Normal bases and \(\lambda\)-invariants of number fields (Q1185134)

The authors use the theory of normal bases of \(S\)-integers in \(\mathbb{Z}_ p\)-extensions to prove the following: ``Let \(F\) be an imaginary quadratic field, \(p\) an odd prime number, \(k^ +\) the maximal real subfield of \(k=F(\zeta_ p)\), \(A^ +\) the \(p\)-primary part of the ideal class-group of \(k^ +\), \(S\) the set of all prime ideals of \(F\) each of which has only one prime factor in \(k(\zeta_{p^ 2})\). Suppose that \(S\) contains all prime ideals above \(p\) and that the component \(e_ 1(A^ +)\) is non-trivial, where \(e_ j\) is the usual \(j\)-th idempotent of \(\mathbb{Z}_ p[\text{Gal}(k/F)]\). Assume moreover that there exists a \(\mathbb{Z}_ p\)- extension \(K\) of \(F\) such that \(K\cap F_ \infty=F\) and all intermediate fields of \(K/F\) have normal bases of \(S\)-integers. Then the Iwasawa invariant \(\lambda^ +\) attached to the cyclotomic \(\mathbb{Z}_ p\)- extension \(k^ +_ \infty\) is non-zero.'' This could give a counter-example to Greenberg's conjecture on the nullity of \(\lambda^ +\), but any such hope is killed by a close scrutiny of the main lines of the proof, which proceeds as follows: the existence of \(K\) as above implies the nullity of \(e_ 1(\text{Ker}(A_ 0\to A_ n))\), where \(A_ n=p\)-primary part of the class-group of \(k_ n\) (this is shown in the paper by a string of lemmas computing norms of units, but is also a consequence --- for all number fields \(F\), not just imaginary quadratic --- of the main exact sequence of \textit{V. Fleckinger} and the reviewer [Manuscr. Math. 71, 183-195 (1991; Zbl 0732.11063)]; then \(e_ 1(A^ +_ 0)\neq 0\) implies \(e_ 1(A^ +_ \infty)\neq 0\); but \(\mu^ +=\lambda^ +=0\) iff \(A^ +_ \infty=0\), by a result of Greenberg.