Table of Contents
Fetching ...

Sums of two units in number fields

Magdaléna Tinková, Robin Visser, Pavlo Yatsyna

TL;DR

This work studies the set $\\mathcal{N}_K = \\{ n \\in \\mathbb{Z}_{>0} : \\exists \\varepsilon,\\delta \\in \\mathcal{O}_K^{\\times} \,\\text{with} \\varepsilon + \\delta = n \\\}$ of positive integers representable as a sum of two units in the ring of integers of a number field $K$. It proves finiteness of $\\mathcal{N}_K$ for number fields not containing a real quadratic subfield, deriving an explicit bound via Amoroso–Viada bounds on unit equations in three unknowns, and it specializes to cubic fields to give a complete classification of unit-sum solutions in both cyclic and complex cubic fields, separating infinite families from finitely many sporadic cases. For cyclic cubic fields $K_a$ and complex cubic fields $L_a$, the paper provides precise descriptions of all solutions to $\\varepsilon + \\delta = n$, identifies the sporadic solutions, and yields explicit $\\mathcal{N}_K$ sets in many instances, including computational tables for several families. Computational methods based on discriminants, minimal polynomials, and Thue equations are developed to determine $\\mathcal{N}_K$ for individual fields, with concrete results and tables for cyclic and complex cubics. The work ends with open problems on averages, extrema, and algorithmic determination of $\\mathcal{N}_K$ and its generalizations to sums of more units.

Abstract

Let $K$ be a number field with ring of integers $\mathcal{O}_K$. Let $\mathcal{N}_K$ be the set of positive integers $n$ such that there exist units $\varepsilon, δ\in \mathcal{O}_K^\times$ satisfying $\varepsilon + δ= n$. We show that $\mathcal{N}_K$ is a finite set if $K$ does not contain any real quadratic subfield. In the case where $K$ is a cubic field, we also explicitly classify all solutions to the unit equation $\varepsilon + δ= n$ when $K$ is either cyclic or has negative discriminant.

Sums of two units in number fields

TL;DR

This work studies the set of positive integers representable as a sum of two units in the ring of integers of a number field . It proves finiteness of for number fields not containing a real quadratic subfield, deriving an explicit bound via Amoroso–Viada bounds on unit equations in three unknowns, and it specializes to cubic fields to give a complete classification of unit-sum solutions in both cyclic and complex cubic fields, separating infinite families from finitely many sporadic cases. For cyclic cubic fields and complex cubic fields , the paper provides precise descriptions of all solutions to , identifies the sporadic solutions, and yields explicit sets in many instances, including computational tables for several families. Computational methods based on discriminants, minimal polynomials, and Thue equations are developed to determine for individual fields, with concrete results and tables for cyclic and complex cubics. The work ends with open problems on averages, extrema, and algorithmic determination of and its generalizations to sums of more units.

Abstract

Let be a number field with ring of integers . Let be the set of positive integers such that there exist units satisfying . We show that is a finite set if does not contain any real quadratic subfield. In the case where is a cubic field, we also explicitly classify all solutions to the unit equation when is either cyclic or has negative discriminant.

Paper Structure

This paper contains 14 sections, 22 theorems, 28 equations, 6 tables.

Key Result

Theorem 1.1

Let $K$ be a degree $d$ number field not containing any real quadratic subfield. Then $\mathcal{N}_K$ is a finite set and satisfies the bound $|\mathcal{N}_K| \leq d \cdot 24^{324(2d+2)} + 2$.

Theorems & Definitions (33)

  • Theorem 1.1
  • Theorem 1.2
  • Theorem 1.3
  • Lemma 2.1
  • proof
  • Proposition 2.2
  • proof
  • Theorem 2.3: AmorosoViada
  • Lemma 3.1
  • Lemma 3.2
  • ...and 23 more