🔢 mathematics

Non-trivial Integer Solutions of $x^r+y^r=Dz^p$

This paper employs the modular method over totally real fields, alongside the Weak Frey–Mazur and Eichler–Shimura conjectures, to demonstrate that infinitely many equations of the form $x^r+y^r=Dz^p$ lack non-trivial primitive integer solutions for a fixed prime $r \geq 5$ when $p$ is sufficiently large.

Original authors: Yasemin Kara, Diana Mocanu, Ekin Özman

Published 2026-08-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yasemin Kara, Diana Mocanu, Ekin Özman

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

Mathematics has long been captivated by equations that look simple on the surface but hide deep, stubborn secrets. Among the most famous of these are problems asking whether whole numbers can be combined in specific ways to equal a power of another number. For centuries, mathematicians have chased a particular type of puzzle: finding whole number solutions to equations where two numbers raised to a high power add up to a third number raised to a different power. While the most famous version of this problem was solved decades ago, a vast family of related equations remains unsolved. These variations involve a coefficient, a multiplier that changes the balance of the equation, and they resist standard methods of proof. The question is not just about finding a single answer, but about understanding the fundamental rules that govern how numbers behave when raised to large powers.

In a recent study, a team of mathematicians has made significant progress on this front, proving that for a wide range of these equations, no non-trivial whole number solutions exist once the power becomes sufficiently large. The researchers focused on equations where two numbers, raised to a fixed high power, sum to a third number multiplied by a constant and raised to a variable, very large power. They demonstrated that if the variable power is large enough, the equation simply cannot be satisfied by any set of whole numbers that are not trivially zero or one. This result does not rely on checking every single possibility one by one, which would be impossible given the infinite nature of numbers. Instead, the team used a sophisticated strategy that connects the world of whole numbers to the geometry of shapes known as elliptic curves.

The approach the authors took is known as the modular method, a powerful technique pioneered in the proof of Fermat's Last Theorem. The process begins by assuming that a solution to the equation does exist. If such a solution were real, it would allow the mathematicians to construct a specific geometric object, an elliptic curve, with very particular properties. This curve acts as a bridge, translating the problem of finding whole numbers into a problem about the behavior of these shapes. The researchers then applied a series of logical steps to show that if the assumed solution existed, the resulting curve would have to match a very specific, limited set of other curves that are already known to mathematicians.

To make this connection, the team worked over a special type of number system called a totally real field, which is a more complex version of the standard number line used in everyday arithmetic. They constructed their geometric curve within this system, ensuring it had the precise mathematical features required to link back to the original equation. A critical part of their work involved analyzing how the curve behaves at specific points of instability, known as primes of bad reduction. By studying these behaviors, they could determine that the curve would have to possess a certain type of symmetry and structure that only a finite number of known curves could possibly have.

The final step of the proof relies on a logical elimination. The researchers showed that the hypothetical curve created from a solution would have to be indistinguishable from one of these few known curves. However, they also demonstrated that the curve derived from a solution would have properties that contradict the known properties of the finite set of candidates. This contradiction implies that the initial assumption—that a solution exists—must be false. The team's argument depends on two widely accepted but unproven ideas in the field: the Weak Frey–Mazur Conjecture and the Eichler–Shimura Conjecture. These conjectures act as guiding principles that allow mathematicians to predict how these geometric shapes relate to one another when the powers involved are large.

The paper establishes two main results based on these principles. First, for a specific type of equation where the fixed power leaves a remainder of three when divided by four, the authors proved that no solutions exist for large enough variable powers, relying only on the Weak Frey–Mazur Conjecture. Second, for cases where the fixed power leaves a remainder of one, they reached the same conclusion, but this required assuming both the Weak Frey–Mazur Conjecture and the Eichler–Shimura Conjecture. In both scenarios, the proof shows that there is a threshold for the variable power; once the power exceeds this threshold, the equation becomes impossible to solve with non-trivial whole numbers.

The significance of this work lies in its ability to clear an infinite number of cases at once. Rather than solving one equation at a time, the authors provided a method that rules out solutions for infinitely many variations of the problem simultaneously. While the results depend on unproven conjectures, these are standard assumptions in modern number theory that have withstood decades of scrutiny. The findings represent a major step forward in understanding the rigid structure of numbers, confirming that for these specific types of equations, the universe of whole numbers simply does not contain the answers mathematicians might hope to find. The work highlights how deep connections between different areas of mathematics can be used to solve problems that seem entirely out of reach when viewed in isolation.

Technical Summary: Non-Trivial Integer Solutions of $x^r + y^r = D z^p$

Problem Statement
This paper investigates the generalized Fermat equation of signature $(r, r, p)$ with a coefficient $D$ :
$x^r + y^r = D z^p, \quad x, y, z \in \mathbb{Z}$
where $r \geq 5$ is a fixed rational prime, $p$ is a rational prime distinct from $r$ , and $\gcd(r, D) = 1$ . The authors seek to prove that for sufficiently large $p$ , this equation admits no non-trivial primitive integer solutions (where $\gcd(x,y,z)=1$ and $xyz \neq 0$ ). The study extends the definition of trivial solutions to exclude cases of the form $x=1, D=\pm(1+y^r), z=\pm 1$ .

