🔢 mathematics

Monogenity of pure quintic fields: the power of sieving

This paper presents a simple algorithm for calculating all generators of power integral bases in pure quintic fields by combining standard techniques like Baker's method and LLL-reduction with a novel, fast sieving method that efficiently eliminates small exponents.

Original authors: István Gaál

Published 2026-05-06

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: István Gaál

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 trying to solve a massive, multi-dimensional puzzle. The puzzle pieces are numbers, and the goal is to find a very specific arrangement where everything fits together perfectly to form a "Power Integral Basis." In the world of mathematics, finding this perfect arrangement is like finding the "master key" that unlocks the entire structure of a specific type of number field (a pure quintic field).

For a long time, mathematicians knew how to solve this puzzle for smaller, simpler fields (like cubic or quartic ones). But when they tried to apply the same logic to quintic fields (which involve fifth-degree equations), the puzzle became impossibly huge. Previous attempts took computers running for 8 hours just to check one case, and for even more complex fields, it took two months. It was like trying to find a single specific grain of sand on a beach by picking up every single grain one by one.

The Problem: The "Combinatorial Explosion"

The core difficulty was a math problem called the "Index Form Equation." To solve it, the researchers had to test millions upon millions of combinations of numbers (exponents). Even after using powerful mathematical shortcuts to narrow down the search, they were still left with about 6 billion combinations to check.

Trying to check 6 billion combinations one by one is like trying to read every book in a library to find one sentence, even if you know the sentence is in there. It's too slow and inefficient.

The Old Way: The "Ellipsoid Method"

In previous attempts, mathematicians used a complex technique called the "ellipsoid method" to try and find these combinations. Think of this as trying to squeeze a giant, wobbly balloon into a tiny box to see what fits inside. It was complicated, slow, and prone to getting stuck.

The New Idea: The "Sieve"

The author of this paper, István Gaál, introduced a clever new trick: Sieving.

Imagine you have a bucket of mixed nuts and bolts, and you need to find only the bolts. Instead of picking up every single item and inspecting it (which is slow), you pour the bucket through a series of sieves (screens with holes).

The First Sieve: You pour the mix through a screen with large holes. Most of the nuts (the wrong answers) fall through or get caught immediately.
The Second Sieve: You take the remaining items and pour them through a finer screen.
The Result: By the time you get to the end, you have very few items left to inspect closely.

In this paper, the "nuts and bolts" are the billions of number combinations. The "sieves" are specific prime numbers. The author realized that by checking the combinations against these prime numbers first, they could instantly eliminate 99.9% of the wrong answers without doing the heavy math required for the final check.

The Secret Ingredient: A Special Shortcut

The paper reveals that this sieve method only works for quintic fields if you use a specific "secret ingredient." The author discovered that in these specific number fields, there is a hidden symmetry related to a smaller "quadratic subfield" (a simpler number system nested inside the complex one).

By exploiting this symmetry, the author was able to reduce the problem from checking 5 numbers at a time to checking 4 numbers at a time.

Before: Checking 5 numbers meant $274$ million possibilities.
After: Checking 4 numbers meant only $274$ million possibilities? No, wait—the math shows it reduced the total work from roughly 12 billion checks down to 274 million.

It's like realizing that in a game of hide-and-seek, you don't need to check every room in the house because you know the person is hiding in a specific wing. You can ignore the rest of the house entirely.

The Results: From Months to Minutes

By combining the standard mathematical tools (Baker's method and LLL-reduction, which are like rough maps to narrow down the search area) with this new "sieving" technique, the results were dramatic:

Old Method: 8 hours per field.
New Method: 3 to 9 minutes per field.

The author tested this on ten different number fields. In every case, the computer found all the correct "master keys" (generators) in under 10 minutes. The computer didn't just guess; it systematically eliminated the wrong answers using the sieve until only the true solutions remained.

Summary

This paper is about finding a faster way to solve a very difficult number puzzle. The author realized that instead of brute-forcing through billions of possibilities, you can use a "sieve" to filter out the wrong answers almost instantly. By using a special property of these specific number fields, the author turned a task that used to take days or months into a task that takes minutes, making it possible to solve problems that were previously considered too difficult for practical calculation.

