Weighted Fruit Diophantine Equations and Hyperelliptic Curves

Jewel Mahajan, Apeksha Sanghi · 2026

Read Weighted Fruit Diophantine Equations and Hyperelliptic Curves by Jewel Mahajan, Apeksha Sanghi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-27322.txt.

Introduction

Mahajan and Sanghi study a weighted generalisation of the fruit Diophantine equation, a polynomial equation whose integer solutions are the prize. Their equation blends a pure power term, squared terms with weights, a product term, and a constant. Under stated hypotheses on the parameters, they prove there are no integer solutions at all for wide families. Primes in a fixed residue class shape the constant, so solutions vanish outside explicit exceptional residue classes. In the 2026 number theory paper, small cases are worked out completely.

For the smallest prime case the insolvability is total, with no exceptions. Weakening the hypotheses still gives non-existence, now with coprimality restrictions on any hypothetical solution. Bounds for positive solutions follow, fencing in where searching could ever succeed. A family of hyperelliptic curves attached to the equation converts the results, via an analogue of the classic Nagell-Lutz theorem, into statements about rational torsion points.

Catalog record

CREATORJewel Mahajan, Apeksha Sanghi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.27322

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Weighted Fruit Diophantine Equations and Hyperelliptic Curves Jewel Mahajan, Apeksha Sanghi Abstract . We study the weighted fruit Diophantine equation a ​ x d − c ⁡ ( m 2 ​ y 2 + n 2 ​ z 2 ) + x ​ y ​ z − b = 0 ax^{d}-c\bigl(m^{2}y^{2}+n^{2}z^{2}\bigr)+xyz-b=0 , generalising the results established in [ 10 , 20 , 13 ] . Subject to specific hypotheses on the parameters, our main result shows that for any prime l ≡ 3 ( mod 4 ) l\equiv 3\pmod{4} and b = a ​ ( 2 ​ c ​ m ​ n ) d − l ​ c s ​ t 2 ​ q b=a(2cmn)^{d}-l\,c^{s}t^{2q} , the above equation has no integer solutions except for certain residue classes of x x modulo 4 ​ l 4l . An analogous result also holds when l l is replaced by an odd power of l l in the definition of b b . We prove some insolvability results for l = − 1 l=-1 . By applying the main result to the small values of l l , such as l ∈ { 3 , 7 , 11 , 19 } l\in\{3,7,11,19\} , we explicitly determine the exceptional residue classes outside of which the equation has no solutions. In particular, for l = 3 l=3 , this yields complete insolvability, and weakening these hypotheses still yields non-existence results, though with specific coprimality restrictions on any possible solutions. We also consider a more general variant of the above Diophantine equation and provide some insolvability results. Subsequently, we establish bounds for the positive solutions of the aforementioned equation. Finally, by associating a family of hyperelliptic curves with the equation under consideration and applying Grant’s analogue of the Nagell–Lutz theorem [ 6 ] , we translate these insolvability results into results about their rational torsion points. 2020 Mathematics Subject Classification Primary 11D41, 11G30; Secondary 11D72, 11A15, 11G05 1. Introduction Diophantine equations have a remarkably rich literature in number theory. A significant part of this research focuses on determining whether a given Diophantine equation admits integer solutions. This area of study also shares deep connections with algebraic geometry, particularly concerning integral points on elliptic (resp., hyperelliptic) curves via the Nagell–Lutz theorem (resp., Grant’s theorem). Consider the Diophantine equation (1) y^{2}-xyz+z^{2}=x^{3}-5. Majumdar and Sury [ 10 ] proved that ( 1 ) has no integer solution. They also showed that the associated elliptic curve y^{2}-mxy=x^{3}-(m^{2}+5) has no integral point for any integer m . Equations of a similar form, ax^{3}+by+c-xyz=0 , were previously investigated by Luca and Togbé [ 8 , 19 ] , Subburam and Thangadurai [ 16 , 18 ] , and Subburam and Togbé [ 17 ] . In an effort to generalise existing results in the literature, we consider the Diophantine equation (2) ax^{d}-c\bigl(m^{2}y^{2}+n^{2}z^{2}\bigr)+xyz-b=0. Extending the work of Majumdar and Sury, who studied ( 2 ) for the