Link Search Menu Expand Document

Serre functors

Another fundamental tool in the study of derived categories is the Serre functor. We work with a triangulated category T\mathcal T which is linear over a field kk.

We will set V:=Homk(V,k)V^\ast := \operatorname{Hom}_k(V,k)

Definition. A Serre functor is an auto-equivalence S:TTS: \mathcal T \to \mathcal T for which there are natural isomorphisms Hom(X,Y)Hom(Y,S(X))\operatorname{Hom}(X,Y)^\ast \cong \operatorname{Hom}(Y,S(X))

Example. Let AA be a dg-algebra over kk. We set ():=RHomA(,A)(-)^\vee := \mathbf{R}\operatorname{Hom}_A(-,A)

There is natural map MLANHom(M,N)M^\vee \overset{\mathbf{L}}{\otimes}_A N \to \operatorname{Hom}(M,N) The subcategory of pairs (M,N)(M,N) where this is an isomorphism

  • contains (A[l],A[s])(A[l],A[s]) for each l,sAl,s \in A,
  • is triangulated in MM and NN,
  • is closed under retracts in both MM and NN, and
  • is closed under arbitrary coproducts in NN.

Thus this is an isomorphism for any perfect MM and any NN.

Using tensor-Hom adjunction, we have Homk(MLAN,k)HomA(N,Homk(M,k))\operatorname{Hom}_k(M^\vee \overset{\mathbf{L}}{\otimes}_A N,k) \cong \operatorname{Hom}_A(N, \operatorname{Hom}_k(M^\vee,k))

Thus, if we have Serre functor, then we know its form must be S(M):=(M)S(M) := \left(M^\vee\right)^\ast The question becomes whether M(M)M \mapsto \left(M^\vee\right)^\ast is an auto-equivalence.

Assuming it is an auto-equivalence, we know the SS commutes with \bigoplus. However, it not reasonable to expect (Mt)((Mt))\bigoplus (M_t^\vee)^\ast \to \left(\left(\bigoplus M_t \right)^\vee \right)^\ast to be an isomorphism for general MM. One needs some finiteness in general, either for MM as an AA-module or HMH^\ast M as a kk-module or both.

We will restrict ourselves now to assuming that AA is proper over kk. Recall this means that dimHA<\dim H^\ast A < \infty and we will assume that MM and NN are perfect AA-modules.

We have a natural map
MLAA(M)M \overset{\mathbf{L}}{\otimes}_A A^\ast (M^\vee)^\ast which one can check is isomorphism for perfect MM.

Given that we can write SS as tensoring with the bimodule AA^\ast, it makes sense to look for another bimodule A!A^! with A!LAAAA^! \overset{\mathbf{L}}{\otimes}_A A^\ast \cong A and ALAA!AA^\ast \overset{\mathbf{L}}{\otimes}_A A^! \cong A

Consider RHomAkAop(A,AkA)LAA\mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, A \otimes_k A) \overset{\mathbf{L}}{\otimes}_A A^\ast If we assume that AA is compact, we have RHomAkAop(A,AkA)LAARHomAkAop(A,AkA)\mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, A \otimes_k A) \overset{\mathbf{L}}{\otimes}_A A^\ast \to \mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, A \otimes_k A^\ast) is an isomorphism. Since HAH^\ast A is finite-dimensional, we have RHomAkAop(A,AkA)RHomAkAop(A,Homk(A,A))\mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, A \otimes_k A^\ast) \to \mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, \operatorname{Hom}_k(A,A)) is also an isomorphism. Now, we can use tensor-hom adjunction ARHomAkAop(AkA,A)RHomAkAop(A,Homk(A,A))A \cong \mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A \otimes_k A, A) \to \mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, \operatorname{Hom}_k(A,A)) to see that RHomAkAop(A,AkA)LAAA\mathbf{R}\operatorname{Hom}_{A \otimes_k A^{op}}(A, A \otimes_k A) \overset{\mathbf{L}}{\otimes}_A A^\ast \cong A We get a similar isomorphism if we tensor in the other order.

Thus, if AA is a compact AA-AA bimodule with dimHA<\dim H^\ast A < \infty, we see that MMLAAM \mapsto M \overset{\mathbf{L}}{\otimes}_A A^\ast is the Serre functor on perfA\operatorname{perf} A. Furthermore, when AA is a compact bimodule, perfA\operatorname{perf} A is equivalence to the category D(modfdA)D(\operatorname{mod}_{fd} A) of AA-modules with total cohomology finite dimensional over kk.

Lemma. If we have an endofunctor S:TTS: \mathcal T \to \mathcal T with natural isomorphisms Hom(X,Y)Hom(Y,S(X))\operatorname{Hom}(X,Y)^\ast \cong \operatorname{Hom}(Y,S(X)) and each Hom(X,Y)\operatorname{Hom}(X,Y) is finite-dimensional, then SS is fully-faithful.

Proof. (Expand to view)

We have a natural isomorphism Hom(X,Y)Hom(X,Y)Hom(Y,S(X))Hom(S(X),S(Y)) \operatorname{Hom}(X,Y) \to \operatorname{Hom}(X,Y)^{\ast\ast} \to \operatorname{Hom}(Y,S(X))^\ast \to \operatorname{Hom}(S(X),S(Y)) and one just needs to check that this coincides with application of SS. We omit that verification.

Proposition. Assume F:TSF: \mathcal T \to \mathcal S is exact and both T\mathcal T and S\mathcal S have Serre functors. If FF has right adjoint R:STR : \mathcal S \to \mathcal T, FF has a left adjoint ST1RSSFS_{\mathcal T}^{-1} R S_{\mathcal S} \vdash F If FF has a left adjoint LL, then FF has a right adjoint FST1LSSF \vdash S_{\mathcal T}^{-1} L S_{\mathcal S}

Proof. (Expand to view)

We have natural isomorphisms Hom(Y,FX)Hom(FX,SSY)Hom(X,RSSY)Hom(RSSY,STX)Hom(ST1RSSY,X) \begin{aligned} \operatorname{Hom}(Y,FX) & \cong \operatorname{Hom}(FX,S_{\mathcal S}Y)^\vee \\ & \cong \operatorname{Hom}(X, RS_{\mathcal S}Y)^\vee \\ & \cong \operatorname{Hom}(RS_{\mathcal S}Y, S_{\mathcal T}X) \\ & \cong \operatorname{Hom}( S_{\mathcal T}^{-1} R S_{\mathcal S}Y,X) \end{aligned} and similarly in the other case.