Documenta Math.

403

Silting Objects, Simple-Minded Collections,

t-Structures and Co-t-Structures for
Finite-Dimensional Algebras

Steffen Koenig, Dong Yang

Received: February 19, 2013

Revised: September 13, 2013

Communicated by Wolfgang Soergel

Abstract. Bijective correspondences are established between (1)

silting objects, (2) simple-minded collections, (3) bounded t-structures
with length heart and (4) bounded co-t-structures. These correspon-
dences are shown to commute with mutations and partial orders. The
results are valid for finite-dimensional algebras. A concrete example
is given to illustrate how these correspondences help to compute the
space of Bridgeland’s stability conditions.

2010 Mathematics Subject Classification: 16E35, 16E45, 18E30

Keywords and Phrases: silting object, simple-minded collection, t-
structure, co-t-structure, mutation.

Contents
1. Introduction 404
2. Notations and preliminaries 405
3. The four concepts 407
4. Finite-dimensional non-positive dg algebras 410
5. The maps 415
6. The correspondences are bijections 421
7. Mutations and partial orders 422
8. A concrete example 431
References 435

Documenta Mathematica 19 (2014) 403–438

404 Steffen Koenig, Dong Yang

1. Introduction
Let Λ be a finite-dimensional associative algebra. Fundamental objects of study
in the representation theory of Λ are the projective modules, the simple mod-
ules and the category of all (finite-dimensional) Λ-modules. Various structural
concepts have been introduced that include one of these classes of objects as
particular instances. In this article, four such concepts are related by explicit
bijections. Moreover, these bijections are shown to commute with the basic
operation of mutation and to preserve partial orders.
These four concepts may be based on two different general points of view, ei-
ther considering particular generators of categories ((1) and (2)) or considering
structures on categories that identify particular subcategories ((3) and (4)):
(1) Focussing on objects that generate categories, the theory of Morita
equivalences has been extended to tilting or derived equivalences. In
this way, projective generators are examples of tilting modules, which
have been generalised further to silting objects (which are allowed to
have negative self-extensions).
(2) Another, and different, natural choice of ‘generators’ of a module cate-
gory is the set of simple modules (up to isomorphism). In the context
of derived or stable equivalences, this set is included in the concept of
simple-minded system or simple-minded collection.
(3) Starting with a triangulated category and looking for particular sub-
categories, t-structures have been defined so as to provide abelian cate-
gories as their hearts. The finite-dimensional Λ-modules form the heart
of some t-structure in the bounded derived category Db (mod Λ).
(4) Choosing as triangulated category the homotopy category K b (proj Λ),
one considers co-t-structures. The additive category proj Λ occurs as
the co-heart of some co-t-structure in K b (proj Λ).
The first main result of this article is:
Theorem (6.1). Let Λ be a finite-dimensional algebra over a field K. There
are one-to-one correspondences between
(1) equivalence classes of silting objects in K b (proj Λ),
(2) equivalence classes of simple-minded collections in Db (mod Λ),
(3) bounded t-structures of Db (mod Λ) with length heart,
(4) bounded co-t-structures of K b (proj Λ).
Here two sets of objects in a category are equivalent if they additively generate
the same subcategory.
A common feature of all four concepts it that they allow for comparisons,
often by equivalences. In particular, each of the four structures to be related
comes with a basic operation, called mutation, which produces a new such
structure from a given one. Moreover, on each of the four structures there is a
partial order. All the bijections in Theorem 6.1 enjoy the following naturality
properties:

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 405

Theorem (7.12). Each of the bijections between the four structures (1), (2),
(3) and (4) commutes with the respective operation of mutation.
Theorem (7.13). Each of the bijections between the four structures (1), (2),
(3) and (4) preserves the respective partial orders.
The four concepts are crucial in representation theory, geometry and topology.
They are also closely related to fundamental concepts in cluster theory such as
clusters ([20]), c-matrices and g-matrices ([21, 40]) and cluster-tilting objects
([7]). We refer to the survey paper [16] for more details. A concrete example
to be given at the end of the article demonstrates one practical use of these
bijections and their properties.
Finally we give some remarks on the literature. For path algebras of Dynkin
quivers, Keller and Vossieck [33] have already given a bijection between
bounded t-structures and silting objects. The bijection between silting ob-
jects and t-structures with length heart has been established by Keller and
Nicolás [32] for homologically smooth non-positive dg algebras, by Assem,
Souto Salorio and Trepode [5] and by Vitória [46], who are focussing on piece-
wise hereditary algebras. An unbounded version of this bijection has been
studied by Aihara and Iyama [1]. The bijection between simple-minded collec-
tions and bounded t-structures has been established implicitely in Al-Nofayee’s
work [3] and explicitely for homologically smooth non-positive dg algebras in
Keller and Nicolás’ work [32] and for finite-dimensional algebras in our preprint
[37], which has been partly incorporated into the present article, and partially
in the work [44] of Rickard and Rouquier. For hereditary algebras, Buan, Reiten
and Thomas [17] studied the bijections between silting objects, simple-minded
collections (=Hom≤0 -configurations in their setting) and bounded t-structures.
The correspondence between silting objects and co-t-structures appears implic-
itly on various levels of generality in the work of Aihara and Iyama [1] and of
Bondarko [12] and explicitly in full generality in the work of Mendoza, Sáenz,
Santiago and Souto Salorio [39] and of Keller and Nicolás [31]. For homologi-
cally smooth non-positive dg algebras, all the bijections are due to Keller and
Nicolás [31]. The intersection of our results with those of Keller and Nicolás is
the case of finite-dimensional algebras of finite global dimension.
Acknowledgement. The authors would like to thank Paul Balmer, Mark Blume,
Martin Kalck, Henning Krause, Qunhua Liu, Yuya Mizuno, David Pauksztello,
Pierre-Guy Plamondon, David Ploog, Jorge Vitória and Jie Xiao for inspiring
discussions and helpful remarks. The second-named author gratefully acknowl-
edges financial support from Max-Planck-Institut für Mathematik in Bonn and
from DFG program SPP 1388 (YA297/1-1). He is deeply grateful to Bernhard
Keller for valuable conversations on derived categories of dg algebras.

2. Notations and preliminaries

2.1. Notations. Throughout, K will be a field. All algebras, modules, vector
spaces and categories are over the base field K, and D = HomK (?, K) denotes

Documenta Mathematica 19 (2014) 403–438

406 Steffen Koenig, Dong Yang

the K-dual. By abuse of notation, we will denote by Σ the suspension functors

of all the triangulated categories.
For a category C, we denote by HomC (X, Y ) the morphism space from X to
Y , where X and Y are two objects of C. We will omit the subscript and write
Hom(X, Y ) when it does not cause confusion. For S a set of objects or a
subcategory of C, call
⊥
S = {X ∈ C | Hom(X, S) = 0 for all S ∈ S}
and
S ⊥ = {X ∈ C | Hom(S, X) = 0 for all S ∈ S}
the left and right perpendicular category of S, respectively.
Let C be an additive category and S a set of objects or a subcategory of
C. Let Add(S) and add(S), respectively, denote the smallest full subcategory
of C containing all objects of S and stable for taking direct summands and
coproducts respectively taking finite coproducts. The category add(S) will be
called the additive closure of S. If further C is abelian or triangulated, the
extension closure of S is the smallest subcategory of C containing S and stable
under taking extensions. Assume that C is triangulated and let thick(S) denote
the smallest triangulated subcategory of C containing objects in S and stable
under taking direct summands. We say that S is a set of generators of C, or
that C is generated by S, when C = thick(S).

2.2. Derived categories. For a finite-dimensional algebra Λ, let Mod Λ (re-

spectively, mod Λ, proj Λ, inj Λ) denote the category of right Λ-modules (respec-
tively, finite-dimensional right Λ-modules, finite-dimensional projective, injec-
tive right Λ-modules), let K b (proj Λ) (respectively, K b (inj Λ)) denote the ho-
motopy category of bounded complexes of proj Λ (respectively, inj Λ) and let
D(Mod Λ) (respectively, Db (mod Λ), D− (mod Λ)) denote the derived category
of Mod Λ (respectively, bounded derived category of mod Λ, bounded above de-
rived category of mod Λ). All these categories are triangulated with suspension
functor the shift functor. We view D− (mod Λ) and Db (mod Λ) as triangulated
subcategories of D(Mod Λ).
The categories mod Λ, Db (mod Λ) and K b (proj Λ) are Krull–Schmidt categories.
An object M of mod Λ (respectively, Db (mod Λ), K b (proj Λ)) is said to be basic
if every indecomposable direct summand of M has multiplicity 1. The finite-
dimensional algebra Λ is said to be basic if the free module of rank 1 is basic
in mod Λ (equivalently, in Db (mod Λ) or K b (proj Λ)).
For a differential graded(=dg) algebra A, let C(A) denote the category of (right)
dg modules over A and K(A) the homotopy category. Let D(A) denote the
derived category of dg A-modules, i.e. the triangle quotient of K(A) by acyclic
dg A-modules, cf. [29, 30], and let Df d (A) denote its full subcategory of dg
A-modules whose total cohomology is finite-dimensional. The category C(A) is
abelian and the other three categories are triangulated with suspension functor
the shift functor of complexes. Let per(A) = thick(AA ), i.e. the triangulated
subcategory of D(A) generated by the free dg A-module of rank 1.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 407

For two dg A-modules M and N , let Hom A (M , N ) denote the complex whose
degree n component consists of those A-linear maps from M to N which are
homogeneous of degree n, and whose differential takes a homogeneous map f
of degree n to dN ◦ f − (−1)n f ◦ dM . Then
(2.1) HomK(A) (M, N ) = H 0 Hom A (M , N ).
A dg A-module M is said to be K-projective if Hom A (M , N ) is acyclic when
N is an acyclic dg A-module. For example, AA , the free dg A-module of rank
1 is K-projective, because Hom A (A, N ) = N . Dually, one defines K-injective
dg modules, and D(A A) is K-injective. For two dg A-modules M and N such
that M is K-projective or N is K-injective, we have
(2.2) HomD(A) (M, N ) = HomK(A) (M, N ).
Let A and B be two dg algebras. Then a triangle equivalence between D(A) and
D(B) restricts to a triangle equivalence between per(A) and per(B) and also to
a triangle equivalence between Df d (A) and Df d (B). If A is a finite-dimensional
algebra viewed as a dg algebra concentrated in degree 0, then D(A) is exactly
D(Mod A), Df d (A) is Db (mod A), per(A) is triangle equivalent to K b (proj A),
and thick(D(A A)) is triangle equivalent to K b (inj A).

2.3. The Nakayama functor. Let Λ be a finite-dimensional algebra. The

Nakayama functor νmod Λ is defined as νmod Λ =? ⊗Λ D(Λ Λ), and the inverse
−1 −1
Nakayama functor νmod Λ is its right adjoint νmod Λ = HomΛ (D(Λ Λ), ?). They
restrict to quasi-inverse equivalences between proj Λ and inj Λ.
−1 −1
The derived functors of νmod Λ and νmod Λ , denoted by ν and ν , restrict to
quasi-inverse triangle equivalences between K b (proj Λ) and K b (inj Λ). When
Λ is self-injective, they restrict to quasi-inverse triangle auto-equivalences of
Db (mod Λ).
The Auslander–Reiten formula for M in K b (proj Λ) and N in D(Mod Λ) (cf. [23,
Chapter 1, Section 4.6]) provides an isomorphism
D Hom(M, N ) ∼
= Hom(N, νM ),
which is natural in M and N . When K b (proj Λ) coincides with K b (inj Λ) (that
is, when Λ is Gorenstein), it has Auslander–Reiten triangles and the Auslander–
Reiten translation is τ = ν ◦ Σ−1 .

3. The four concepts

In this section we introduce silting objects, simple-minded collections, t-
structures and co-t-structure. Let C be a triangulated category with suspension
functor Σ.

