Table of Contents
Fetching ...

On the Exponential Diophantine Equation $(a^n-1)(b^n-1)=x^2$

Armand Noubissie, Alain Togbe, Zhongfeng Zhang

TL;DR

The article investigates the exponential Diophantine equation $(a^n-1)(b^n-1)=x^2$ for fixed $a,b>1$, delivering two main no-solution results. It uses a Pell-equation framework with $a^n-1=Dy^2$ and $b^n-1=Dz^2$, combined with gcd analysis and primitive-divisor theory to derive stringent parity and divisibility constraints. Specifically, it proves no solutions when $a$ is even and $b$ is a prime with $b\equiv3\pmod{8}$, and, under several congruence conditions on $a$ and $b$, shows that the only possible $n$ is $2$, correcting a previous incomplete argument related to Szalay's work. The results extend Szalay’s findings, clarify the role of Pell-type recurrences in this problem, and illustrate how primitive divisors can rule out larger exponents in exponential Diophantine equations.

Abstract

Let $a$ and $b$ be two distinct fixed positive integers such that $\min \{a,b\}>1.$ First, we correct an oversight from \cite{X-Z}. Then, we show that the equation in the title with $b \equiv 3 \pmod 8$, $b$ prime and $a$ even has no solution in positive integers $n, x$. This generalizes a result of Szalay \cite{L}.

On the Exponential Diophantine Equation $(a^n-1)(b^n-1)=x^2$

TL;DR

The article investigates the exponential Diophantine equation for fixed , delivering two main no-solution results. It uses a Pell-equation framework with and , combined with gcd analysis and primitive-divisor theory to derive stringent parity and divisibility constraints. Specifically, it proves no solutions when is even and is a prime with , and, under several congruence conditions on and , shows that the only possible is , correcting a previous incomplete argument related to Szalay's work. The results extend Szalay’s findings, clarify the role of Pell-type recurrences in this problem, and illustrate how primitive divisors can rule out larger exponents in exponential Diophantine equations.

Abstract

Let and be two distinct fixed positive integers such that First, we correct an oversight from \cite{X-Z}. Then, we show that the equation in the title with , prime and even has no solution in positive integers . This generalizes a result of Szalay \cite{L}.

Paper Structure

This paper contains 4 sections, 5 theorems, 15 equations.

Key Result

Theorem 1

Let $a,~b \in \mathbb{N}$ such that $a,~b>1.$ Suppose that one of the following properties is satisfied: Then, the only possible solution of equation eq1 is $n=2.$

Theorems & Definitions (5)

  • Theorem 1
  • Theorem 2
  • Lemma 3
  • Lemma 4
  • Lemma 5