🔢 mathematics

Rational 2-Cycles for $x^3+bx+a$ and the Elliptic Family $Y^2=X^3+4X^2+16t^2$

This paper establishes a birational correspondence between rational 2-cycles of the cubic family $x^3+bx+a$ and rational points on a specific elliptic curve, proving that the curve has positive rank for all nonzero $a \in \mathbb{Q}$ and thus guaranteeing infinitely many rational values of $b$ yielding such cycles, while also characterizing the curve's torsion structure and excluding specific rational isogenies and point orders.

Original authors: Sompong Chuysurichay, Sawian Jaidee, Chatchawan Panraksa, Teerapol Sukhonwimolmal

Published 2026-06-16

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Sompong Chuysurichay, Sawian Jaidee, Chatchawan Panraksa, Teerapol Sukhonwimolmal

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 mystery about numbers. Specifically, you are looking at a family of mathematical machines called cubic polynomials. Think of these machines as simple factories: you feed them a number ( $x$ ), they crunch it through a specific formula ( $x^3 + bx + a$ ), and spit out a new number.

The big question this paper asks is: Can we find a "loop" in these machines?

The Mystery: The 2-Cycle Loop

In this story, a "2-cycle" is a special kind of loop. Imagine you put a number into the machine, and it spits out a second number. If you put that second number back in, the machine spits out your original number. You are now stuck in a perfect, endless dance between two numbers.

The authors are asking: If we fix one part of the machine's formula (let's call it $a$ ), can we tweak the other part (let's call it $b$ ) to create this loop using only rational numbers (fractions like 1/2 or 3/4, rather than messy decimals or square roots)?

The Detective's Tool: The Elliptic Curve Map

The authors discovered that finding these loops is incredibly hard if you look at the polynomial directly. It's like trying to find a needle in a haystack by staring at the hay.

Instead, they realized there is a secret map. They found a way to translate the problem of finding these loops into a different, well-studied world called Elliptic Curves.

The Analogy: Think of the polynomial as a complex, winding maze. The Elliptic Curve is a straight, paved highway that runs parallel to the maze. If you can find a specific spot on the highway, you can instantly know where the loop is in the maze.
For every fixed value of $a$ , there is a specific highway (an elliptic curve) defined by the equation $Y^2 = X^3 + 4X^2 + 16a^2$ .

The Big Discovery: Infinite Possibilities

The authors proved a stunning fact about this highway. They found a specific "landmark" on the curve (a point they call $P_t$ ) that has infinite order.

What does "infinite order" mean? Imagine standing on a number line. If you take a step, you land on a new number. If you take another step, you land on another. If you have "infinite order," it means you can keep taking steps forever, and you will never land on the same spot twice, and you will never return to where you started.
The Result: Because this landmark never repeats, it generates an infinite number of unique points on the highway. Since every point on the highway corresponds to a valid setting for $b$ , this proves that for any fixed $a$ , there are infinitely many different values of $b$ that create a rational 2-cycle.

In simple terms: No matter how you set up the first part of your machine, there are endless ways to tune the second part to make it dance in a perfect two-step loop using fractions.

The Side Quest: Checking for "Bad" Loops

While the main discovery was about finding these infinite loops, the authors also did a very thorough "safety inspection" of the highway itself. They wanted to know if there were any other, shorter, or "weird" loops hidden in the structure of these curves.

They checked to see if the curves could have loops of specific lengths (like 3 steps, 5 steps, or 7 steps).

The Method: They used powerful computer algebra systems (like Magma and SageMath) to build complex, high-dimensional shapes (genus-3 curves) that would only exist if these "bad" loops were possible.
The Result: They proved that for almost all settings, these "bad" loops do not exist. The curves are "clean" of these specific types of torsion (repeating patterns). This part of the paper is like a rigorous engineering stress test, confirming that the mathematical structure is stable and behaves exactly as predicted, with no hidden surprises.

Summary

The Problem: Can we find two numbers that swap places when fed into a specific type of math machine?
The Trick: Translate the problem into a map of an Elliptic Curve.
The Main Finding: There is a "magic point" on this map that never repeats. This proves there are infinitely many ways to tune the machine to create a perfect 2-step loop.
The Safety Check: They also proved that the machine doesn't accidentally create other weird, repeating patterns of length 3, 5, or 7.

The paper is a triumph of connecting two different areas of math (dynamics and geometry) to show that a specific mathematical phenomenon is not just possible, but abundant.

Technical Summary: Rational 2-Cycles for $x^3 + bx + a$ and the Elliptic Family $Y^2 = X^3 + 4X^2 + 16t^2$

Problem Statement
The paper investigates the existence and arithmetic properties of rational 2-cycles for the cubic polynomial family $f_{b,a}(x) = x^3 + bx + a$ over the rational numbers $\mathbb{Q}$ , where $a$ and $b$ are fixed rational parameters. Specifically, for a fixed nonzero $a \in \mathbb{Q}$ , the authors seek to determine the set of rational values $b$ for which the map $f_{b,a}$ admits a rational point of exact period 2. A rational 2-cycle is defined as an ordered pair $(x_1, x_2) \in \mathbb{Q}^2$ with $x_1 \neq x_2$ such that $f_{b,a}(x_1) = x_2$ and $f_{b,a}(x_2) = x_1$ .