3.1. Silting objects. A subcategory M of C is called a silting subcate-

gory [33, 1] if it is stable for taking direct summands and generates C (i.e.
C = thick(M)) and if Hom(M, Σm N ) = 0 for m > 0 and M, N ∈ M.

Documenta Mathematica 19 (2014) 403–438

408 Steffen Koenig, Dong Yang

Theorem 3.1. ([1, Theorem 2.27]) Assume that C is Krull–Schmidt and has a
silting subcategory M. Then the Grothendieck group of C is free and its rank
is equal to the cardinality of the set of isomorphism classes of indecomposable
objects of M.
An object M of C is called a silting object if add M is a silting subcategory
of C. This notion was introduced by Keller and Vossieck in [33] to study t-
structures on the bounded derived category of representations over a Dynkin
quiver. Recently it has also been studied by Wei [47] (who uses the terminology
semi-tilting complexes) from the perspective of classical tilting theory. A tilting
object is a silting object M such that Hom(M, Σm M ) = 0 for m < 0. For an
algebra Λ, a tilting object in K b (proj Λ) is called a tilting complex in the liter-
ature. For example, the free module of rank 1 is a tilting object in K b (proj Λ).
Assume that Λ is finite-dimensional. Theorem 3.1 implies that (a) any silting
subcategory of K b (proj Λ) is the additive closure of a silting object, and (b)
any two basic silting objects have the same number of indecomposable direct
summands. We will rederive (b) as a corollary of the existence of a certain
derived equivalence (Corollary 5.1).
3.2. Simple-minded collections.
Definition 3.2. A collection X1 , . . . , Xr of objects of C is said to be simple-
minded (cohomologically Schurian in [3]) if the following conditions hold for
i, j = 1, . . . , r
· Hom(Xi , Σm Xj ) = 0, ∀ m < 0,
· End(Xi ) is a division algebra and Hom(Xi , Xj ) vanishes for i 6= j,
· X1 , . . . , Xr generate C (i.e. C = thick(X1 , . . . , Xr )).
Simple-minded collections are variants of simple-minded systems in [36] and
were first studied by Rickard [43] in the context of derived equivalences of
symmetric algebras. For a finite-dimensional algebra Λ, a complete collection
of pairwise non-isomorphic simple modules is a simple-minded collection in
Db (mod Λ). A natural question is: do any two simple-minded collections have
the same collection of endomorphism algebras?
3.3. t-structures. A t-structure on C ([8]) is a pair (C ≤0 , C ≥0 ) of strict (that
is, closed under isomorphisms) and full subcategories of C such that
· ΣC ≤0 ⊆ C ≤0 and Σ−1 C ≥0 ⊆ C ≥0 ;
· Hom(M, Σ−1 N ) = 0 for M ∈ C ≤0 and N ∈ C ≥0 ,
· for each M ∈ C there is a triangle M ′ → M → M ′′ → ΣM ′ in C with
M ′ ∈ C ≤0 and M ′′ ∈ Σ−1 C ≥0 .
The two subcategories C ≤0 and C ≥0 are often called the aisle and the co-aisle of
the t-structure respectively. The heart C ≤0 ∩ C ≥0 is always abelian. Moreover,
Hom(M, Σm N ) vanishes for any two objects M and N in the heart and for any
m < 0. The t-structure (C ≤0 , C ≥0 ) is said to be bounded if
[ [
Σn C ≤0 = C = Σn C ≥0 .
n∈Z n∈Z

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 409

A bounded t-structure is one of the two ingredients of a Bridgeland stability

condition [15]. A typical example of a t-structure is the pair (D≤0 , D≥0 ) for the
derived category D(Mod Λ) of an (ordinary) algebra Λ, where D≤0 consists of
complexes with vanishing cohomologies in positive degrees, and D≥0 consists
of complexes with vanishing cohomologies in negative degrees. This t-structure
restricts to a bounded t-structure of Db (mod Λ) whose heart is mod Λ, which is
a length category, i.e. every object in it has finite length. The following lemma
is well-known.
Lemma 3.3. Let (C ≤0 , C ≥0 ) be a bounded t-structure on C with heart A.
(a) The embedding A → C induces an isomorphism K0 (A) → K0 (C) of
Grothendieck groups.
(b) C ≤0 respectively C ≥0 is the extension closure of Σm A for m ≥ 0 respec-
tively for m ≤ 0.
(c) C = thick(A).
Assume further A is a length category with simple objects {Si | i ∈ I}.
(d) C ≤0 respectively C ≥0 is the extension closure of Σm {Si | i ∈ I} for
m ≥ 0 respectively for m ≤ 0.
(e) C = thick(Si , i ∈ I).
(f) If I is finite, then {Si | i ∈ I} is a simple-minded collection.
3.4. Co-t-structures. According to [41], a co-t-structure on C (or weight
structure in [12]) is a pair (C≥0 , C≤0 ) of strict and full subcategories of C such
that
· both C≥0 and C≤0 are additive and closed under taking direct sum-
mands,
· Σ−1 C≥0 ⊆ C≥0 and ΣC≤0 ⊆ C≤0 ;
· Hom(M, ΣN ) = 0 for M ∈ C≥0 and N ∈ C≤0 ,
· for each M ∈ C there is a triangle M ′ → M → M ′′ → ΣM ′ in C with
M ′ ∈ C≥0 and M ′′ ∈ ΣC≤0 .
The co-heart is defined as the intersection C≥0 ∩ C≤0 . This is usually not an
abelian category. For any two objects M and N in the co-heart, the morphism
space Hom(M, Σm N ) vanishes for any m > 0. The co-t-structure (C ≤0 , C ≥0 ) is
said to be bounded [12] if
[ [
Σn C≤0 = C = Σn C≥0 .
n∈Z n∈Z

A bounded co-t-structure is one of the two ingredients of a Jørgensen–

Pauksztello costability condition [27]. A typical example of a co-t-structure
is the pair (K≥0 , K≤0 ) for the homotopy category K b (proj Λ) of a finite-
dimensional algebra Λ, where K≥0 consists of complexes which are homotopy
equivalent to a complex bounded below at 0, and K≤0 consists of complexes
which are homotopy equivalent to a complex bounded above at 0. The co-heart
of this co-t-structure is proj Λ.

Documenta Mathematica 19 (2014) 403–438

410 Steffen Koenig, Dong Yang

Lemma 3.4. ([39, Theorem 4.10 (a)]) Let (C≥0 , C≤0 ) be a bounded co-t-structure
on C with co-heart A. Then A is a silting subcategory of C.
Proof. For the convenience of the reader we give a proof. It suffices to show that
C = thick(A). Let M be an object of C. Since the co-t-structure is bounded,
there are integers m ≥ n such that M ∈ Σm C≥0 ∩ Σn C≤0 . Up to suspension
and cosuspension we may assume that m = 0. If n = 0, then M ∈ A. Suppose
n < 0. There exists a triangle

M′ /M / M ′′ / ΣM ′

with M ′ ∈ Σ−1 C≥0 and M ′′ ∈ C≤0 . In fact, M ′′ ∈ A, see [12, Proposition

1.3.3.6]. Moreover, ΣM ′ ∈ Σn+1 C≤0 due to the triangle

M ′′ / ΣM ′ / ΣM / ΣM ′′

since both M ′′ and ΣM belong to Σn+1 C≤0 and C≤0 is extension closed (see [12,
Proposition 1.3.3.3]). So ΣM ′ ∈ C≥0 ∪ Σn+1 C≤0 . We finish the proof by √
induction on n.

Proposition 3.5. ([1, Proposition 2.22], [12, (proof of) Theorem 4.3.2], [39,
Theorem 5.5] and [31]) Let A be a silting subcategory of C. Let C≤0 respectively
C≥0 be the extension closure of Σm A for m ≥ 0 respectively for m ≤ 0. Then
(C≥0 , C≤0 ) is a bounded co-t-structure on C with co-heart A.

4. Finite-dimensional non-positive dg algebras

In this sectionLwe study derived categories of non-positive dg algebras, i.e. dg
algebras A = i∈Z Ai with Ai = 0 for i > 0, especially finite-dimensional non-
positive dg algebras, i.e. , non-positive dg algebras which, as vector spaces, are
finite-dimensional. These results will be used in Sections 5.1 and 5.4.
Non-positive dg algebras are closely related to silting objects. A triangulated
category is said to be algebraic if it is triangle equivalent to the stable category
of a Frobenius category.
Lemma 4.1. (a) Let A be a non-positive dg algebra. The free dg A-module
of rank 1 is a silting object of per(A).
(b) Let C be an algebraic triangulated category with split idempotents and
let M ∈ C be a silting object. Then there is a non-positive dg algebra A
∼
together with a triangle equivalence per(A) → C which takes A to M .
Proof. (a) This is because Homper(A) (A, Σi A) = H i (A) vanishes for i > 0.
(b) By [30, Theorem 3.8 b)] (which is a ‘classically generated’ version of [29,
Theorem 4.3]), there is a dg algebra A′ together with a triangle equivalence
∼
per(A′ ) → C. In particular, there are isomorphisms Homper(A′ ) (A′ , Σi A′ ) ∼
=
HomC (M, Σi M ) for all i ∈ Z. Since M is a silting object, A′ has vanishing
cohomologies in positive degrees. Therefore, if A = τ≤0 A′ is the standard

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 411

truncation at position 0, then the embedding A ֒→ A′ is a quasi-isomorphism.

It follows that there is a composite triangle equivalence
∼ / per(A′ ) ∼ /C
per(A)
√
which takes A to M .

In the sequel of this section we assume that A is a finite-dimensional non-

positive dg algebra. The 0-th cohomology Ā = H 0 (A) of A is a finite-
dimensional K-algebra. Let Mod Ā and mod Ā denote the category of (right)
modules over Ā and its subcategory consisting of those finite-dimensional mod-
ules. Let π : A → Ā be the canonical projection. We view Mod Ā as a subcat-
egory of C(A) via π.
The total cohomology H ∗ (A) of A is a finite-dimensional graded algebra with
multiplication induced from the multiplication of A. Let M be a dg A-module.
Then the total cohomology H ∗ (M ) carries a graded H ∗ (A)-module structure,
and hence a graded Ā = H 0 (A)-module structure. In particular, a stalk dg
A-module concentrated in degree 0 is an Ā-module.

4.1. The standard t-structure. We follow [22, 4, 34], where the dg algebra
is not necessarily finite-dimensional.
di−1 di
Let M = . . . → M i−1 → M i → M i+1 → . . . be a dg A-module. Consider the
standard truncation functors τ≤0 and τ>0 :
−2
/ M −2 d / M −1 d−1
τ≤0 M = . . . / kerd0 /0 /0 /0 / ...

d0 d1 d2
τ>0 M = . . . /0 /0 / M 0 /kerd0 / M1 / M2 / M3 / ...

Since A is non-positive, τ≤0 M is a dg A-submodule of M and τ>0 M is the

corresponding quotient dg A-module. Hence there is a distinguished triangle
in D(A)
τ≤0 M → M → τ>0 M → Στ≤0 M.
These two functors define a t-structure (D≤0 , D≥0 ) on D(A), where D≤0 is the
subcategory of D(A) consisting of dg A-modules with vanishing cohomology
in positive degrees, and D≥0 is the subcategory of D(A) consisting of dg A-
modules with vanishing cohomology in negative degrees.
By the definition of the t-structure (D≤0 , D≥0 ), the heart H = D≤0 ∩ D≥0
consists of those dg A-modules whose cohomology is concentrated in degree 0.
Thus the functor H 0 induces an equivalence
H0 : H −→ Mod Ā.
M 7→ H 0 (M )
See also [26, Theorem 1.3]. The t-structure (D≤0 , D≥0 ) on D(A) restricts to a
bounded t-structure on Df d (A) with heart equivalent to mod Ā.

