Greenberg’s conjecture and Inverse Galois problem: metacyclic-nonmodular group of type 1 whose abelianization is ℤ / 2 ​ ℤ × ℤ / 2 m ​ ℤ \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{m}\mathbb{Z} , m ≥ 2 m\geq 2

Mohamed Mahmoud Chems-Eddin, Hamza El Mamry · 2025

Read Greenberg’s conjecture and Inverse Galois problem: metacyclic-nonmodular group of type 1 whose abelianization is ℤ / 2 ​ ℤ × ℤ / 2 m ​ ℤ \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{m}\mathbb{Z} , m ≥ 2 m\geq 2 by Mohamed Mahmoud Chems-Eddin, Hamza El Mamry (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-2508-18402.txt.

Introduction

Mohamed Chems-Eddin and Hamza El Mamry ask which symmetry groups can arise over certain number fields. They target metacyclic groups whose abelian shadow is a product of two cyclic 2-groups. The fields in play are number fields of 2-power degree and their cyclotomic towers. The question is whether such a group can govern the largest unramified 2-extension.

New techniques for Greenberg's conjecture on class numbers come along the way. Real quadratic fields built from four primes get a close look. Real biquadratic and multiquadratic fields built from three odd primes are treated too. The work ties the inverse Galois question to deep conjectures in the area.

Catalog record

CREATORMohamed Mahmoud Chems-Eddin, Hamza El Mamry
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-18402

