Is normalized chain complex functor the unique Quillen equivalence?

82 Views Asked by At

I don't have a grasp of model categories. I asked the question through Quillen equivalences in order to make it as general as possible. This might be too general to answer and/or might be above my reach. I just have a grasp of Dold-Kan correspondence in the classical setting and wanted to have an idea about this question.

What I really want to ask is the following:

By Dold-Kan we have $(\Gamma \dashv N) : sAb \stackrel{\overset{\Gamma}{\leftarrow}}{\underset{N}{\to}} Ch_\bullet^+$ an equivalence of categories where for $A \in sAb$, ${\pi}_n(A,0) \simeq H_n(NA, \mathbb{Z})$.

Let $(F \dashv G) : sAb \stackrel{\overset{F}{\leftarrow}}{\underset{G}{\to}} Ch_\bullet^+ $ be an equivalence of categories such that ${\pi}_n(A,0) \simeq H_n(GA, \mathbb{Z})$.

I want to investigate if $F$ and $G$ are in some way equivalent to $\Gamma$ and $N$. One can also exclude the condition of $ {\pi}_n(A,0) \simeq H_n(GA, \mathbb{Z}) $ and look at this new question too, but this looks like wildly different than the other two questions, so this is just a side-question. It would be nice to have a reference to see how the "shifting the indices one down" operation in chain complexes corresponds to a functor for simplicial abelian groups, but as I said before, I don't want to make this a mega question. However, in usual references for Dold-Kan, I couldn't find something related to other possible equivalences in general.

Any reference, hint, idea regarding these questions would be nice. Thanks!

1

There are 1 best solutions below

0
On BEST ANSWER

Since equivalences of categories are cocontinuous, it suffices to show that F and G restrict to an equivalence of categories on chain complexes of the form Z[n] and simplicial abelian groups of the form Γ(Z[n]), and these restrictions are isomorphic to Γ and N respectively.

Indeed, Z[n] and Γ(Z[n]) are uniquely characterized by the property that their nth homology respectively homotopy group is Z and the other groups vanish. This proves the first claim.

For the second claim, observe that morphisms from Z[m] to Z[n] vanish if m≠n and are canonically isomorphic to Z if m=n. Likewise for Γ(Z[m]) and Γ(Z[n]). This means that any family of isomorphisms F(Z[n])→Γ(Z[n]) is automatically natural and likewise for G and N.

Hence, F is isomorphic to Γ and G is isomorphic to N.