Documenta Mathematica 19 (2014) 403–438

412 Steffen Koenig, Dong Yang

4.2. Morita reduction. Let d be the differential of A. Then d(A0 ) = 0.

Let e be an idempotent of A. For degree reasons, e must belong to A0 , and
the graded subspace eA of A is a dg submodule: d(ea) = d(e)a + ed(a) =
ed(a). Therefore for each decomposition 1 = e1 + . . . + en of the unity into a
sum of primitive orthogonal idempotents, there is a direct sum decomposition
A = e1 A ⊕ . . . ⊕ en A of A into indecomposable dg A-modules. Moreover, if e
and e′ are two idempotents of A such that eA ∼ = e′ A as ordinary modules over
the ordinary algebra A, then this isomorphism is also an isomorphism of dg
modules. Indeed, there are two elements of A such that f g = e and gf = e′ .
Again for degree reasons, f and g belong to A0 . So they induce isomorphisms
of dg A-modules: eA → e′ A, a 7→ ga and e′ A → eA, a 7→ f a. It follows that
the above decomposition of A into a direct sum of indecomposable dg modules
is essentially unique. Namely, if 1 = e′1 + . . . + e′n is another decomposition of
the unity into a sum of primitive orthogonal idempotents, then m = n and up
to reordering, e1 A ∼= e′1 A, . . ., en A ∼
= e′n A.
4.3. The perfect derived category. Since A is finite-dimensional (and
thus has finite-dimensional total cohomology), per(A) is a triangulated subcat-
egory of Df d (A).
We assume, as we may, that A is basic. Let 1 = e1 +. . .+en be a decomposition
of 1 in A into a sum of primitive orthogonal idempotents. Since d(x) = λ1 ei1 +
. . . + λs eis implies that d(eij x) = λj eij , the intersection of the space spanned
by e1 , . . . , en with the image of the differential d has a basis consisting of some
ei ’s, say er+1 , . . . , en . So, er+1 A, . . . , en A are homotopic to zero.
We say that a dg A-module M is strictly perfect if its underlying graded module
LN
is of the form j=1 Rj , where Rj belongs to add(Σtj A) for some tj with t1 <
t2 < . . . < tN , and if its differential is of the form dint + δ, where dint is
the direct sum of the differential of the Rj ’s, and δ, as a degree 1 map from
LN
j=1 Rj to itself, is a strictly upper triangular matrix whose entries are in
A. It is minimal if in addition no shifted copy of er+1 A, . . . , en A belongs to
add(R1 , . . . , Rj ), and the entries of δ are in the radical of A, cf. [42, Section
2.8]. Strictly perfect dg modules are K-projective. If A is an ordinary algebra,
then strictly perfect dg modules are precisely bounded complexes of finitely
generated projective modules.
Lemma 4.2. Let M be a dg A-module belonging to per(A). Then M is quasi-
isomorphic to a minimal strictly perfect dg A-module.
Proof. Bearing in mind that e1 A, . . . , er A have local endomorphism algebras
and er+1 A, . . . , en A are homotopic to zero, we prove the assertion as in [42,
√
Lemma 2.14].
4.4. Simple modules. Assume that A is basic. According to the preceding
subsection, we may assume that there is a decomposition 1 = e1 + . . . + er +
er+1 + . . .+ en of the unity of A into a sum of primitive orthogonal idempotents
such that 1 = ē1 + . . . + ēr is a decomposition of 1 in Ā into a sum of primitive
orthogonal idempotents.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 413

Let S1 , . . . , Sr be a complete set of pairwise non-isomorphic simple Ā-modules

and let R1 , . . . , Rr be their endomorphism algebras. Then
(
Rj Rj if i = j,
HomA (ei A, Sj ) =
0 otherwise.
Therefore, by (2.1) and (2.2),
(
m Rj Rj if i = j and m = 0,
HomD(A) (ei A, Σ Sj ) =
0 otherwise.
Moreover, {e1 A, . . . , er A} and {S1 , . . . , Sr } characterise each other by this
property. On the one hand, if M is a dg A-module such that for some in-
teger 1 ≤ j ≤ r
(
Rj Rj if i = j and m = 0,
HomD(A) (ei A, Σm M ) =
0 otherwise,
then M is isomorphic in D(A) to Sj . On the other hand, let M be an object
of per(A) such that for some integer 1 ≤ i ≤ r
(
m Rj Rj if i = j and m = 0,
HomD(A) (M, Σ Sj ) =
0 otherwise.
Then by replacing M by its minimal perfect resolution (Lemma 4.2), we see
that M is isomorphic in D(A) to ei A.
Further, recall from Section 4.1 that Df d (A) admits a standard t-structure
whose heart is equivalent to mod Ā. This implies that the simple modules
S1 , . . . , Sr form a simple-minded collection in Df d (A).

4.5. The Nakayama functor. For a complex M of K-vector spaces, we

define its dual as D(M ) = HomK (M, K), where K in the second argument is
considered as a complex concentrated in degree 0. One checks that D defines
a duality between finite-dimensional dg A-modules and finite-dimensional dg
Aop -modules.
Let e be an idempotent of A and M a dg A-module. Then there is a canonical
isomorphism
HomA (eA, M ) ∼
= M e.
If in addition each component of M is finite-dimensional , there are canonical
isomorphisms
= Me ∼
HomA (eA, M ) ∼ = DHomA (M, D(Ae)).
Let C(A) denote the category of dg A-modules. The Nakayama functor ν :
C(A) → C(A) is defined by ν(M ) = DHomA (M, A) [29, Section 10]. There are
canonical isomorphisms
DHom A (M , N ) ∼
= Hom A (N , νM )

Documenta Mathematica 19 (2014) 403–438

414 Steffen Koenig, Dong Yang

for any strictly perfect dg A-module M and any dg A-module N . Then

ν(eA) = D(Ae) for an idempotent e of A, and the functor ν induces a tri-
angle equivalences between the subcategories per(A) and thick(D(A)) of D(A)
with quasi-inverse given by ν −1 (M ) = HomA (D(A), M ). Moreover, we have
the Auslander–Reiten formula

D Hom(M, N ) ∼
= Hom(N, νM ),

which is natural in M ∈ per(A) and N ∈ D(A).

Let e1 , . . . , er , S1 , . . . , Sr and R1 , . . . , Rr be as in the preceding subsection.
Then
(
∼ (Rj )Rj if i = j,
HomA (Sj , D(Aei )) = DHom A (ei A, Sj ) =
0 otherwise.

Therefore, by (2.1) and (2.2),

(
(Rj )Rj if i = j and m = 0,
HomD(A) (Sj , Σm D(Aei )) =
0 otherwise.

Moreover, {D(Ae1 ), . . . , D(Aer )} and {S1 , . . . , Sr } characterise each other in

D(A) by this property. This follows from the arguments in the preceding
subsection by applying the functors ν and ν −1 .

4.6. The standard co-t-structure. Let P≤0 (respectively, P≥0 ) be the

smallest full subcategory of per(A) containing {Σm A | m ≥ 0} (respectively,
{Σm A | m ≤ 0}) and closed under taking extensions and direct summands.
The following lemma is a special case of Proposition 3.5. For the convenience
of the reader we include a proof.

Lemma 4.3. The pair (P≥0 , P≤0 ) is a co-t-structure on per(A). Moreover, its
co-heart is add(AA ).

Proof. Since Hom(A, Σm A) = 0 for m ≥ 0, it follows that Hom(X, ΣY ) = 0

for M ∈ P≥0 and N ∈ P≤0 . It remains to show that any object M in
per(A) fits into a triangle whose outer terms belong to P≥0 and P≤0 , respec-
tively. By Lemma 4.2, we may assume that M is minimal perfect. Write
LN
M = ( j=1 Rj , dint + δ) as in Section 4.3. Let N ′ ∈ {1, . . . , N } be the unique
L ′
integer such that tN ′ ≥ 0 but tN ′ +1 < 0. Let M ′ be the graded module N j=1 Rj
endowed with the differential restricted from dint + δ. Because dint + δ is upper
triangular, M ′ is a dg submodule of M . Clearly M ′ belongs to P≥0 and the
quotient M ′′ = M/M ′ belongs to ΣP≤0 . Thus we obtain the desired triangle

M′ /M / M ′′ / ΣM ′
√
with M ′ in P≥0 and M ′′ in ΣP≤0 .

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 415

5. The maps
Let Λ be a finite-dimensional basic K-algebra. This section is devoted to defin-
ing the maps in the following diagram.

φ41
equivalence classes of silt- ✲ bounded co-t-structures on
ing objects in K b (proj Λ) ✛ K b (proj Λ)
φ14

✻ ❅ φ31
φ12 φ21 φ34
❅
❄ ❅
❘
❅ ❄

equivalence classes of φ32 bounded t-structures on

✲
simple-minded collections Db (mod Λ) with length
in Db (mod Λ) ✛ heart
φ23

5.1. Silting objects induce derived equivalences. Let M be a basic silt-

ing object of the category K b (proj Λ). By definition, M is a bounded complex of
finitely generated projective Λ-modules such that HomK b (proj Λ) (M, Σm M ) van-
ishes for all m > 0. By Lemma 4.1, there is a non-positive dg algebra whose
perfect derived category is triangle equivalent to K b (proj Λ). This equivalence
sends the free dg module of rank 1 to M . Below we explicitly construct such a
dg algebra.
Consider Hom Λ (M , M ). Recall that the degree n component of Hom Λ (M , M )
consists of those Λ-linear maps from M to itself which are homogeneous of
degree n. The differential of Hom Λ (M , M ) takes a homogeneous map f of
degree n to d ◦ f − (−1)n f ◦ d, where d is the differential of M . This dif-
ferential and the composition of maps makes Hom Λ (M , M ) into a dg al-
gebra. Therefore Hom Λ (M , M ) is a finite-dimensional dg algebra. More-
over, H m (Hom Λ (M , M )) = HomD(Λ) (M , Σ m M ) for any integer m, by (2.1)
and (2.2). Because M is a silting object, Hom Λ (M , M ) has cohomology
concentrated in non-positive degrees. Take the truncated dg algebra Γ̃ =
τ≤0 Hom Λ (M , M ), where τ≤0 is the standard truncation at position 0. Then
the embedding Γ̃ → Hom Λ (M , M ) is a quasi-isomorphism of dg algebras, and
hence Γ̃ is a finite-dimensional non-positive dg algebra. Therefore, the derived
category D(Γ̃) carries a natural t-structure (D≤0 , D≥0 ) with heart D≤0 ∩ D≥0
equivalent to Mod Γ, where Γ = H 0 (Γ̃) = EndD(A) (M ). This t-structure re-
stricts to a t-structure on Df d (Γ̃), denoted by (Df≤0 ≥0
d , Df d ), whose heart is
equivalent to mod Γ. Moreover, there is a standard co-t-structure (P≥0 , P≤0 )
on per(Γ̃), see Section 4.