Technical Summary: Monogenity of Pure Quintic Fields

Problem Statement
The paper addresses the problem of determining all generators of power integral bases in pure quintic fields $K = \mathbb{Q}(\xi)$ , where $\xi$ is a root of an irreducible polynomial $f(x) = x^5 - m$ . A number field is monogenic if its ring of integers $\mathbb{Z}_K$ admits a basis of the form $(1, \xi, \dots, \xi^{n-1})$ . Finding such generators is equivalent to solving the index form equation $I(\alpha) = 1$ , where $\alpha \in \mathbb{Z}_K$ . While efficient algorithms exist for cubic and quartic fields (reducible to Thue equations), and for certain higher-degree fields with quadratic subfields, solving index form equations in general pure quintic fields has historically been computationally prohibitive. Previous attempts, such as those by the author in reference [4], required approximately 8 hours of CPU time for quintic cases due to the difficulty of enumerating small exponent tuples after reduction.

Methodology
The proposed algorithm integrates standard techniques from algebraic number theory with a novel sieving strategy:

Baker's Method and Reduction: The process begins by establishing an upper bound ( $A_B \approx 10^{34}$ ) for the exponents $a_h$ in the unit representation of the index form components. This is achieved using linear forms in logarithms (Baker's method) and subsequent reduction via the LLL algorithm. This step reduces the bound to a manageable range ( $A_R \approx 40\text{--}50$ ).
The Enumeration Challenge: Even with a reduced bound $A_R$ , the number of exponent tuples $(a_1, \dots, a_5)$ to check is $(2A_R + 1)^5$ . For $A_R = 45$ , this exceeds $6 \times 10^9$ , rendering brute-force enumeration infeasible within reasonable timeframes.
Novel Sieving Strategy: The core contribution of the paper is a new sieving method designed to drastically reduce the search space before full enumeration.
- Structural Observation: The author exploits the structure of the degree 10 fields $L_{i,j} = \mathbb{Q}(\xi^{(i)} + \xi^{(j)}, \xi^{(i)}\xi^{(j)})$ . It is shown that one of the fundamental units in these fields originates from the quadratic subfield $\mathbb{Q}(\sqrt{5})$ .
- Reduction of Variables: By selecting a specific cyclic sequence of indices $(1,2), (2,3), (3,4), (4,5), (5,1)$ , the author constructs an identity (Equation 11) where the unit corresponding to the quadratic subfield cancels out. This allows the problem to be reformulated such that the exponent $a_1$ (associated with the quadratic unit) is eliminated from the primary congruence test.
- Modular Sieving: Instead of testing all $(2A_R + 1)^5$ tuples, the algorithm tests only $(2A_R + 1)^4$ tuples modulo a set of small primes $p$ . The polynomial $x^5 - m$ is chosen to split into linear factors modulo $p$ , allowing the conjugates of the units and field elements to be computed as integers.
- Verification: Only tuples surviving the modular tests for five distinct primes are subjected to the final verification steps: checking the full identity involving $a_1$ and solving the resulting linear system for integer coefficients.

Key Results
The algorithm was implemented and tested on ten pure quintic fields generated by $x^5 - m$ for the smallest positive integers $m$ (from 2 to 17).

Efficiency: The new method reduced the CPU time from the previously observed 8 hours to approximately 3 to 9 minutes per field on a standard PC (2.5 GHz).
Search Space Reduction: For a typical case with $A_R = 45$ , the sieving reduced the number of fraction calculations from $\approx 1.2 \times 10^{10}$ (for 5 variables) to $\approx 2.7 \times 10^8$ (for 4 variables), a reduction of two orders of magnitude.
Solutions Found: The algorithm successfully identified all inequivalent generators of power integral bases for the tested fields. For example, in the case $m=2$ , solutions $(1, 0, 0, 0)$ and $(1, 1, 1, 1)$ were found; for $m=6$ , solutions $(1, 0, 0, 0)$ and $(1, -1, -1, 0)$ were identified.

Significance and Claims
The paper claims that the primary novelty lies in the sieving idea which allows for the efficient testing of remaining small exponents. The author states that this approach "turns out to be surprisingly fast" and makes the solution of index form equations in pure quintic fields feasible within a few minutes, a task previously considered impractical.

The author modestly notes that while the method is presented for monogenic polynomials $x^5 - m$ , it can be extended to non-monogenic cases by introducing a common denominator. The paper concludes that this new idea is likely to have "many further applications" in similar cases involving the enumeration of exponent tuples in higher-degree number fields, though it does not specify other specific fields or problems beyond the general class of "similar cases."