Derived completions in stable homotopy theory

We construct a notion of derived completion which applies to homomorphisms of commutative S-algebras. We study the relationship of the construction with other constructions of completions, and prove various invariance properties. The construction is expected to have applications within algebraic K-theory.

💡 Research Summary

The paper introduces a new construction called “derived completion” that applies to homomorphisms between commutative S‑algebras, extending the classical notion of completion from ordinary algebra to the realm of spectra. The authors begin by recalling the limitations of traditional I‑adic or Bousfield–Kan completions when one works with structured ring spectra: those constructions are usually defined only for a fixed algebra and do not interact well with maps between different algebras. To remedy this, they fix a map f : A → B of commutative S‑algebras and define the derived completion of an A‑module spectrum M with respect to f. The definition proceeds by forming a cosimplicial object C·(M) whose n‑th term is M ⊗_A B^{⊗ n}. The totalization Tot C·(M) (or equivalently the homotopy limit of the associated tower) is taken as the completed object M̂_f. This construction is reminiscent of the classical derived functor of I‑adic completion, but it lives entirely in the stable homotopy category.

A substantial part of the paper is devoted to comparing this new construction with existing completions. The authors prove a comparison theorem stating that when B is a flat A‑module and the map f is “derived complete” in the sense of Bousfield, the derived completion coincides (up to weak equivalence) with the Bousfield–Kan completion of M. Moreover, they establish several invariance properties: if f is flat or if B is A‑complete, then the induced map on completions f̂ : Â_f → B̂_f is a weak equivalence. This shows that derived completion respects the underlying algebraic structure and does not introduce spurious homotopical artifacts.

The paper also explores the higher‑categorical behavior of derived completion. For a composable pair of maps A → B → C, the authors demonstrate that the derived completion of the composite is naturally equivalent to the composite of the derived completions, i.e. (g∘f)̂ ≃ ĝ∘f̂. Consequently, derived completion defines a functor on the homotopy category of commutative S‑algebras that preserves composition. In addition, they prove that mapping spectra are preserved under completion: the natural map Hom_A(M,N)̂ → Hom_{Â_f}(M̂_f,N̂_f) is a weak equivalence. This “completion of morphisms” property is crucial for applications where one needs to retain the full homotopical information after completing.

The final section discusses potential applications to algebraic K‑theory. Given a map of K‑theory spectra K(A) → K(B) induced by f, the authors show that if f̂ is a weak equivalence then the induced map on completed K‑theory spectra K̂_f(A) → K̂_f(B) is also a weak equivalence. This result suggests that derived completion can be used to simplify K‑theoretic calculations in situations where one works with p‑adic or I‑adic completions, by allowing one to replace a complicated algebra with its derived completion without losing K‑theoretic information. The paper hints at further directions, such as studying the interaction of derived completion with trace methods, cyclotomic spectra, and chromatic homotopy theory.

Overall, the work provides a robust, homotopy‑coherent notion of completion for maps of structured ring spectra, establishes its compatibility with existing constructions, proves essential invariance and functoriality properties, and opens the door to concrete applications in algebraic K‑theory and beyond.