Documenta Mathematica 19 (2014) 403–438

416 Steffen Koenig, Dong Yang

The object M has a natural dg Γ̃-Λ-bimodule structure. Moreover, since it

generates K b (proj Λ), it follows from [29, Lemma 6.1 (a)] that there are triangle
equivalences
L ∼ / D(Λ)
F =? ⊗Γ̃ M : D(Γ̃) D(Mod Λ)
O O O
? ? ?
∼
Df d (Γ̃) / Df d (Λ) Db (mod Λ)
O O O
? ? ?
∼
per(Γ̃) / per(Λ) K b (proj Λ)

These equivalences take Γ̃ to M . The following special case of Theorem 3.1 is

a consequence.
Corollary 5.1. The number of indecomposable direct summands of M equals
the rank of the Grothendieck group of K b (proj Λ). In particular, any two basic
silting objects of K b (proj Λ) have the same number of indecomposable direct
summands.
Proof. The number of indecomposable direct summands of M equals the rank
of the Grothendieck group of mod Γ, which equals the rank of the Grothendieck
group of Df d (Γ̃) ∼
= Db (mod Λ) since mod Γ is the heart of a bounded t-structure
√
(Lemma 3.3).
Write M = M1 ⊕ . . . ⊕ Mr with Mi indecomposable. Suppose that X1 , . . . , Xr
are objects in Db (mod Λ) such that their endomorphism algebras R1 , . . . , Rr
are division algebras and that the following formula holds for i, j = 1, . . . , r
and m ∈ Z
(
Rj Rj if i = j and m = 0,
Hom(Mi , Σm Xj ) =
0 otherwise.
Then up to isomorphism, the objects X1 , . . . , Xr are sent by the derived equiv-
L
alence ? ⊗Γ̃ M to a complete set of pairwise non-isomorphic simple Γ-modules,
see Section 4.4.
Lemma 5.2. (a) Let X1′ , . . . , Xr′ be objects of Db (mod Λ) such that the fol-
lowing formula holds for 1 ≤ i, j ≤ r and m ∈ Z
(
m ′ Rj Rj if i = j and m = 0,
Hom(Mi , Σ Xj ) =
0 otherwise.
Then Xi ∼ = Xi′ for any i = 1, . . . , r.
(b) Let M1′ , . . . , Mr′ be objects of K b (proj Λ) such that the following formula
holds for 1 ≤ i, j ≤ r and m ∈ Z
(
′ m Rj Rj if i = j and m = 0,
Hom(Mi , Σ Xj ) =
0 otherwise.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 417

Then Mi ∼
= Mi′ for any i = 1, . . . , r.
√
Proof. This follows from the corresponding result in D(Γ̃), see Section 4.4.

5.2. From co-t-structures to silting objects. Let (C≥0 , C≤0 ) be a

bounded co-t-structure of K b (proj Λ). By Lemma 3.4, the co-heart A =
C≥0 ∩ C≤0 is a silting subcategory of K b (proj Λ). Since Λ is a silting object
of K b (proj Λ), it follows from Theorem 3.1 that A has an additive generator,
say M , i.e. A = add(M ). Then M is a silting object in K b (proj Λ). Define
φ14 (C≥0 , C≤0 ) = M.

5.3. From t-structures to simple-minded collections. Let (C ≤0 , C ≥0 )

be a bounded t-structure of Db (mod Λ) with length heart A. Boundedness
implies that the Grothendieck group of A is isomorphic to the Grothendieck
group of Db (mod Λ), which is free, say, of rank r. Therefore, A has precisely
r isomorphism classes of simple objects, say X1 , . . . , Xr . By Lemma 3.3 (f),
X1 , . . . , Xr is a simple-minded collection in Db (mod Λ). Define
φ23 (C ≤0 , C ≥0 ) = {X1 , . . . , Xr }.

5.4. From silting objects to simple-minded collections, t-

structures and co-t-structures. Let M be a silting object of K b (proj Λ).
Define full subcategories of C
C ≤0 = {N ∈ Db (mod Λ) | Hom(M, Σm N ) = 0, ∀ m > 0},
C ≥0 = {N ∈ Db (mod Λ) | Hom(M, Σm N ) = 0, ∀ m < 0},
C≤0 = the additive closure of the extension closure
of Σm M , m ≥ 0 in K b (proj Λ),
C≥0 = the additive closure of the extension closure
of Σm M , m ≤ 0 in K b (proj Λ).
Lemma 5.3. (a) The pair (C ≤0 , C ≥0 ) is a bounded t-structure on
b
D (mod Λ) whose heart is equivalent to mod Γ for Γ = End(M ).
Write M = M1 ⊕ . . . ⊕ Mr and let X1 , . . . , Xr be the corresponding
simple objects of the heart with endomorphism algebras R1 , . . . , Rr
respectively. Then the following formula holds for 1 ≤ i, j ≤ r and
m∈Z
(
m Rj Rj if i = j and m = 0,
Hom(Mi , Σ Xj ) =
0 otherwise.

(b) The pair (C≥0 , C≤0 ) is a bounded co-t-structure on K b (proj Λ) whose

co-heart is add(M ).
The first statement of part (a) is proved by Keller and Vossieck [33] in the
case when Λ is the path algebra of a Dynkin quiver and by Assem, Souto and
Trepode [5] in the case when Λ is hereditary.

Documenta Mathematica 19 (2014) 403–438

418 Steffen Koenig, Dong Yang

Proof. Let Γ̃ be the truncated dg endomorphism algebra of M , see Section 5.1.

Then per(Γ̃) has a standard bounded co-t-structure (P≥0 , P≤0 ) and Df d (Γ̃)
has a standard bounded t-structure (Df≤0 ≥0
d , Df d ) with heart equivalent to mod Γ.
L
One checks that the triangle equivalence ? ⊗Γ̃ M takes (P≥0 , P≤0 ) to (C≥0 , C≤0 )
√
and it takes (Df≤0 ≥0
d , Df d ) to (C
≤0
, C ≥0 ).
Define
φ31 (M ) = (C ≤0 , C ≥0 ),
φ41 (M ) = (C≥0 , C≤0 ),
φ21 (M ) = {X1 , . . . , Xr }.
5.5. From simple-minded collections to t-structures. Let X1 , . . . , Xr
be a simple-minded collection of Db (mod Λ). Let C ≤0 (respectively, C ≥0 ) be
the extension closure of {Σm Xi | i = 1, . . . , r, m ≥ 0} (respectively, {Σm Xi |
i = 1, . . . , r, m ≤ 0}) in Db (mod Λ).
Proposition 5.4. The pair (C ≤0 , C ≥0 ) is a bounded t-structure on Db (mod Λ).
Moreover, the heart of this t-structure is a length category with simple objects
X1 , . . . , Xr . The same results hold true with Db (mod Λ) replaced by a Hom-
finite Krull–Schmidt triangulated category C.
Proof. The first two statements are [3, Corollary 3 and Proposition 4]. The
√
proof there still works if we replace Db (mod Λ) by C.
Define
φ32 (X1 , . . . , Xr ) = (C ≤0 , C ≥0 ).
Later we will show that the heart of this t-structure always is equivalent to
the category of finite-dimensional modules over a finite-dimensional algebra
(Corollary 6.2). This was proved by Al-Nofayee for self-injective algebras Λ,
see [3, Theorem 7].
Corollary 5.5. Any two simple-minded collections in Db (mod Λ) have the
same cardinality.
Proof. By Proposition 5.4, the cardinality of a simple-minded collection equals
√
the rank of the Grothendieck group of Db (mod Λ). The assertion follows.
5.6. From simple-minded collections to silting objects. Let
X1 , . . . , Xr be a simple-minded collection in Db (mod Λ). We will construct a
silting object ν −1 T of K b (proj Λ) following a method of Rickard [43]. Then we
define
φ12 (X1 , . . . , Xr ) = ν −1 T.
The same construction is studied by Keller and Nicolás [32] in the context
of positive dg algebras. In the case of Λ being hereditary, Buan, Reiten and
Thomas [17] give an elegant construction of ν −1 (T ) using the Braid group

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 419

action on exceptional sequences. Unfortunately, their construction cannot be

generalised.
Let R1 , . . . , Rr be the endomorphism algebras of X1 , . . . , Xr , respectively.
(0) (n−1)
Set Xi = Xi . Suppose Xi is constructed. For i, j = 1, . . . , r and m < 0,
(n−1)
let B(j, m, i) be a basis of Hom(Σm Xj , Xi ) over Rj . Put
(n−1)
M M M
Zi = Σm X j
m<0 j B(j,m,i)

(n−1) (n−1) (n−1)

and let αi : Zi → be the map whose component corresponding
Xi
to f ∈ B(j, m, i) is exactly f .
(n) (n−1)
Let Xi be a cone of αi and form the corresponding triangle
(n−1) (n−1)
αi βi
(n−1) / X (n−1) / X (n) / ΣZ (n−1) .
Zi i i i

Inductively, a sequence of morphisms in D(Mod Λ) is constructed:

(0) (n−1)
βi βi
Xi
(0) / X (1) / ... / X (n−1) / X (n) / ....
i i i

Let Ti be the homotopy colimit of this sequence. That is, up to isomorphism,

Ti is defined by the following triangle
(n) id−β /L (n) / Ti / ΣL (n)
.
L
n≥0 Xi n≥0 Xi n≥0 Xi

Here β = (βmn ) is the square matrix with rows and columns labeled by non-
(n)
negative integers and with entries βmn = βi if n + 1 = m and 0 otherwise.
These properties of Ti ’s were proved by Rickard in [43] for symmetric algebras Λ
over algebraically closed fields. Rickard remarked that they hold for arbitrary
fields, see [43, Section 8]. In fact, his proofs verbatim carry over to general
finite-dimensional algebras.
Lemma 5.6. (a) ([43, Lemma 5.4]) For 1 ≤ i, j ≤ r, and m ∈ Z,
(
m (Rj )Rj if i = j and m = 0,
Hom(Xj , Σ Ti ) =
0 otherwise.
(b) ([43, Lemma 5.5]) For each 1 ≤ i ≤ r, Ti is quasi-isomorphic to a
bounded complex of finitely generated injective Λ-modules.
(c) ([43, Lemma 5.8]) Let C be an object of D− (mod Λ). If
Hom(C, Σm Ti ) = 0 for all m ∈ Z and all 1 ≤ i ≤ r, then C = 0.
From now on we assume that Ti is a bounded complex of finitely generated
injective Λ-modules. Recall from Section 2.3 that the Nakayama functor ν
and the inverse Nakayama functor ν −1 are quasi-inverse triangle equivalences
between K b (proj Λ) and K b (inj Λ) The following is a consequence of Lemma 5.6
and the Auslander–Reiten formula.

Documenta Mathematica 19 (2014) 403–438

420 Steffen Koenig, Dong Yang

Lemma 5.7. (a) For 1 ≤ i, j ≤ r, and m ∈ Z,

(
−1 m Rj Rj if i = j and m = 0,
Hom(ν Ti , Σ Xj ) =
0 otherwise.

(b) For each 1 ≤ i ≤ r, ν −1 Ti is a bounded complex of finitely generated

projective Λ-modules.
(c) Let C be an object of D− (mod Λ). If Hom(ν −1 Ti , Σm C) = 0 for all
m ∈ Z and all 1 ≤ i ≤ r, then C = 0.
Put T = ri=1 Ti and ν −1 T = ri=1 ν −1 Ti .
L L

Lemma 5.8. We have Hom(ν −1 T, Σm T ) = 0 for m < 0. Equivalently,

Hom(ν −1 T, Σm ν −1 T ) = Hom(T, Σm T ) = 0 for m > 0.
Proof. Same as the proof of [43, Lemma 5.7], with the Ti in the first entry √
of
Hom there replaced by ν −1 Ti .

It follows from Lemma 5.7 (c) that ν −1 T generates K b (proj Λ). Combining this
with Lemma 5.8 implies
Proposition 5.9. ν −1 T is a silting object of K b (proj Λ).

Rickard’s construction was originally motivated by constructing tilting com-

plexes over symmetric algebras which yield certain derived equivalences, see
[43, Theorem 5.1]. His work was later generalised by Al-Nofayee to self-injective
algebras, see [2, Theorem 4].

5.7. From co-t-structures to t-structures. Let (C≥0 , C≤0 ) be a bounded

co-t-structure of K b (proj Λ). Let
C ≤0 = {N ∈ Db (mod Λ) | Hom(M, N ) = 0, ∀ M ∈ Σ−1 C≥0 }
C ≥0 = {N ∈ Db (mod Λ) | Hom(M, N ) = 0, ∀ M ∈ ΣC≤0 }.

Lemma 5.10. The pair (C ≤0 , C ≥0 ) is a bounded t-structure on Db (mod Λ) with

length heart.
√
Proof. Because (C ≤0 , C ≥0 ) = φ31 ◦ φ14 (C≥0 , C≤0 ).