Browse the full library · About the Tower of Babel

Full text

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

Greenberg’s conjecture and Inverse Galois problem: metacyclic-nonmodular group of type 1 whose abelianization is ℤ / 2 ​ ℤ × ℤ / 2 m ​ ℤ \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{m}\mathbb{Z} , m ≥ 2 m\geq 2 Mohamed Mahmoud Chems-Eddin, Hamza El Mamry Abstract . For an integer m ≥ 2 m\geq 2 , we aim to investigate the realizability of types of metacyclic-nonmodular groups, whose abelianization is ℤ / 2 ​ ℤ × ℤ / 2 m ​ ℤ \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{m}\mathbb{Z} , as the Galois group of the maximal unramified 2 2 -extension (resp. pro- 2 2 -extension) of certain number fields of 2 2 -power degree (resp. cyclotomic ℤ 2 \mathbb{Z}_{2} -extensions). Furthermore, we show up some new techniques for studying Greenberg’s conjecture for some number fields. In particular, the reader can find results concerning the real quadratic fields F = ℚ ⁡ ( η ​ q ​ r ​ s ) F=\mathbb{Q}(\sqrt{\eta qrs}) , the real biquadratic fields K = ℚ ⁡ ( η ​ q , r ​ s ) K=\mathbb{Q}(\sqrt{\eta q},\sqrt{rs}) , with η ∈ { 1 , 2 } \eta\in\{1,2\} , and the Fröhlich multiquadratic fields of the form 𝔽 = ℚ ⁡ ( q , r , s ) \mathbb{F}=\mathbb{Q}(\sqrt{q},\sqrt{r},\sqrt{s}) , where q q , r r and s s are odd primes numbers. 2010 Mathematics Subject Classification 20D15, 11R27, 11R04, 11R23, 11R29, 11R37. 1. Introduction Let n\geq 1 be a natural number and G be a finite metabelian 2 -group of rank 2 such that G^{ab}:=G/G^{\prime}\simeq\mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{n}\mathbb{Z} , where G^{\prime}:=[G,G] is the derived subgroup of G and so G^{ab} is the abelianization of G . The study of the realizability of these 2 -groups as the Galois group of the maximal unramified extension of quadratic fields was developed by Kisilevsky, Benjamin and Snyder (cf. [ 6 , 7 , 8 , 9 , 19 ] ). For n=1 , it is known that G is in one of the following structures: Klein four-group, the quaternion, dihedral or semidihedral groups (cf. [ 19 ] ). The study of the realizability of these 2 -groups as the Galois group of the maximal unramified pro- 2 -extension of certain cyclotomic \mathbb{Z}_{2} -extensions was motivated by Mizusawa in his thesis (cf. [ 24 ] ). Subsequently, this has been extensively investigated by several authors in previous years (cf. [ 23 , 25 , 26 ] and references cited therein). For readers’ convenience, recall that a \mathbb{Z}_{2} -extension of a number field k is an infinite extension of k denoted by k_{\infty} such that \mathrm{Gal}(k_{\infty}/k)\simeq\mathbb{Z}_{2} , where \mathbb{Z}_{2} is the ring of 2 -adic numbers. For each s\geq 1 , the extension k_{\infty}/k contains a unique field of degree 2^{s} denoted by k_{s} and called the s th layer of the \mathbb{Z}_{2} -extension of k . Furthermore, we have: k=k_{0}\subset k_{1}\subset k_{2}\subset\cdots\subset k_{s}\subset\cdots\subset k_{\infty}=\bigcup_{s\geq 0}k_{s}. In particular, if \mathbb{Q}_{2,n} denote the field \mathbb{Q}(2\cos({2\pi}/{2^{s+2}})) , for all s\geq 1 , then k_{\infty}=\bigcup k_{s} , where k_{s}=k\mathbb{Q}_{2,s} , is called the cyclotomic \mathbb{Z}_{2} -extension of k . The inverse limit A(k_{\infty}):=\varprojlim A(k_{s}) with respect to the norm maps is called the 2 -Iwasawa module of k , where A(k) denotes the 2 -class group of a number field k . For more details we refer the reader to [ 28 , Chapter 13] . For n>1 , it is known that G is generated by two elements a and b and admits a representation by generators and relations as follows: G=\langle a,b\ |\ a^{2}\equiv b^{2^{n}}\equiv 1\text{ mod }G^{\prime}\rangle. The 2 -group G is classified (cf. [ 9 , Sec. 2 and 3] ) in one of the following structures: abelian , modular , metacyclic-nonmodular or nonmetacyclic . For convenience of readers, we recall that a modular group is defined as follows: Definition . A modular group M is a 2 -group of order 2^{m} , with m>3 , that is generated by two elements a and b and admits the following presentation: M=\langle a,b\ |\ a^{2}=b^{2^{m-1}}=1\text{ and }[a,b]=b^{2^{m-2}}\rangle, where [a,b] is the commutator of a and b . Notice that for us a metacyclic-nonmodular is not abelian. The inverse Galois problem for this case was recently investigated by some authors, including Azizi, Rezzougui and Zekhnini, where they interested in the realizability of finite metacyclic 2 -groups G as Galois group of the maximal unramified pro- 2 -extension of the cyclotomic \mathbb{Z}_{2} -extension of some number fields (cf. [ 3 , 4 ] ). However, a finite metacyclic 2 -groups G such that G^{ab}:=G/G^{\prime}\simeq\mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{n}\mathbb{Z} is in one of the following structures: abelian , modular or metacyclic-nonmodular . With this in mind, it is legitimate to ask the following question: Question 1: Are all these different structures of metacyclic 2 -groups realizable as Galois group of the maximal unramified pro- 2 -extension of certain cyclotomic \mathbb{Z}_{2} -extensions ? Moreover, Benjamin and Snyder [ 9 ] showed that metacyclic-nonmodular groups whose abelianization is isomorphic to \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{n}\mathbb{Z} are classified as follows: Proposition ( [ 9 ] , Proposition 2) . Let M be a metacyclic-nonmodular group whose abelianization is isomorphic to \mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{n}\mathbb{Z} , with n>1 , and such that M=\langle a,b\ |\ a^{2}\equiv b^{2^{n}}\equiv 1\text{ mod }M^{\prime}\rangle . Then M admits a presentation in one of the following types, where \alpha , s and k are integers: \begin{array}[]{cl}Type\ 1:&a^{2^{\alpha}}=1,\ b^{2^{n}}=1,\ b^{-1}ab=a^{-1},\ \alpha>1.\\ Type\ 2:&a^{2^{\alpha}}=1,\ b^{2^{n}}=a^{2^{\alpha-1}},\ b^{-1}ab=a^{-1},\ \alpha>1.\\ Type\ 3:&a^{2^{\alpha}}=1,\ b^{2^{n}}=1,\ b^{-1}ab=a^{-1+k2^{s}},\ \alpha>1,\ \alpha>s>1,k\text{ is odd}.\\ Type\ 4:&a^{2^{\alpha}}=1,\ b^{2^{n}}=a^{2^{\alpha-1}},\ b^{-1}ab=a^{-1+k2^{s}},\ \alpha>1,\ \alpha>s>1,k\text{ is odd}.\end{array} This also led us to ask the following second question: Question 2: Are all these four types of metacyclic-nonmodular groups realizable as Galois group of the maximal unramified pro- 2 -extension of certain cyclotomic \mathbb{Z}_{2} -extensions (resp. 2 -extension of some number fields)? In this paper, we investigate these two questions. Our investigations lead us to examine Greenberg’s conjecture for certain multiquadratic fields, focusing in particular on the 4 -rank of their 2 -Iwasawa module. In fact, the main results are summarized as follows: The following result gives a necessary and sufficient condition for a families of real quadratic fields to have a metacyclic-nonmodular pro- 2 -group of Type 1 with \alpha=2 as the Galois group of the maximal unramified pro- 2 -extension. Let \eta be an element of \{1,2\} , h_{2}(d) and \varepsilon_{d} be respectively the 2 -class number and the fundamental unit of a real quadratic number field \mathbb{Q}(\sqrt{d}) . The notation \delta_{a,b} is reserved to the Kronecker delta. We have: Theorem (Theorem 3.9 ) . Let q\equiv 3\pmod{4} and r\equiv s\equiv 5\pmod{8} be prime numbers such that \left(\frac{q}{r}\right)=\left(\frac{q}{s}\right)=(-1)^{\delta_{\eta,2}} , \left(\frac{r}{s}\right)=1 and N(\varepsilon_{sr})=1 . Put F=\mathbb{Q}(\sqrt{\eta qrs}) and \varepsilon_{\eta qrs}=\gamma+\gamma^{\prime}\sqrt{\eta qrs} . Assume that 2^{\delta_{\eta,1}}q(\gamma-1) is not a square in \mathbb{N} . Then, we have: A(F)\simeq A(F_{\infty})\simeq\mathbb{Z}/2\mathbb{Z}\times\mathbb{Z}/2^{m}\mathbb{Z}, where m is such that h_{2}(\eta qrs)=2^{m+1} . Furthermore, \mathrm{Gal}(\mathcal{L}(F_{\infty})/F_{\infty}) , the Galois group of the maximal unramified pro- 2 -extension \mathcal{L}(F_{\infty}) of F_{\infty} , is a metacyclic-nonmodular pro- 2 -group of Type 1 with \alpha=2 if and only if \left(\frac{r}{s}\right)_{4}\not=\left(\frac{s}{r}\right)_{4} . The stability of the 2 -rank of the class group of intermediate fields of the cyclotomic \mathbb{Z}_{2} -extensions of quadratic and biquadratic fields has been extensively investigated, especially in [ 12 , 22 ] . However, the study of the stability of the 4 -rank

Read the complete text