Abstract:We study the behaviour of the Stark conjecture for an abelian extension K/k of totally real number fields as K varies in a cyclotomic Z_p-tower. We consider possible strengthenings of the natural norm-coherence in the tower of putative solution of the complex conjecture and,especially, the consequences for the analogous p-adic conjecture. More precisely, under two principal assumptions - (i) that these solutions are given by exterior powers of norm-coherent sequences of global (or p-semilocal) units and (ii) that p splits in k - we show how the values of p-adic twisted zeta-functions for K/k are determined at any integer by a group-ring-valued regulator formed from the appropriate Coates-Wiles homomorphisms.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is centered around the generalization and verification of the complex and p - adic variants of Stark's conjecture in abelian extensions. Specifically, the author D. Solomon continues to study the conjecture proposed by H. M. Stark regarding abelian extensions of number fields \(K/k\), and proposes new versions of the conjecture, which use the so - called "twisted zeta functions" in place of the more commonly used L - functions.
The main problems can be summarized as follows:
1. **Complex and p - adic versions of Stark's conjecture**:
- The author proposes the complex and p - adic versions of Stark's conjecture (referred to as C1 and C2 respectively) using the twisted zeta function at \(s = 1\).
- These conjectures are expressed through group - ring - valued regulators, similar to the method used to formulate the integral variant of Stark's conjecture at \(s = 0\).
2. **Combined conjecture**:
- A combined version of the conjecture (C3) is proposed, which links the complex and p - adic versions of the conjecture, especially in the case where \(K\) is a totally real projective class field.
3. **Existence of special elements**:
- Under appropriate conditions, C1 and C2 predict the existence of "special" elements in the \(r\) - th exterior power over \(Q\otimes\mathbb{Z}E(K)\) (where \(r:=[k:\mathbb{Q}]\) and \(E(K)\) represents the unit group of \(K\)).
- The behavior of these elements is considered as \(K\) varies along a cyclotomic \(\mathbb{Z}_p\)-tower \(K_\infty=\bigcup_{n = 0}^{\infty}K_n\).
4. **Properties P1 and P2**:
- Two new properties P1 and P2 are introduced, which describe the existence and consistency of solutions to the complex and p - adic versions of Stark's conjecture in the cyclotomic \(\mathbb{Z}_p\)-tower.
- P1 requires the existence of an element \(\eta\) such that for all \(n\geq1\), \(\eta\) satisfies the complex version of Stark's conjecture C1 at each level.
- P2 focuses on the behavior of the p - adic version of Stark's conjecture C2 in the \(\mathbb{Z}_p\)-extension.
5. **Higher - order p - adic regulators and special - value conjecture**:
- A set of "higher - order" p - adic regulators \(\check{R}_{t,n}\) are constructed and defined by replacing the usual p - adic logarithm with the twisted generalized Coates - Wiles homomorphism.
- The relationship between these regulators and the p - adic twisted zeta function is explored, and a new special - value conjecture is proposed for the values of the p - adic twisted zeta function at arbitrary integers \(s\).
6. **Theoretical background and comparison**:
- The paper also explores the connections between these conjectures and the Leopoldt - Kubota p - adic L - function, Iwasawa theory, and Perrin - Riou theory.
In summary, the main objective of this paper is to further understand the behavior of Stark's conjecture in the complex and p - adic settings by introducing new conjectures and properties, and to explore the behavior of these conjectures in the cyclotomic \(\mathbb{Z}_p\)-tower. This not only helps to deepen the understanding of Stark's conjecture itself, but also provides new research directions and tools for the field of number theory.
### Key formulas
- **Complex version of Stark's conjecture C1**:
\[
\Phi_{g,T}(1)=\frac{2^r}{\sqrt{d_k}}\prod_{p\in T}N_pR_{k(g)/k}(\eta)\quad\text{for some }\eta\in\bigwedge^r_{Q}