specific case where a=c=m=n=1 , b=5 , and d=3 , Vaishya and Sharma [ 20 ] proved the insolvability of ( 2 ) when a\equiv 1\pmod{12} , c=m=n=1 , b=8a-3 , and d=3 . This yielded an infinite family of elliptic curves with torsion-free Mordell–Weil group over \mathbb{Q} (see [ 20 , Corollary 1.4] ). Along the same line, further extending the work of Vaishya and Sharma, Prakash and Chakraborty [ 13 ] established the insolvability of ( 2 ) for the case when a\equiv 1\pmod{12} , c=m=n=1 , b=a2^{d}-3 , and d is an odd multiple of 3 . By applying Grant’s hyperelliptic analogue of the Nagell–Lutz theorem [ 6 ] , they obtained infinitely many hyperelliptic curves with torsion-free Mordell–Weil groups. This is consistent with Faltings’ finiteness theorem [ 5 ] and Bhargava’s result [ 1 ] that most hyperelliptic curves over \mathbb{Q} have no rational points. We remark that [ 13 , Theorem 3.3] and [ 20 , Theorem 1.2] are valid only for a=1 , genus g=1 , and when m specified therein is even (see Remark 8.6 ). This article aims to investigate the insolvability of ( 2 ) under some hypotheses on its parameters. The remainder of this article is organised as follows. In Section 2 , we define the sets \mathcal{N}_{1}(l) and \mathcal{N}_{2}(l) (see Definition 2.1 ) and establish several propositions and lemmas for later use. In Section 3 , assuming the parameters a,c,d,m,n,q,s,t satisfy the hypotheses of Theorem 3.1 , we establish the insolvability of the generalised Diophantine equation ( 2 ) for a prime l\equiv 3\pmod{4} and parameter b=a(2cmn)^{d}-lc^{s}t^{2q} . Our proof generalises methods from previous work by analysing the residue classes of x modulo 4l , with the careful observation made in Theorem 3.1 (d) . A significant portion of the proof of Theorem 3.1 remains valid even if l is replaced by an odd power of l in the definition of b . This yields a similar insolvability result, with the only exception that integer solutions (x,y,z) with x\equiv 2cmn\pmod{l} cannot be ruled out (see Corollary 3.3 ). Subsequently, we apply Theorem 3.1 to some small values of l in Section 4 , such as l\in\{3,7,11,19\} , and explicitly determine the exceptional residue classes outside of which ( 2 ) has no solutions. In particular, for l=3 , this yields complete insolvability (see Theorem 4.3 ), while for the small primes l=7,11 , and 19 , we list the set of residues modulo 28,44 , and 76 , respectively, and exhibit solutions in the exceptional classes, showing that Theorems 4.6 , 4.8 and 4.10 are sharp. Similar ideas also yield some insolvability results for l=-1 . In subsequent sections, we investigate ( 2 ) under various hypotheses, establishing a connection between an associated family of hyperelliptic curves and the insolvability of ( 2 ). In Section 5 , we explore a variant of Theorem 4.3 under weakened hypotheses. We observe that non-existence results continue to hold, though with specific coprimality restrictions on any potential solutions (see Theorem 5.1 ). Section 6 considers a more general variant of ( 2 ), where the coefficients of y^{2} and z^{2} are independent, specifically allowing them to be squarefree, and we also provide some insolvability results depending on the residue of a modulo 4 . In Section 7 , we focus on positive solutions to ( 2 ). In particular, under the hypothesis on the parameters of ( 2 ), we show that there are only finitely many positive integer solutions (x,y,z) satisfying x\leq 2cmn . Finally, in Section 8 , we associate a family of hyperelliptic curves with ( 2 ) and analyse their rational torsion points. As a consequence, Theorem 4.3 yields a family of elliptic curves with torsion-free Mordell–Weil groups (see Theorem 8.5 ). We believe that this framework of connecting rational torsion points to the solution of certain Diophantine equations could possibly be extended to superelliptic curves, assuming an analogue to the Grant or Nagell-Lutz theorem exists in that setting. We leave this direction for future work. 2. Preliminaries Throughout this article, \bigl(\frac{\cdot}{\cdot}\bigr) denotes the Jacobi symbol. Note that this reduces to the Legendre symbol when the denominator is an odd prime p . This section establishes the preliminary definitions, lemmas, and propositions necessary

Read the complete text