By definition (C ≤0 , C ≥0 ) is right orthogonal to the given co-t-structure in the

sense of Bondarko [11, Definition 2.5.1]. Define
φ34 (C≥0 , C≤0 ) = (C ≤0 , C ≥0 ).
If Λ has finite global dimension, then K b (proj Λ) is identified with Db (mod Λ).
As a consequence, C≤0 = C ≤0 and C ≥0 = νC≥0 . Thus the t-structure (C ≤0 , C ≥0 )
is right adjacent to the given co-t-structure (C≥0 , C≤0 ) in the sense of Bon-
darko [12, Definition 4.4.1].

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 421

5.8. Some remarks. Some of the maps φij are defined in more general setups:
– φ14 and φ41 are defined for all triangulated categories, with silt-
ing objects replaced by silting subcategories, by Proposition 3.5 and
Lemma 3.4, see also [12, 31, 39].
– φ23 is defined for all triangulated categories, with simple-minded col-
lections allowed to contain infinitely many objects (Lemma 3.3).
– φ32 is defined for all algebraic triangulated categories (see [32]) and for
Hom-finite Krull–Schmidt triangulated categories (see Proposition 5.4).
– φ21 and φ31 are defined for all algebraic triangulated categories (replac-
ing K b (proj Λ)), with Db (mod Λ) replaced by a suitable triangulated
category; then we may follow the arguments in Sections 4.1 and 5.4.
– φ34 is defined for all algebraic triangulated categories (replacing
K b (proj Λ)), with Db (mod Λ) replaced by a suitable triangulated cate-
gory. Then we may follow the argument in Section 5.7.
– φ12 is defined for finite-dimensional non-positive dg algebras, since
these dg algebras behave like finite-dimensional algebras from the per-
spective of derived categories. Similarly, φ12 is defined for homologi-
cally smooth non-positive dg algebras, see [31].

6. The correspondences are bijections

Let Λ be a finite-dimensional K-algebra. In the preceding section we defined
the maps φij . In this section we will show that they are bijections. See [5, 46]
for related work, focussing on piecewise hereditary algebras.
Theorem 6.1. The φij ’s defined in Section 5 are bijective. In particular, there
are one-to-one correspondences between
(1) equivalence classes of silting objects in K b (proj Λ),
(2) equivalence classes of simple-minded collections in Db (mod Λ),
(3) bounded t-structures on Db (mod Λ) with length heart,
(4) bounded co-t-structures on K b (proj Λ).
There is an immediate consequence:
Corollary 6.2. Let A be the heart of a bounded t-structure on Db (mod Λ).
If A is a length category, then A is equivalent to mod Γ for some finite-
dimensional algebra Γ.
Proof. By Theorem 6.1, such a t-structure is of the form φ31 (M ) for some silting
√
object M of K b (proj Λ). The result then follows from Lemma 5.3 (a).
The proof of the theorem is divided into several lemmas, which are consequences
of the material collected in the previous sections.
Lemma 6.3. The maps φ14 and φ41 are inverse to each other.
Proof. Let M be a basic silting object. The definitions of φ14 and φ41 and
Lemma 5.3 (b) imply that φ14 ◦ φ41 (M ) ∼ = M.
Let (C≥0 , C≤0 ) be a bounded co-t-structure on K b (proj Λ). It follows from
√
Lemma 3.4 that φ41 ◦ φ14 (C≥0 , C≤0 ) = (C≥0 , C≤0 ).

Documenta Mathematica 19 (2014) 403–438

422 Steffen Koenig, Dong Yang

Recall from Section 5.8 that φ14 and φ41 are defined in full generality.
Lemma 6.3 holds in full generality as well, see [39, Corollary 5.8] and [31].
Lemma 6.4. The maps φ21 and φ12 are inverse to each other.
Proof. This follows from the Hom-duality: Lemma 5.7 (a), Lemma 5.3 (a) and
√
Lemma 5.2.
Lemma 6.5. The maps φ23 and φ32 are inverse to each other.
Proof. Let X1 , . . . , Xr be a simple-minded collection in Db (mod Λ). It follows
from Proposition 5.4 that φ23 ◦ φ32 (X1 , . . . , Xr ) = {X1 , . . . , Xr }.
Let (C ≤0 , C ≥0 ) be a bounded t-structure on Db (mod Λ) with length heart. √ It
follows from Lemma 3.3 that φ32 ◦ φ23 (C ≤0 , C ≥0 ) = (C ≤0 , C ≥0 ).
Lemma 6.6. For a triple i, j, k such that φij , φjk and φik are defined, there is
the equality φij ◦ φjk = φik . In particular, φ31 and φ34 are bijective.
Proof. In view of the preceding three lemmas, it suffices to prove φ23 ◦φ31 = φ21
√
and φ31 ◦ φ14 = φ34 , which is clear from the definitions.

7. Mutations and partial orders

In this section we introduce mutations and partial orders on the four concepts
in Section 3, and we show that the maps defined in Section 5 commute with
mutations and preserve the partial orders.
Let C be a Hom-finite Krull–Schmidt triangulated category with suspension
functor Σ.
7.1. Silting objects. We follow [1, 18] to define silting mutation. Let M be
a silting object in C. We assume that M is basic and M = M1 ⊕ . . . ⊕ Mr is a
decomposition into indecomposable objects. Let i = 1, . . . , r. The
L left mutation
of M at the direct summand Mi is the object µ+ i (M ) = M i
′
⊕ j6=i Mj where
Mi′ is the cone of the minimal left add( j6=i Mj )-approximation of Mi
L

Mi / E.

Similarly one can define the right mutation µ−

i (M ).

Theorem 7.1. ([1, Theorem 2.31 and Proposition 2.33]) The objects µ+ i (M )
and µ− (M ) are silting objects. Moreover, µ +
◦ µ −
(M ) ∼
= M ∼
= µ −
◦ µ +
(M ).
i i i i i

Let silt C be the set of isomorphism classes of basic tilting objects of C. The
silting quiver of C has the elements in silt C as vertices. For P, P ′ ∈ silt C, there
are arrows from P to P ′ if and only if P ′ is obtained from P by a left mutation,
in which case there is precisely one arrow. See [1, Section 2.6].
For P, P ′ ∈ silt C, define P ≥ P ′ if Hom(P, Σm P ′ ) = 0 for any m > 0. According
to [1, Theorem 2.11], ≥ is a partial order on silt C.
Theorem 7.2. ([1, Theorem 2.35]) The Hasse diagram of (silt C, ≥) is the
silting quiver of C.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 423

Next we define (a generalisation of) the Brenner–Butler tilting module for

a finite-dimensional algebra, and show that it is a left mutation of the free
module of rank 1. The corresponding right mutation is the Okuyama–Rickard
complex, see [1, Section 2.7]. Let Λ be a finite-dimensional basic algebra and
1 = e1 + . . . + en be a decomposition of the unity into the sum of primitive
idempotents and Λ = P1 ⊕ . . . ⊕ Pn the corresponding decomposition of the
free module of rank 1. Fix i = 1, . . . , n and let Si be the corresponding simple
module and let Si+ = D(Λ/Λ(1 − ei )Λ). Assume that
· Si+ is not injective,
−1 +
· the projective dimension of τmod Λ Si is at most 1.

Definition 7.3. Define the BB tilting module with respect to i by

M
−1 +
T = τmod Λ Si ⊕ Pj .
j6=i

We call it the APR tilting module if Λ/Λ(1 − ei )Λ is projective as a Λ-module.

When Λ/Λ(1 − ei )Λ is a division algebra (i.e. there are no loops in the quiver
of Λ at the vertex i), this specialises to the ‘classical’ BB tilting module [13]
and APR tilting module [6]. The following proposition generalises [1, Theorem
2.53].
Proposition 7.4. (a) T is isomorphic to the left mutation µ+
i (Λ) of Λ.
(b) T is a tilting Λ-module of projective dimension at most 1.
Proof. We modify the proof in [1]. Take a minimal injective copresentation of
Si+ :
f
0 / S+ / D(ei Λ) / I.
i

Since Ext1Λ (Si , Si+ ) = Ext1Λ/Λ(1−ei )Λ (Si , Si+ ) = 0, it follows that the injective
module I belongs to add D((1 − ei )Λ). Applying the inverse Nakayama functor
−1
νmod Λ yields an exact sequence
−1
νmod Λf
Pi / ν −1 I / τ −1 S + / 0.
mod Λ mod Λ i
−1
Moreover, νmod Λ f is a minimal left approximation of Pi in add(Pj , j 6= i).
−1 + −1
Since the projective dimension of τmod Λ Si is at most 1, it follows that νmod Λ f
is injective. This completes the proof for (a). √
(b) follows from [1, Theorem 2.32].
7.2. Simple-minded collections. Let X1 , . . . , Xr be a simple-minded col-
lection in C and fix i = 1, . . . , r. Let Xi denote the extension closure of Xi in C.
Assume that for any j the object Σ−1 Xj admits a minimal left approximation
gj : Σ−1 Xj → Xij in Xi .
Definition 7.5. The left mutation µ+ i (X1 , . . . , Xr ) of X1 , . . . , Xr at Xi is a
new collection X1′ , . . . , Xr′ such that Xi′ = ΣXi and Xj′ (j = 6 i) is the cone of

Documenta Mathematica 19 (2014) 403–438

424 Steffen Koenig, Dong Yang

the above left approximation

gj
Σ−1 Xj / Xij .

Similarly one defines the right mutation µ−

i (X1 , . . . , Xr ).
This generalises Kontsevich–Soibelman’s mutation of spherical collections [38,
Section 8.1] and appeared in [35] in the case of derived categories of acyclic
quivers.
Proposition 7.6. (a) µ+ ◦ µ− (X1 , . . . , Xr ) ∼
i i = (X1 , . . . , Xr ) ∼
= µ− ◦ i
µ+
i (X1 , . . . , Xr ).
(b) Assume that
· for any j 6= i the object Σ−1 Xj admits a minimal left approxima-
tion gj : Σ−1 Xj → Xij in Xi ,
· the induced map Hom(gj , Xi ) : Hom(Xij , Xi ) → Hom(Σ−1 Xj , Xi )
is injective,
· the induced map Hom(gj , ΣXi ) : Hom(Xij , ΣXi ) →
Hom(Σ−1 Xj , ΣXi ) is injective.
Then the collection µ+ i (X1 , . . . , Xr ) is simple-minded.
(c) Assume that
· for any j 6= i the object Xj admits a minimal right approximation
gj− : Σ−1 Xij −
→ Xj in Σ−1 Xi ,
· the induced map Hom(Xi , Σgj− ) : Hom(Xi , Xij −
) → Hom(Xi , ΣXj )
is injective,
· the induced map Hom(Xi , Σ2 gj− , ) : Hom(Xi , ΣXij −
) →
2
Hom(Xi , Σ Xj ) is injective.
Then the collection µ− i (X1 , . . . , Xr ) is simple-minded.
Proof. (a) Because in the triangle
gj gj−
Σ−1 Xj / Xij / Xj′ / Xj

gj is a minimal left approximation of Σ−1 Xj in Xi if and only if gj− is a minimal

right approximation of Xj in Xi = Σ−1 (ΣXi ).
(b) and (c) The proof uses long exact Hom sequences induced from the defining √
triangles of the Xj′ . We leave it to the reader.
Remark 7.7. In the course of the proof of Proposition 7.6 (b) and (c), one
notices that the collection of endomorphism algebras of the mutated simple-
minded collection is the same as that of the given simple-minded collection.
If Hom(Xi , ΣXi ) = 0, then Xi = add(Xi ). In this case, all six assumptions in
Proposition 7.6 (b) and (c) are satisfied.
Lemma 7.8. Let Λ be a finite-dimensional algebra and let X1 , . . . , Xr be a
simple-minded collection in Db (mod Λ). Let i = 1, . . . , r. Then the left muta-
tion µ+ −
i (X1 , . . . , Xr ) and the right mutation µi (X1 , . . . , Xr ) are again simple-
minded collections.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 425

