KU-local zeta-functions of finite CW-complexes

TL;DR

This paper introduces a novel $KU$-local zeta-function for finite CW-complexes, extending ideas from algebraic geometry to topology, and demonstrates its analytic properties and relation to $KU$-local stable homotopy groups.

Contribution

It defines and analyzes a new $KU$-local zeta-function for finite CW-complexes, including cases with torsion, and establishes its analytic continuation and connection to homotopy groups.

Findings

The $KU$-local zeta-function admits meromorphic continuation.

Special values recover $KU$-local stable homotopy groups.

Analytic continuation holds under broad conditions.

Abstract

Begin with the Hasse-Weil zeta-function of a smooth projective variety over the rational numbers. Replace the variety with a finite CW-complex, replace etale cohomology with complex K-theory $KU^*$, and replace the $p$-Frobenius operator with the $p$th Adams operation on $K$-theory. This simple idea yields a kind of "$KU$-local zeta-function" of a finite CW-complex. For a wide range of finite CW-complexes $X$ with torsion-free $K$-theory, we show that this zeta-function admits analytic continuation to a meromorphic function on the complex plane, with a nice functional equation, and whose special values in the left half-plane recover the $KU$-local stable homotopy groups of $X$ away from $2$. We then consider a more general and sophisticated version of the $KU$-local zeta-function, one which is suited to finite CW-complexes $X$ with nontrivial torsion in their $K$-theory. This more sophisticated $KU$-local zeta-function involves a product of $L$-functions of complex representations of the torsion subgroup of $KU^0(X)$, similar to how the Dedekind zeta-function of a number field factors as a product of Artin $L$-functions of complex representations of the Galois group. For a wide range of such finite CW-complexes $X$, we prove analytic continuation, and we show that the special values in the left half-plane recover the $KU$-local stable homotopy groups of $X$ away from $2$ if and only if the skeletal filtration on the torsion subgroup of $KU^0(X)$ splits completely.