Power integral bases in a family of octic fields

István Gaál · 2025

Read Power integral bases in a family of octic fields by István Gaál (2025) 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-2501-13519.txt.

Introduction

István Gaál wrote this math paper in two thousand twenty-five as the sole author. It asks whether a family of octic number fields, fields from degree-eight polynomials, holds power integral bases beyond the known one.

When a polynomial is monogenic, its root spawns a power integral basis for the field it makes. That leaves open whether other spawns exist in the same field. Gaál takes up a family of octic polynomials where monogenity was shown in recent work by Jones. He hunts for all further generators of power integral bases there. The paper extends his past hunts in smaller families to the octic case.

Catalog record

CREATORIstván Gaál
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-13519

Browse the full library · About the Tower of Babel

Full text

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

Power integral bases in a family of octic fields István Gaál Abstract Several recent results prove the monogenity of some polynomials. In these cases the root of the polynomial generates a power integral basis in the number field generated by the root. A straightforward question is whether such a number field admits other generators of power integral bases? We have investigated this problem in some previous papers and here we extend this research to a family of octic polynomials, following a recent result of L. Jones [ 12 ] . Mathematics Subject Classification: Primary 11Y50, 11R04; Secondary 11D25 Key words and phrases: monogenity; power integral basis; octic fields; relative quartic fields; relative Thue equations 1 Introduction A number field K of degree n with ring of integers {\mathbb{Z}}_{K} is called monogenic (cf. [ 3 ] ) if there exists \xi\in{\mathbb{Z}}_{K} such that (1,\xi,\ldots,\xi^{n-1}) is an integral basis, called power integral basis. We call \xi the generator of this power integral basis. \alpha,\beta\in{\mathbb{Z}}_{K} are called equivalent , if \alpha+\beta\in{\mathbb{Z}} or \alpha-\beta\in{\mathbb{Z}}_{K} . Obviously, \alpha generates a power integral basis in K if and only if any \beta , equivalent to \alpha , does. As it is known, any algebraic number field admits up to equivalence only finitely many generators of power integral bases. An irreducible polynomial f(x)\in{\mathbb{Z}}[x] is called monogenic , if a root \xi of f(x) generates a power integral basis in K={\mathbb{Q}}(\xi) . If f(x) is monogenic, then K is monogenic, but the converse is not true. For \alpha\in{\mathbb{Z}}_{K} (generating K over {\mathbb{Q}} ) we call the module index I(\alpha)=({\mathbb{Z}}_{K}:{\mathbb{Z}}[\alpha]) the index of \alpha . \alpha generates a power integral basis in K if and only if I(\alpha)=1 . If \alpha^{(i)}\;(1\leq i\leq n) are the conjugates of \alpha in K of degree n with discriminant D_{K} , then I(\alpha)=\frac{1}{\sqrt{|D_{K}|}}\prod_{1\leq i<j\leq n}|\alpha^{(i)}-\alpha^{(j)}|. For more details concerning monogenity and power integral bases cf. [ 3 ] . In some recent papers we investigated number fields generated by a root of a monogenic polynomial and made calculations to figure out, whether these fields admit any additional generators of power integral bases. We refer to [ 4 ] for some sextic trinomials, [ 5 ] for pure sextic fields, [ 6 ] for pure octic fields, [ 7 ] for certain quartic trinomials and [ 8 ] for some quartic polynomials with given Galois groups. L. Jones [ 11 ] (see also [ 12 ] ) gave conditions for the monogenity of certain even octic polynomials of type x^{8}+ax^{6}+bx^{4}+ax^{2}+1 . In the present paper we extend our calculations to this type of polynomials. Among others we prove the existence of a non-trivial generator of power integral basis. The octic field, generated by a root of the above polynomial is a quadratic extension of a quartic field. It is an interesting point of our arguments, that this octic field can also be considered as a quartic extension of a quadratic field, which makes it much easier to deal with. All tools used in our calculations are optimized to this special case, in order to make our calculations more efficient. 