Proof. We will show that the three assumptions in Proposition 7.6 (b) are
satisfied, so the left-mutated collection µ+ i (X1 , . . . , Xr ) is a simple-minded col-
lection. The case for µ− i (X 1 , . . . , X r ) is similar.
By Proposition 5.4, X1 , . . . , Xr are the simple objects in the heart of a bounded
t-structure on Db (mod Λ). Moreover, by Corollary 6.2, the heart is equivalent to
mod Γ for some finite-dimensional algebra Γ. We identify mod Γ with the heart
via this equivalence. In this way we consider X1 , . . . , Xr as simple Γ-modules.
By [8, Section 3.1], there is a triangle functor
real : Db (mod Γ) → Db (mod Λ)
such that
– restricted to mod Γ, real is the identity;
– for M, N ∈ mod Γ, the induced map
Ext1Γ (M, N ) = HomDb (mod Γ) (M, ΣN ) → HomDb (mod Λ) (M, ΣN )
is bijective;
– for M, N ∈ mod Γ, the induced map
Ext2Γ (M, N ) = HomDb (mod Γ) (M, Σ2 N ) → HomDb (mod Λ) (M, Σ2 N )
is injective.
For j = 1, . . . , r, there is a short exact sequence
0 / ΩXj / Pj / Xj / 0,

where Pj is the projective cover of Xj and ΩXj is the first syzygy of Xj . Let
Xi be the extension closure of Xi in mod Γ (by the second property of real listed
in the preceding paragraph, this is the same as the extension closure of Xi in
Db (mod Λ)) and let Xij denote the maximal quotient of ΩXj belonging to Xi .
There is the following push-out diagram
0 / ΩXj / Pj / Xj /0

 
ξ: 0 / Xij / Xj′ / Xj /0

(a) Suppose we are given an object Y of Xi and a short exact sequence

η: 0 /Y /Z / Xj / 0.

Then there is a commutative diagram

0 / ΩXj / Pj / Xj /0

 
η: 0 /Y /Z / Xj / 0.

Because Xij is the maximal quotient of ΩXj belonging to Xi , this morphism

of short exact sequences factors through ξ. In other words, the morphism
gj : Xj → ΣXij corresponding to ξ is a minimal left ΣXi -approximation.

Documenta Mathematica 19 (2014) 403–438

426 Steffen Koenig, Dong Yang

(b) The dimension of the space HomDb (mod Λ) (ΣXij , ΣXi ) ∼ = HomΓ (Xij , Xi )
over End(Xi ) equals the number of indecomposable direct summands
of top(Xij ), which clearly equals the dimension of Ext1Γ (Xj , Xi ) ∼ =
HomDb (mod Λ) (Xj , ΣXi ) over End(Xi ). Therefore the induced map
Hom(gj , ΣXi ) : HomDb (mod Λ) (ΣXij , ΣXi ) −→ HomDb (mod Λ) (Xj , ΣXi )
is injective since by (a) it is surjective.
(c) First observe that the following diagram is commutative
HomΛ (gj ,Σ2 Xi )
HomDb (mod Λ) (ΣXij , Σ2 Xi ) / HomDb (mod Λ) (Xj , Σ2 Xi )
O O
real real
2
HomΓ (gj ,Σ Xi )
HomDb (mod Γ) (ΣXij , Σ2 Xi ) / HomDb (mod Γ) (Xj , Σ2 Xi )

The left vertical map is a bijection and the right vertical map is injective, so
to prove the injectivity of HomΛ (gj , Σ2 Xi ) it suffices to prove the injectivity of
HomΓ (gj , Σ2 Xi ). Writing
HomDb (mod Γ) (ΣXij , Σ2 Xi ) = Ext1Γ (Xij , Xi ) = HomΓ (ΩXij , Xi )
and
HomDb (mod Γ) (Xj , Σ2 Xi ) = Ext2Γ (Xj , Xi ) = Ext1Γ (ΩXj , Xi ) = HomΓ (Ω2 Xj , Xi ),
we see that HomΓ (gj , Σ2 Xi ) is HomΓ (α, Xi ), where α is defined by the following
commutative diagram
0 / Ω2 X j / P0 / ΩXj /0

α β γ
  
0 / ΩXij / Q0 / Xij / 0,

Here, P 0 and Q0 are projective covers of ΩXj and Xij , respectively, and γ is
the canonical quotient map. As the map γ is surjective, the map β is a split
epimorphism. By the snake lemma, there is an exact sequence
ker(γ) / cok(α) / 0.

Since Xij is the maximal quotient of ΩXj in Xi , it follows that

HomΓ (ker(γ), Xi ) = 0, and hence HomΓ (cok(α), Xi ) = 0. Therefore
√
HomΓ (α, Xi ) is injective.
For two simple-minded collections {X1 , . . . , Xr } and {X1′ , . . . , Xr′ } of C, define
{X1 , . . . , Xr } ≥ {X1′ , . . . , Xr′ }
if Hom(Xi′ , Σm Xj ) = 0 for any m < 0 and any i, j = 1, . . . , r.
Proposition 7.9. The relation ≥ defined above is a partial order on the set of
equivalence classes of simple-minded collections of C.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 427

Proof. The reflexivity is clear by the definition of a simple-minded collec-

tion. Next we show the antisymmetry and transitivity. Let {X1 , . . . , Xr } and
{X1′ , . . . , Xr′ } be two simple-minded collections of C and let (C ≤0 , C ≥0 ) and
(C ′≤0 , C ′≥0 ) be the corresponding t-structures given in Proposition 5.4 (the
general case). Then

{X1 , . . . , Xr } ≥ {X1′ , . . . , Xr′ }

⇔ Hom(Xi′ , Σm Xj ) = 0 for any m < 0 and i, j = 1, . . . , r
′
⇔ Hom(Σm Xi′ , Σm Xj ) = 0 for any m < 0, m′ ≥ 0
and i, j = 1, . . . , r
⇔ C ′≤0 ⊥ Σ−1 C ≤0
⇔ C ′≤0 ⊆ C ≤0 .

(a) If {X1 , . . . , Xr } ≥ {X1′ , . . . , Xr′ } and {X1′ , . . . , Xr′ } ≥ {X1 , . . . , Xr }, then

(C ≤0 , C ≥0 ) = (C ′≤0 , C ′≥0 ). In particular, the two t-structures have the same
heart. Therefore, both {X1 , . . . , Xr } and {X1′ , . . . , Xr′ } are complete sets of
pairwise non-isomorphic simple objects of the same abelian category, and hence
they are equivalent.
(b) Let {X1′′ , . . . , Xr′′ } be a third simple-minded collection of C, with corre-
sponding t-structure (C ′′≤0 , C ′′≥0 ). Suppose {X1 , . . . , Xr } ≥ {X1′ , . . . , Xr′ } and
{X1′ , . . . , Xr′ } ≥ {X1′′ , . . . , Xr′′ }. Then C ′′≤0 ⊆ C ′≤0 ⊆ C ≤0 . Consequently, √
{X1 , . . . , Xr } ≥ {X1′′ , . . . , Xr′′ }.

7.3. t-structures. Let (C ≤0 , C ≥0 ) be a bounded t-structure of C such that

the heart A is a length category which has only finitely many simple objects
S1 , . . . , Sr up to isomorphism. Then {S1 , . . . , Sr } is a simple-minded collection.
Let F = Si be the extension closure of Si in A and let T = ⊥ Si be the left
perpendicular category of Si in A. It is easy to show that (T , F ) is a torsion
pair of A. Define the left mutation µ+ i (C
≤0
, C ≥0 ) = (C ′≤0 , C ′≥0 ) by

C ′≤0 = {M ∈ C | H m (M ) = 0 for m > 0 and H 0 (M ) ∈ T },

C ′≥0 = {M ∈ C | H m (M ) = 0 for m < −1 and H −1 (M ) ∈ F }.

Similarly one defines the right mutation µ− i (C

≤0
, C ≥0 ). These mutations pro-
vide an effective method to compute the space of Bridgeland’s stability condi-
tions on C by gluing different charts, see [14, 48].

Proposition 7.10. The pairs µ+ i (C

≤0
, C ≥0 ) and µ−
i (C
≤0
, C ≥0 ) are bounded t-
+ ≤0 ≥0
structures of C. The heart of µi (C , C ) has a torsion pair (ΣF , T ) and
the heart of µ−i (C
≤0
, C ≥0 ) has a torsion pair (Si⊥ , Σ−1 Si ). Moreover, µ+ i ◦
− ≤0 ≥0 ≤0 ≥0 − + ≤0 ≥0
µi (C , C ) = (C , C ) = µi ◦ µi (C , C ).

Proof. This follows from [24, Proposition 2.1, Corollary 2.2] and [14, Proposi-
√
tion 2.5].

Documenta Mathematica 19 (2014) 403–438

428 Steffen Koenig, Dong Yang

In general the heart of the mutation of a bounded t-structure with length heart
is not necessarily a length category. For an example, let Q be the quiver
:1o 2
and consider the bounded derived category C = Db (nil. rep Q) of finite-
dimensional nilpotent representations of Q. Let S1 and S2 be the one-
dimensional nilpotent representations associated to the two vertices. Let
F = S1 be the extension closure of S1 and T = ⊥ F = {M ∈ nil. rep Q |
top(M ) ∈ add(S2 )}. Then the heart A′ of the left mutation at 1 of the stan-
dard t-structure has a torsion pair (ΣF , T ). Due to nil. rep Q being hereditary,
there are no extensions of ΣF by T , and hence any indecomposable object of
A′ belongs to either T or ΣF . Suppose that A′ is a length category. Then A′
has two isomorphism classes of simple modules, which respectively belong to
T and ΣF , say S2′ ∈ T and S1′ ∈ ΣF . For n ∈ N define an indecomposable
object Mn in T as
tr
(0,...,0,1)
Jn (0) 9ko k,
where Jn (0) is the (upper triangular) Jordan block of size n and with eigenvalue
0. There are no morphisms from S1′ to Mn for any n. Suppose that the Loewy
length of S2′ in A is l. Then for n > l, any morphism from S2′ to Mn factors
through radn−l Mn which lies in F , and hence the morphism has to be zero.
Therefore Mn (n > l), considered as an object in A′ , does not have finite
length, a contradiction.
For two bounded t-structures (C ≤0 , C ≥0 ) and (C ′≤0 , C ′≥0 ) on C, define
(C ≤0 , C ≥0 ) ≥ (C ′≤0 , C ′≥0 )
if C ≤0 ⊇ C ′≤0 . This defines a partial order on the set of bounded t-structures
on C.

7.4. Co-t-structures. Let (C≥0 , C≤0 ) be a bounded co-t-structure of C. As-

sume that the co-heart admits a basic additive generator M = M1 ⊕ . . . ⊕ Mr
with Mi indecomposable. Then M is a silting object of C. Let i = 1, . . . , r.
′
Define C≤0 as the additive closure of the extension closure of Σm Mj , j 6= i, and
m+1 ′ ′
Σ Mi for m ≥ 0 and define C≥0 as the left perpendicular category of ΣC≤0 .
+ ′ ′
The left mutation µi (C≥0 , C≤0 ) is defined as the pair (C≥0 , C≤0 ). Similarly one
defines the right mutation µ− i (C≥0 , C≤0 ).

Proposition 7.11. The pairs µ+ −

i (C≥0 , C≤0 ) and µi (C≥0 , C≤0 ) are bounded
+
co-t-structures on C. Moreover, µi ◦ µi (C≥0 , C≤0 ) = (C≥0 , C≤0 ) = µ−
−
i ◦
µ+
i (C ≥0 , C ≤0 ).
Proof. This can be proved directly. Here we alternatively make use of the re-
sults in Sections 3.1 and 7.1. Recall from Theorem 7.1 that there is a mutated
silting object µ+i (M ). It is straightforward to check, using the defining tri-
angle for µ+i (M ), that µ+
i (C≥0 , C≤0 ) is the bounded co-t-structure associated

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 429

to µ+ −
i (M ) as defined in Proposition 3.5, and similarly for µi . The second
√
statement follows from Theorem 7.1.