Methodology
The authors employ a combination of arithmetic dynamics, the theory of elliptic curves, and explicit computational number theory. The core methodology involves:

Reduction to Elliptic Curves: The authors utilize the classification of cubic polynomials over $\mathbb{Q}$ into two normal forms. While Normal Form I (odd polynomials) yields a rational period-2 locus, Normal Form II (general polynomials) yields an elliptic moduli space. The family $f_{b,a}$ is shown to be linearly conjugate over $\mathbb{Q}$ to the "square-coefficient slice" of Normal Form II. This reduces the dynamical problem to the arithmetic of a specific one-parameter family of elliptic curves:
$E_t: Y^2 = X^3 + 4X^2 + 16t^2$
where the specialization $t=a$ corresponds to the dynamical parameter $a$ .
Birational Dictionary: An explicit birational map is constructed between rational points on the specialized curve $E_a$ (excluding degenerate loci where $X=0$ or $Y=0$ ) and rational 2-cycles of $f_{b,a}$ . This dictionary allows the translation of dynamical questions into questions about rational points on elliptic curves.
Division Polynomials and Mazur's Theorem: To prove the existence of infinitely many such parameters $b$ , the authors analyze the Mordell–Weil group of the generic family $E_t$ . They focus on a distinguished section $P_t = (0, -4t)$ . Using explicit evaluations of division polynomials at $P_t$ , $[2]P_t$ , and $[4]P_t$ , they prove that $P_t$ has infinite order for every $t \in \mathbb{Q}^\times$ . This proof relies solely on Mazur's theorem regarding possible torsion orders over $\mathbb{Q}$ and does not depend on the subsequent torsion analysis of the family.
Torsion Analysis via Modular Curves: Independently of the dynamical application, the paper conducts a finer arithmetic study of the torsion subgroup of $E_t(\mathbb{Q})$ .
- 2-Torsion: The locus of parameters $t$ admitting a rational point of order 2 is explicitly determined.
- Odd Torsion (3, 5, 7): The authors prove that for $t \neq 0$ $t \neq = 0$ , $E_t$ $E_{t}$ admits no rational points of order 3, 5, or 7.
  - Order 3 is excluded by reducing the problem to the Mordell–Weil group of an auxiliary elliptic curve of rank 0.
  - Order 5 is excluded by analyzing the fiber product of the $j$ -map with the modular curve $X_0(5)$ . This results in a genus-3 curve $C_5$ . Using a Prym variety construction (Ritzenthaler–Romagny), the authors show the relevant Jacobian has rank 0 and determine the rational points on $C_5$ .
  - Order 7 is excluded by analyzing the fiber product with $X_1(7)$ , leading to a hyperelliptic genus-3 curve $C_7$ . The rational points on $C_7$ are determined using a two-cover descent (Bruin–Stoll) followed by elliptic Chabauty over the quadratic field $\mathbb{Q}(\sqrt{21})$ .

Key Contributions and Results

Explicit Dictionary: The paper provides explicit formulas relating rational 2-cycles of $x^3 + bx + a$ to rational points on $Y^2 = X^3 + 4X^2 + 16a^2$ .
Infinite Order of a Section: It is proven that the section $P_t = (0, -4t)$ has infinite order on $E_t$ for all $t \in \mathbb{Q}^\times$ . This is established via a self-contained division-polynomial argument that excludes all torsion orders permitted by Mazur's theorem.
Positive Rank and Infinitely Many Parameters: As a direct consequence, for every fixed nonzero $a \in \mathbb{Q}$ , the specialized curve $E_a$ has positive Mordell–Weil rank. Consequently, there exist infinitely many rational values of $b$ such that $f_{b,a}$ admits a rational 2-cycle.
Torsion Classification: The paper determines the rational torsion structure of the generic family $E_t$ . It is shown that $E_t(\mathbb{Q})$ contains no rational points of order 3, 5, or 7. The torsion subgroup is either trivial or cyclic of order 2, 4, or 8, with nontrivial torsion occurring only on a specific one-parameter locus defined by $t = u(u^2+4)/4$ .
Certified Computations: The exclusion of torsion of orders 5 and 7 relies on certified computations using Magma on explicit genus-3 curves. The paper provides detailed verification scripts and checksums for reproducibility.

Significance and Claims
The paper claims that its primary novelty lies not in the existence of an elliptic moduli description for cubic 2-cycles (which was previously noted by Ingram), but in the specialized arithmetic analysis of the family arising from the specific form $x^3 + bx + a$ .

Dynamical Conclusion: The work establishes a uniform result: for any fixed nonzero rational $a$ , the family $x^3 + bx + a$ admits rational 2-cycles for infinitely many rational $b$ . This is derived from the infinite order of a specific section on the associated elliptic curve.
Arithmetic Independence: The proof of the infinite order of the section is logically independent of the detailed torsion analysis (Sections 5–8). The torsion analysis is presented as a study of independent interest regarding the generic family $E_t$ , providing a complete picture of its rational torsion subgroups.
Methodological Rigor: The paper emphasizes the use of certified computational methods (division polynomials, Prym varieties, two-cover descent, and elliptic Chabauty) to rigorously exclude specific torsion orders, ensuring that the results are not merely heuristic.

The authors do not claim to solve the general problem of rational preperiodic points for all cubic polynomials, nor do they propose new experimental applications. The work is strictly confined to the arithmetic study of the 2-cycle locus for the specified cubic family and the associated elliptic curve family.