Units of ring spectra and Thom spectra

TL;DR

This paper reviews and extends the theory of Thom spectra, exploring their orientations, classifications, and two approaches—rigidified models and infinity categories—to better understand their structure and applications.

Contribution

It introduces and compares two frameworks for Thom spectra—rigidified A-infinity/E-infinity models and infinity categories—and relates them through Morita theory.

Findings

Characterization of Thom spectra via Morita theory.

Comparison of rigidified and infinity category approaches.

Extension of Thom spectrum theory to orientations and classifications.

Abstract

We review and extend the theory of Thom spectra and the associated obstruction theory for orientations. We recall (from May, Quinn, and Ray) that a commutative ring spectrum A has a spectrum of units gl(A). To a map of spectra f: b -> bgl(A), we associate a commutative A-algebra Thom spectrum Mf, which admits a commutative A-algebra map to R if and only if b -> bgl(A) -> bgl(R) is null. If A is an associative ring spectrum, then to a map of spaces f: B -> BGL(A) we associate an A-module Thom spectrum Mf, which admits an R-orientation if and only if B -> BGL(A) -> BGL(R) is null. We also note that BGL(A) classifies the twists of A-theory. We develop and compare two approaches to the theory of Thom spectra. The first involves a rigidified model of A-infinity and E-infinity spaces. Our second approach is via infinity categories. In order to compare these approaches to one another and to the classical theory, we characterize the Thom spectrum functor from the perspective of Morita theory.