′ ′
For two bounded co-t-structures (C≥0 , C≤0 ) and (C≥0 , C≤0 ) on C, define
′ ′
(C≥0 , C≤0 ) ≥ (C≥0 , C≤0 )
′
if C≤0 ⊇ C≤0 . This defines a partial order on the set of bounded co-t-structures
on C.

7.5. The bijections commute with mutations. Let Λ a finite-dimensional

algebra over K.

Theorem 7.12. The φij ’s defined in Section 5 commute with the left and right
mutations defined in previous subsections.

A priori it it not known that the heart of the mutation of a bounded t-structure
with length heart is again a length category. So the theorem becomes well-
stated only when the proof has been finished.

Proof. In view of Lemma 6.6, Theorem 7.1, and Propositions 7.6, 7.10 and 7.11,
it suffices to prove that φ41 , φ31 and φ23 commute with the corresponding left
mutations.
(a) φ41 commutes with µ+ i : this was already shown in the proof of Proposi-
tion 7.11.
(b) φ31 commutes with µ+ i : Let M = M1 ⊕ . . . ⊕ Mr be a silting object with Mi
indecomposable and (C ≤0 , C ≥0 ) = φ31 (M ). We want to show µ+ i (C
≤0
, C ≥0 ) =
+
φ31 (µi (M )).
Let Γ̃ be the truncated dg endomorphism algebra of M as in Section 5.1. Then
L
there is a triangle equivalence F =? ⊗Γ̃ M : Df d (Γ̃) → Db (mod Λ), which takes
Γ̃ to M and takes the standard t-structure (D≤0 , D≥0 ) on Df d (Γ̃) to (C ≤0 , C ≥0 ).
There is a decomposition 1 = e1 + . . . + er , where e1 , . . . , er are (not necessarily
primitive) idempotents of Γ̃ such that F takes ej Γ̃ to Mj for 1 ≤ j ≤ r.
Let Γ = H 0 (Γ̃) and π : Γ̃ → Γ be the canonical projection. By abuse of
notation, write e1 = π(e1 ), . . . , er = π(er ). Then e1 Γ, . . . , er Γ are indecom-
posable projective Γ-modules. Let S1 , . . . , Sr be the corresponding simple
modules. Recall that the heart of the t-structure (D≤0 , D≥0 ) is mod Γ. Let
F = add(Si ) ⊆ mod Γ and T = ⊥ Si . Define D′≤0 (respectively, D′≥0 ) to be
the extension closure of ΣD≤0 and T (respectively, of ΣF and D≥0 ). Then
F (D′≤0 , D′≥0 ) = µ+
i (C
≤0
, C ≥0 ).
The left mutation of Γ̃ at ei Γ̃ is µ+
L
i (Γ̃) = Qi ⊕ j6=i ej Γ̃, where Qi is defined
by the triangle

f
(7.1) ei Γ̃ /E / P′ / Σe Γ̃ ,
i i

Documenta Mathematica 19 (2014) 403–438

430 Steffen Koenig, Dong Yang

where f is a minimal left add( j6=i ej Γ̃)-approximation. Then F (µ+

L
i (Γ̃)) =
µ+
i (M ). Define

D′′≤0 = {N ∈ Df d (Γ̃) | Hom(µ+ m

i (Γ̃), Σ N ) = 0, ∀m > 0},
D′′≥0 = {N ∈ Df d (Γ̃) | Hom(µ+ m
i (Γ̃), Σ N ) = 0, ∀m < 0}.

Thus showing µ+ i (C
≤0
, C ≥0 ) = φ31 (µ+
i (M )) is equivalent to showing the equal-
ity (D′≤0 , D′≥0 ) = (D′′≤0 , D′′≥0 ), equivalently, the inclusions D′≤0 ⊆ D′′≤0 and
D′≥0 ⊆ D′′≥0 . It suffices to prove T ⊆ D′′≤0 , ΣD≤0 ⊆ D′′≤0 , ΣF ⊆ D′′≥0 and
D≥0 ⊆ D′′≥0 . We only show the first inclusion, the other three are easy.
Let T ∈ T . To show T ∈ D′′≤0 , it suffices to show Hom(Qi , ΣT ) = 0. Applying
Hom(?, T ) to the triangle (7.1), we obtain a long exact sequence
f∗
Hom(E, T ) / Hom(ei Γ̃, ΣT ) / Hom(Qi , ΣT ) / Hom(E, ΣT ) = 0 .

We claim that f ∗ is surjective. Then the desired result follows. Consider the
commutative diagram
πi∗
Hom(ei Γ, T ) / Hom(ei Γ̃, T )
O O
(7.2) H 0 (f )∗ f∗
∗
πE
Hom(H 0 (E), T ) / Hom(E, T ),

where πi : ei Γ̃ → ei Γ and πE : E → H 0 (E) are the canonical projections. Let

C = ker(πi ). Then there is a triangle
πi
C / e Γ̃ / ei Γ / ΣC .
i

Note that C belongs to ΣD≤0 , which implies that Hom(C, T ) = 0 =

Hom(ΣC, T ). It follows that the map πi∗ is bijective. Similarly, the map πE ∗
0 ∗
is also bijective. Thus it suffices to show the surjectivity of H (f ) L . Now let
PT be a projective cover of T in mod Γ. Then PT belongs to add( j6=i ej Γ)
because T ∈ T = ⊥ Si . It follows that any morphism ei Γ → T factors through
PT , and hence factors through H 0 (f ) : ei Γ → H 0 (E), since H 0 (f ) is a mini-
mal left add( j6=i ej Γ)-approximation (for H 0 |add(Γ̃) : add(Γ̃) → add(Γ) is an
L

equivalence). This shows that H 0 (f )∗ is surjective, completing the proof of the

claim.
(c) φ23 commutes with µ+ i : Let (C
≤0
, C ≥0 ) be a bounded t-structure on
b
D (mod Λ) with length heart. Let {X1 , . . . , Xr } = φ23 (C ≤0 , C ≥0 ). The
mutated simple-minded collection µ+ i (X1 , . . . , Xr ) is contained in the heart
of the mutated t-structure µ+ i (C ≤0
, C ≥0 ). Consequently, the aisle and co-
+
aisle of φ32 ◦ µi (X1 , . . . , Xr ) are respectively contained in the aisle and co-
aisle of µ+i (C
≤0
, C ≥0 ), and hence φ32 ◦ µ+ + ≤0
i (X1 , . . . , Xr ) = µi (C , C ≥0 ), i.e.
+ ≤0 ≥0 + ≤0 ≥0 √
φ23 ◦ µi (C , C ) = µi ◦ φ23 (C , C ).

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 431

7.6. The bijections are isomorphisms of partially ordered sets. Let

Λ be a finite-dimensional algebra over K.

Theorem 7.13. The φij ’s defined in Section 5 are isomorphisms of partially

ordered sets with respect to the partial orders defined in previous subsections.

Proof. In view of Theorem 6.1 and Lemma 6.6, it suffices to show that f (x) ≥
f (y) if and only if x ≥ y for f = φ41 , φ32 and φ34 .
(a) For φ41 the desired result follows from [1, Proposition 2.14].
(b) For φ32 the desired result is included in the proof of Proposition 7.9.
′ ′
(c) Let (C≥0 , C≤0 ) and (C≥0 , C≤0 ) be two bounded co-t-structures on C and
≤0 ≥0 ′≤0 ′≥0
let (C , C ) and (C , C ) be their respective images under φ34 . Then by
definition

C ≤0 = {M ∈ Db (mod Λ) | Hom(N, M ) = 0 ∀N ∈ Σ−1 C≥0 },

C ′≤0 = {M ∈ Db (mod Λ) | Hom(N, M ) = 0 ∀N ∈ Σ−1 C≥0
′
}.

Here, C ≤0 ⊇ C ′≤0 if and only if C≥0 ⊇ C≥0

′
, and hence by definition (C ≤0 , C ≥0 ) ≥
′≤0 ′≥0 ′ ′ √
(C , C ) if and only if (C≥0 , C≤0 ) ≥ (C≥0 , C≤0 ).

8. A concrete example
Let Λ be the finite-dimensional K-algebra given by the quiver
α /
1o 2
β

with relation αβ = 0. This algebra has many manifestations: It is, possibly

up to Morita equivalence, the Auslander algebra of k[x]/x2 , the Schur algebra
S(2, 2) (charK = 2) and the principal block of the category O for sl2 (C) (K =
C). In this section we will compute the derived Picard group for Λ and classify
all silting objects/simple-minded collections in Db (mod Λ). As a consequence of
this classification and a result of Woolf [48, Theorem 3.1], the space of stability
conditions on Db (mod Λ) is exactly C2 .

8.1. Indecomposable objects. Let P1 and P2 be the indecomposable pro-

jective Λ-modules corresponding to the vertices 1 and 2. Then up to isomor-
phism and up to shift an indecomposable object in Db (mod Λ) belongs to one
of the following four families (see for example [19, 9])
· P1 (n) = P1 → P1 → . . . → P1 → P1 , n ≥ 1,
· R(n) = P1 → P1 → . . . → P1 → P1 → P2 , n ≥ 0,
· L(n) = P2 → P1 → P1 → . . . → P1 → P1 , n ≥ 0,
· B(n) = P2 → P1 → P1 → . . . → P1 → P1 → P2 , n ≥ 1,
where the homomorphisms are the unique non-isomorphisms, n is the number
of occurrences of P1 and the rightmost components have been put in degree 0.

Documenta Mathematica 19 (2014) 403–438

432 Steffen Koenig, Dong Yang

8.2. The Auslander–Reiten quiver. The Auslander–Reiten quiver of

Db (mod Λ) consists of three components: two ZA∞ components and one ZA∞
∞
component (see [10, 28])

◦
? ??
ΣS1 ◦
?  ? ??
?◦? I2 ΣP2 ?◦?
 ?
?  ? ? ? ◦ ??  ?
? ◦ ?? ? ◦ ?? ◦? ? ?◦?
◦? ◦ ◦
◦
  
◦

◦
? ? S1  ? ?? ?  ?  ?
? ? ? ? ? ? ?◦? ?◦? ?◦?
? ? P2 ? ?
◦? ◦
 −1       −1
◦ Σ P1 P1 ΣP1 ◦ ΣS2 S2 Σ S2
◦? ◦
? ?
◦

The abelian category mod Λ has five indecomposable objects up to isomor-

phism: the two simple modules S1 and S2 , their projective covers P1 and P2
and their injective envelopes I1 = P1 and I2 . They are marked on the above
Auslander–Reiten quiver.
The left ZA∞ component consists of shifts of P1 (n), n ≥ 1. The Auslander–
Reiten translation τ takes P1 (n) to Σ−1 P1 (n). It is straightforward to check
that P1 is a 0-spherical object of Db (mod Λ) in the sense of Seidel and
Thomas [45]. The additive closure of this component is the triangulated sub-
category generated by P1 . This component will be referred to as the 0-spherical
component.
The ZA∞ ∞ component consists of shifts of R(n) and L(n), n ≥ 0. Note that S1 =
L(1), P2 = R(0) = L(0) and I2 = L(2). The Auslander–Reiten translation
τ takes R(n) (n ≥ 2) to ΣR(n − 2), takes R(1) to L(1) and takes L(n) to
Σ−1 L(n + 2).
The right ZA∞ component consists of shifts of B(n), n ≥ 1. The Auslander–
Reiten translation takes B(n) to ΣB(n). The simple module S2 = B(1) is a
2-spherical object of Db (mod Λ) and the additive closure of this component is
the triangulated subcategory generated by S2 . This component will be referred
to as the 2-spherical component.

