Sur le sous-groupe des éléments de hauteur infinie du K2 d'un corps de nombres

Jean-François Jaulent,Florence Soriano-Gafiuk

DOI: https://doi.org/10.48550/arXiv.0801.0916

2008-01-07

Abstract:By using the logarithmic approach of the classical kernels for the K2 of number fields, we compute the 2-rank of the wild kernel WK2(F) and the 2-rank of the subgroup of infinite heigh elements in K2(F) in terms of positive class groups for any number field F.

Number Theory

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to determine the 2 - rank of the wild kernel \(WK_2(F)\) and the subgroup of elements of infinite height \(K^\infty_2(F)\) in the number field \(F\). Specifically, the author calculates the 2 - rank of these two subgroups by using positive class groups, and combines the criterion on the equality of these two subgroups previously obtained by K. Hutchinson and the logarithmic description of the wild kernel to provide a simple logarithmic characteristic to determine in which cases \(K^\infty_2(F)\) is a direct factor of the wild kernel \(WK_2(F)\).

### Main problem decomposition

1. **Determine 2 - rank**:
- For any number field \(F\), calculate the 2 - rank of the wild kernel \(WK_2(F)\) and the subgroup of elements of infinite height \(K^\infty_2(F)\).

2. **Combine existing results**:
- Combine K. Hutchinson's criterion on the equality of these two subgroups and the logarithmic description method to give a logarithmic characteristic.

3. **Determine direct factor conditions**:
- Determine in which cases \(K^\infty_2(F)\) is a direct factor of the wild kernel \(WK_2(F)\).

### Specific formulas and symbols

- \(WK_2(F)\): The wild kernel of the number field \(F\).
- \(K^\infty_2(F)=\bigcap_{n\geq1}K^n_2(F)\): The subgroup of elements of infinite height.
- \(C\ell^{\text{pos}}_F\): The positive class group.
- \(\tilde{C}\ell^{\text{pos}}_F\): The subgroup of degree zero in the positive class group.
- \(\mathbb{Z}_2\): The ring of 2 - adic integers.
- \(\otimes\): Tensor product.

### Main conclusions

According to the main theorem of the paper:

1. **Non - exceptional cases** (i.e., \(i\in F^c\)):
\[
K^\infty_2(F)/K^\infty_2(F)^2 = WK_2(F)/WK_2(F)^2\cong\{\pm1\}\otimes_{\mathbb{Z}_2}\tilde{C}\ell_F
\]
At this time, \(WK_2(F) = K^\infty_2(F)\).

2. **Exceptional cases** (i.e., \(i\notin F^c\)):
\[
WK_2(F)/WK_2(F)^2\cong\{\pm1\}\otimes_{\mathbb{Z}_2}C\ell^{\text{pos}}_F
\]
\[
K^\infty_2(F)/K^\infty_2(F)^2\cong\{\pm1\}\otimes_{\mathbb{Z}_2}\tilde{C}\ell^{\text{pos}}_F
\]
In this case, \(WK_2(F)=K^\infty_2(F)\) if and only if \(F\) is logarithmically primitive; otherwise, \([WK_2(F):K^\infty_2(F)] = 2\).

In summary, this paper solves the problem of how to use logarithmic characteristics to describe whether \(K^\infty_2(F)\) is a direct factor of \(WK_2(F)\) by introducing the concept of positive class groups.