Methodology: The Modular Method over Totally Real Fields
The authors employ the modular method, a strategy pioneered by Wiles for Fermat's Last Theorem, adapted to elliptic curves over totally real number fields. The approach proceeds in three distinct steps:

Construction of a Frey Elliptic Curve:
Following the recipes of Freitas [Fre15], the authors associate a putative solution $(x,y,z)$ with an elliptic curve $E$ defined over the maximal totally real subfield $K = \mathbb{Q}(\zeta_r + \zeta_r^{-1})$ of the $r$ -th cyclotomic field. The curve is given by the equation:
$E: Y^2 = X(X - A)(X + B)$
where $A, B, C$ are constructed from the factorization of $x^r + y^r$ over $K$ . Specifically, $A = (\alpha_1 - \alpha_2)f_0$ , $B = (\alpha_2 - 2)f_1$ , and $C = (2 - \alpha_1)f_2$ , with $f_i$ being quadratic factors of the sum of powers. The discriminant of $E$ is explicitly given as $\Delta_E = 16(ABC)^2$ , satisfying the necessary ramification properties.
Modularity and Level Lowering:
The authors rely on modularity results for elliptic curves over totally real fields established by Freitas and Siksek [FS15a]. They prove that for sufficiently large $p$ , the Frey curve $E$ is modular and the residual Galois representation $\bar{\rho}_{E,p}$ is irreducible. Using level lowering theorems, they deduce that $\bar{\rho}_{E,p}$ is isomorphic to the representation $\bar{\rho}_{f,p}$ attached to a Hilbert newform $f$ of parallel weight 2 and level $N_p$ , where $N_p$ depends only on $D$ and $r$ , not on the specific solution $(x,y,z)$ .
Elimination via Conjectures:
The final step involves eliminating the possibility that such a newform $f$ exists. The authors transform the problem of eliminating Hilbert newforms into a problem of eliminating elliptic curves with specific conductors. This is achieved through two main theoretical tools:
- Eichler–Shimura Correspondence: In specific cases (detailed in Theorem 3.9), rational Hilbert newforms correspond to elliptic curves over $K$ .
- Weak Frey–Mazur Conjecture: The authors assume this conjecture (Conjecture 3.10), which posits that for a fixed elliptic curve $E_0$ , if $p$ is large enough, any curve $E$ with $\bar{\rho}_{E,p} \cong \bar{\rho}_{E_0,p}$ is isogenous to $E_0$ .

Key Contributions and Results
The paper establishes the non-existence of non-trivial primitive solutions for infinitely many instances of the equation, contingent on standard conjectures. The results are divided based on the congruence of $r$ modulo 4:

Theorem 1.1 (First Main Theorem): Assuming the Weak Frey–Mazur Conjecture:
- If $r \equiv 3 \pmod 4$ , there exists a constant $B_{r,D}$ such that for all $p > B_{r,D}$ , the equation has no non-trivial primitive solutions. In this case, the Eichler–Shimura correspondence holds unconditionally because the degree of $K$ over $\mathbb{Q}$ is odd.
- If $r \equiv 1 \pmod 4$ , there exists a constant $B'_{r,D}$ such that for all $p > B'_{r,D}$ , the equation has no non-trivial primitive solutions where $r \mid z$ . This case requires an additional analysis of the inertia image at the prime $\beta$ above $r$ to satisfy the conditions of Theorem 3.9(2).
Theorem 1.3 (Second Main Theorem): Assuming both the Weak Frey–Mazur Conjecture and the Eichler–Shimura Conjecture (Conjecture 3.8), the result holds for $r \equiv 1 \pmod 4$ without the restriction $r \mid z$ .

The constants $B_{r,D}, B'_{r,D}, B''_{r,D}$ are non-effective, as they depend on the constants inherent in the Weak Frey–Mazur Conjecture.

Significance and Scope
The authors position this work as a completion of the resolution for the generalized Fermat equation $(r,r,p)$ for fixed $r \geq 5$ and large $p$ , filling gaps left by previous unconditional results (summarized in Table 1.1). The paper explicitly states that its primary contribution is twofold:

Providing a proof for these instances by assuming strong but standard conjectures in the field.
Offering a concise survey of the modular method over totally real fields, specifically highlighting the interplay between the Eichler–Shimura correspondence and the Frey–Mazur conjecture in the elimination step.

The authors note that while the approach is familiar to experts, the specific application of the Eichler–Shimura correspondence combined with the Frey–Mazur conjecture to eliminate elliptic curves in this context has not been extensively discussed in existing literature. The work avoids the computational difficulty of explicitly computing all Hilbert newforms at a given level by instead reducing the problem to the study of elliptic curves with specific reduction properties.

More like this