\mathbb{Q}$ or imaginary quadratic fields, the defect $\delta(L, p)$ is either 0 or bounded below by the smallest degree of a non-linear irreducible character of $G$ (specifically $>1$ ). This rules out the possibility of a defect of exactly 1.
Defect Relations (Theorem 1.10): Derived explicit linear relations between the defects of intermediate fields using Brauer relations. For example, for a Frobenius group $G = N \rtimes H$ , they proved:
$\delta(L, p) + |H|\delta(K, p) = \delta(L_N, p) + |H|\delta(L_H, p)$

Construction of Infinite Families

The authors successfully constructed infinite families of non-abelian totally real extensions where the conjecture is proven for specific primes:

$S_3$ -Extensions (Theorem 1.12):
- Family: Totally real $S_3$ -extensions of $\mathbb{Q}$ (Galois closures of non-Galois cubic fields defined by $x^3 - t^2x - 1 = 0$ ).
- Result: For any finite set of primes $P$ , there exists an infinite family $\mathcal{F}$ of such fields where Leopoldt's conjecture holds for all $F \in \mathcal{F}$ and all $p \in P$ .
- Mechanism: Proved the conjecture for the cubic subfield using explicit units and the Buchmann-Sands method, then lifted it to the $S_3$ -closure using the idempotent relation for $S_3$ .
$D_8$ -Extensions (Theorem 1.13):
- Family: Totally real $D_8$ -extensions of $\mathbb{Q}$ (Galois closures of quartic fields defined by $x^4 - tx^3 - x^2 + tx + 1 = 0$ ).
- Result: There exists an infinite family $\mathcal{F}$ where the conjecture holds for all $p \le 10^6$ (excluding $p=3$ ).
- Mechanism: Similar to the $S_3$ case, utilizing explicit units for the quartic subfield and the relation $\delta(L, p) = 2\delta(F, p)$ derived from the representation theory of $D_8$ .

4. Significance

First Proofs for Non-Abelian Totally Real Fields: This paper provides the first known examples of infinite families of totally real non-abelian Galois extensions of $\mathbb{Q}$ for which Leopoldt's conjecture is proven to hold for odd primes. Previously, such results were limited to $p=2$ or abelian cases.
Methodological Shift: The work moves away from relying on the specific structure of the unit group as a $\mathbb{Q}_p[G]$ -module (which is difficult to determine) and instead uses the existence of the module structure combined with idempotent relations to transfer results from subfields.
Quantitative Bounds: The results provide new lower bounds for the Leopoldt defect in cases where the conjecture fails, showing that the defect cannot be 1 for non-abelian extensions over $\mathbb{Q}$ .
Computational Feasibility: By combining theoretical reduction with explicit unit calculations, the authors demonstrate a practical pathway to verifying the conjecture for large classes of number fields, potentially extendable to other groups (e.g., $A_4$ , $S_4$ ) if explicit units can be found.

In summary, Ferri and Johnston bridge abstract representation theory with explicit arithmetic geometry to resolve a long-standing gap in the understanding of Leopoldt's conjecture for non-abelian totally real fields.