2 The octic polynomial Let f(x)=x^{8}+ax^{6}+bx^{4}+ax^{2}+1 (1) with a,b\in{\mathbb{Z}} . Set W_{1}=b+2-2a,W_{2}=b+2+2a,W_{3}=a^{2}-4b+8 . L. Jones [ 12 ] proved: Theorem 1 If W_{1}W_{2}W_{3} is square free and (a\mod 4,\;b\mod 4)\in\{(1,3),(3,1),(3,3)\} (2) then the polynomial f(x) in ( 1 ) is monogenic. Assume f(x) is monogenic, not necessarily satisfying ( 2 ). We wonder how many generators of power integral bases the number field K has, generated by a root \alpha of f(x) . Set g(x)=x^{4}+ax^{3}+bx^{2}+ax+1. (3) Obviously, a root \beta of g(x) satisfies \alpha^{2}=\beta . Therefore the octic number field K={\mathbb{Q}}(\alpha) is a quadratic extension of the quartic field L={\mathbb{Q}}(\beta) : {\mathbb{Q}}\overset{4}{\subset}L={\mathbb{Q}}(\beta)\overset{2}{\subset}K={\mathbb{Q}}(\alpha). Unfortunately there exist feasible algorithms for solving index form equations, that is for determining generators of power integral bases, only for a restricted class of number fields. Apart from low degree fields, like cubic and quartic fields, there exist such algorithms only for some higher degree fields with special structure. These are e.g. sextic fields with a quadratic subfield and octic fields with a quadratic subfield (cf. [ 3 ] ). Above we have an octic field with a quartic subfield, but using the reciprocal structure of f(x) and g(x) we can help this problem. If \beta^{4}+a\beta^{3}+b\beta^{2}+a\beta+1=0, then \beta^{2}+a\beta+b+\frac{a}{\beta}+\frac{1}{\beta^{2}}=0, hence \left(\beta^{2}+\frac{1}{\beta^{2}}\right)+a\left(\beta+\frac{1}{\beta}\right)+b=0, \left(\beta+\frac{1}{\beta}\right)^{2}+a\left(\beta+\frac{1}{\beta}\right)+b-2=0. This yields, that \delta=\beta+\frac{1}{\beta} (4) satisfies the quadratic equation \delta^{2}+a\delta+(b-2)=0. (5) Consequently the number field M={\mathbb{Q}}(\delta) is a quadratic subfield of K : {\mathbb{Q}}\overset{2}{\subset}M={\mathbb{Q}}(\delta)\overset{4}{\subset}K={\mathbb{Q}}(\alpha). By ( 4 ) we have \beta^{2}-\delta\beta+1=0, therefore \alpha^{4}-\delta\alpha^{2}+1=0. This means that h(x)=x^{4}-\delta x^{2}+1 (6) is the relative defining polynomial of \alpha over M , and this is what we need for our procedure. Note that in some previous papers (cf. [ 10 ] ) we have developed an algorithm for the complete resolution of index form equations in octic fields with a quadratic subfield. This takes quite a long CPU time, since one has to solve a unit equation in the octic field. Also, it can take long to calculate the fundamental units of that field, which is a necessary input data for the calculations. Therefore, if we would like to have an overall picture about the generators of power integral bases of our octic fields, we have to restrict ourselves to the calculation of the so called "small solutions", that means we calculate all generators of power integral bases having coefficients, say \leq 10^{200} in absolute value in a given integral basis. Since the generators of power integral bases usually have very small coefficients, such an algorithm determines all generators of power integral bases with a very high probability. Moreover it certainly indicates, if a number field, generated by a root of a monogenic polynomial, has also other generators of power integral bases, in addition to the root of the polynomial. Moreover, as we shall see in the following, a crucial point in this algorithm is the resolution of a relative quartic Thue equation over the quadratic subfield. The fast algorithm [ 2 ] for determining "small" solutions of quartic relative Thue equations over quadratic fields is only efficient, if the quadratic subfield is complex. Therefore in our calculations we assume a^{2}-4b+8<0, (7) which guarantees by ( 5 ), that M is a complex quadratic subfield. On the other hand, we shall not restrict ourselves to those monogenic polynomials f(x) , satisfying all conditions of Theorem 1 . We shall run the parameters a,b in certain regions and consider all irreducible polynomials f(x) that are monogenic. The only condition we keep is that W_{3}=a^{2}-4b+8 is square-free, in order to fix the basis element of M and to make our arguments simpler. Note, that we made calculations also for non-squarefree W_{3} , and had comletely the same experiences, including also the non-trivial generator of power integral bases (cf. Theorm 6 ). 3 Integral basis We have M={\mathbb{Q}}(\delta) , with \delta=\frac{-a+\sqrt{a^{2}-4b+8}}{2}. (8) According to the above arguments, we assume W_{3}=a^{2}-4b+8<0 is square-free. To keep usual notation we set m=W_{3} . This number can only be square-free if a\equiv\;\pm 1\;(\bmod\;4) , whence m\equiv\;1\;(\bmod\;4) , therefore the integral basis of the complex quadratic field M={\mathbb{Q}}(\sqrt{m}) is (1,\omega) , where \omega=\frac{1+\sqrt{m}}{2}. We shall make calculations for monogenic polynomials f(x) .

Read the complete text