8.3. The derived Picard group. Let E be a spherical object of a triangu-

lated category C in the sense of Seidel and Thomas [45]. Then the twist functor
ΦE defined by
ev
M
ΦE (M ) = Cone( Hom(Σm E, M ) ⊗ Σm E −→ M ),
m∈Z

where ev is the evaluation map, is an auto-equivalence of C by [45, Proposition

2.10].

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 433

Recall from the preceding subsection that P1 is a 0-spherical object and S2 is

a 2-spherical object of Db (mod Λ). Thus the associated twist functors ΦP1 and
ΦS2 are two auto-equivalences of Db (mod Λ).
Lemma 8.1. For M in Db (mod Λ) there are isomorphisms ΦS2 (M ) ∼ = ΦP1 ◦
Σ−1 (M ) and Φ2P1 (M ) ∼
= ν −1
◦ Σ 2
(M ). Moreover, if M is indecomposable and
belongs to the ZA∞ component, there exists a unique pair of integers (n, n′ )
∞

such that M ∼
′
= ΦnP1 ◦ ΦnS2 (P2 ).
Proof. Observe that ΦP1 (S1 ) ∼= ΣP2 , ΦP1 (P1 ) ∼ = ΣP1 , ΦP1 (S2 ) ∼= S2 and
ΦS2 (S1 ) ∼ P
= 2 , Φ S2 (P )
1 =
∼ P1 , Φ S2 (S )
2 =
∼ Σ −1
S 2 . Since auto-equivalences
√
preserve the shape of the Auslander–Reiten quiver, the statements follow.

Remark 8.2. Inspecting the action of ΦP1 and ΦS2 on maps shows that the
= ν −1 ◦ Σ2 (M ) is functorial, while ΦS2 (M ) ∼
isomorphism Φ2P1 (M ) ∼ = ΦP1 ◦
−1
Σ (M ) is not.
Let Aut Db (mod Λ) denote the group of algebraic auto-equivalences of
Db (mod Λ), i.e. those which admits a dg lift. By [29, Lemma 6.4], such
an auto-equivalence is naturally isomorphic to the derived tensor functor of a
complex of bimodules.
Lemma 8.3. Aut Db (mod Λ) is isomorphic to Z2 × K × .
Proof. Let F ∈ Aut Db (mod Λ). Since F preserves the Auslander–Reiten
quiver, the object F (P2 ) is in the ZA∞ ∞ component. Thus there is a pair
∼ n′
of integers (nF , nF ) such that F (P2 ) = ΦnPF2 ◦ ΦSF2 (P2 ). This allows us to define
′

a map
f: Aut Db (mod Λ) / Z2

F
 / (nF , n′ ).
F

This map is clearly a surjective group homomorphism. Moreover, the group

homomorphism
Z2 / Aut Db (mod Λ)

(n, n′ )
 / Φn ◦ Φn′
P2 S2

is a retraction of f . Therefore Aut Db (mod Λ) ∼= Z2 × ker(f ).

∼
Let F ∈ ker(f ). Then F (P2 ) = P2 . This forces F (P1 ) ∼ = P1 , and hence F
is induced from an outer automorphism of Λ which fixes the two primitive √
idempotents e1 and e2 . Thus ker(f ) ∼= K × , finishing the proof.

8.4. Morphism spaces. We first compute the morphism spaces between the
two ZA∞ components.
Lemma 8.4. (a) For n ≥ 2, Hom(P1 (n), Σm P1 (n)) does not vanish for
some m > 0 and for some m < 0. For n = 1, Hom(P1 , Σm P1 ) is
isomorphic to K[x]/x2 for m = 0 and vanishes for m 6= 0.

Documenta Mathematica 19 (2014) 403–438

434 Steffen Koenig, Dong Yang

(b) For n ≥ 2, Hom(B(n), Σm B(n)) does not vanish for some m > 0 and
for some m < 0. For n = 1, Hom(S2 , Σm S2 ) is K for m = 0, 2 and
vanishes for m 6= 0, 2.
Proof. Direct computation, or apply some general result (e.g. [25, Section 2])
√
to the triangulated categories generated by P1 and S2 .
Next we compute the morphism spaces between P2 and the objects on the ZA∞
∞
component.
Lemma 8.5. Let n ≥ 0.
(a) Hom(P2 , Σm R(n)) is K if −n ≤ m ≤ 0 and is 0 otherwise.
(a’) Hom(R(n), Σm P2 ) is K if 2 ≤ m ≤ n or if n = 0, m = 0 and is 0
otherwise.
(b) Hom(P2 , Σm L(n)) is K if 2 − n ≤ m ≤ 0 or if n = 0, m = 0 and is 0
otherwise.
(b’) Hom(L(n), Σm P2 ) is K if 0 ≤ m ≤ n and is 0 otherwise.
Proof. (a) and (b) Because Hom(P2 , M ) = H 0 (M )e2 .
(a’) and (b’) are obtained from (a) and (b) by applying the Auslander–Reiten
√
formula D Hom(M, N ) ∼ = Hom(N, τ ΣM ).
8.5. Silting objects and simple-minded collections. Now we are ready
to classify the silting objects and simple-minded collections in Db (mod Λ).
Proposition 8.6. Up to isomorphism, any basic silting object of Db (mod Λ)
belongs to one of the following two families
′
· ΦnP1 ◦ ΦnS2 (P1 ⊕ P2 ), n, n′ ∈ Z, the corresponding simple-minded collec-
′
tion is ΦnP1 ◦ ΦnS2 {S1 , S2 },
′
· ΦnP1 ◦ ΦnS2 (Σm S1 ⊕ P2 ), n, n′ ∈ Z and m ≤ −1, the corresponding
′
simple-minded collection is ΦnP1 ◦ ΦnS2 {Σm S1 , I2 }.
Proof. Let N be an indecomposable direct summand of a silting object. By
Lemma 8.4, N does not belong to the 2-spherical component, and N belongs
to the 0-spherical component if and only if N is a shift of P1 . Moreover, a basic
silting object can have at most one shift of P1 as a direct summand. It follows
that a silting object has at least one indecomposable direct summand from the
ZA∞ ∞ component.
Let M = M1 ⊕M2 be a silting object with M1 and M2 indecomposable. Assume
that M1 belongs to the ZA∞ ∞ component. Up to an auto-equivalence of the
′
form ΦnP1 ◦ ΦnS2 , we may assume that M1 = P2 . Then, if M2 belongs to
the 0-spherical component it has to be P1 . Thus we assume that M2 also
belongs to the ZA∞ ∞ component. Then it follows from Lemma 8.5 that M2 is
isomorphic to Σm S1 for some m ≤ −1 or to Σm R(1) for some m ≥ 0. Observing
−m−1
P2 ⊕ Σm R(1) = ΦP 1
◦ ΦmS2 (P2 ⊕ Σ
−m−1
S1 ) for m ≥ 0 finishes the proof for
the silting-object part.
That the simple-minded collection corresponding to a silting object is the de- √
sired one follows from the Hom-duality they satisfy.

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 435

8.6. The silting quiver. Recall from [1] that the silting quiver has as vertices
the isomorphism classes of basic silting objects and there is an arrow from M
to M ′ if M ′ can be obtained from M by a left mutation.
The vertex set of the silting quiver of Db (mod Λ) is {(n, n′ , m) | n, n′ ∈ Z, m ∈
′
Z≤0 }, where (n, n′ , 0) represents the silting object ΦnP1 ◦ ΦnS2 (P1 ⊕ P2 ) and
′
(n, n′ , m) (m ≤ −1) represents the silting object ΦnP1 ◦ ΦnS2 (Σm S1 ⊕ P2 ). It is
straightforward to show that from each vertex (n, n′ , m) there are precisely two
outgoing arrows whose targets are respectively
· (n, n′ − 1, m) and (n + 1, n′ , m − 1) if m = 0,
· (n + 1, n′ − 1, m − 1) and (n, n′ , m + 1) if m ≤ −1.
8.7. Hearts and the space of stability conditions.
Lemma 8.7. The heart of any t-structure on Db (mod Λ) is a length category.
Proof. Let A be the heart of a t-structure on Db (mod Λ). We will show that
A has only finitely many isomorphism classes of indecomposable objects. Such
an abelian category must be a length category.
Due to vanishing of negative extensions, it follows from Lemma 8.4 that A
contains at most one indecomposable object from the 0-spherical component
respectively the 2-spherical component.
Suppose that A contains an indecomposable object from the ZA∞ ∞ compo-
nent. Without loss of generality we may assume that it is P2 . It follows from
′
Lemma 8.5 that for n ≥ 3 and m ∈ Z either Hom(P2 , Σm Σm R(n)) 6= 0 for
′
some m′ < 0 or Hom(Σm R(n), Σm P2 ) 6= 0 for some m′ < 0. Similarly for
L(n). Therefore an indecomposable object M belongs to the heart only if it is
isomorphic to one of Σm P2 , Σm R(1), Σm R(2), Σm L(1) and Σm L(2), m ∈ Z.
But at most one shift of a nonzero object can belong to a heart. So A contains
√
at most 7 indecomposable objects up to isomorphism.
In view of Lemma 8.7, the result in the preceding subsection shows that all
bounded t-structures on Db (mod Λ) are related to each other by a sequence of
left or/and right mutations. In particular, this implies that the t-structures
Woolf considered in [48, Section 3.1] are already all bounded t-structures on
Db (mod Λ). Therefore we have
Corollary 8.8. (a) The Bridgeland space of stability conditions on
Db (mod Λ) is C2 .
(b) An abelian category is the heart of some bounded t-structure on
Db (mod Λ) if and only if it is equivalent to mod Γ for Γ = Λ or
Γ = K( · / · ) or Γ = K ⊕ K.

References
[1] Takuma Aihara and Osamu Iyama, Silting mutation in triangulated cate-
gories, J. Lond. Math. Soc. (2) 85 (2012), no. 3, 633–668.
[2] Salah Al-Nofayee, Equivalences of derived categories for selfinjective alge-
bras, J. Algebra 313 (2007), no. 2, 897–904.

Documenta Mathematica 19 (2014) 403–438

436 Steffen Koenig, Dong Yang

[3] , Simple objects in the heart of a t-structure, J. Pure Appl. Algebra

213 (2009), no. 1, 54–59.
[4] Claire Amiot, Cluster categories for algebras of global dimension 2 and
quivers with potential, Ann. Inst. Fourier (Grenoble) 59 (2009), no. 6, 2525–
2590.
[5] Ibrahim Assem, Marı́a José Souto Salorio, and Sonia Trepode, Ext-
projectives in suspended subcategories, J. Pure Appl. Algebra 212 (2008),
no. 2, 423–434.
[6] Maurice Auslander, Marı́a Inés Platzeck, and Idun Reiten, Coxeter functors
without diagrams, Trans. Amer. Math. Soc. 250 (1979), 1–46.
[7] Aslak Bakke Buan, Robert J. Marsh, Markus Reineke, Idun Reiten, and
Gordana Todorov, Tilting theory and cluster combinatorics, Adv. Math.
204 (2) (2006), 572–618.
[8] Alexander A. Beilinson, Joseph Bernstein, and Pierre Deligne, Analyse et
topologie sur les espaces singuliers, Astérisque, vol. 100, Soc. Math. France,
1982 (French).
[9] Viktor Bekkert and Héctor A. Merklen, Indecomposables in derived cate-
gories of gentle algebras, Algebr. Represent. Theory 6 (2003), no. 3, 285–
302.
[10] Grzegorz Bobiński, Christof Geiß, and Andrzej Skowroński, Classification
of discrete derived categories, Cent. Eur. J. Math. 2 (2004), no. 1, 19–49
(electronic).
[11] Mikhail V. Bondarko, Motivically functorial coniveau spectral sequences;
direct summands of cohomology of function fields, Doc. Math. (2010),
no. Extra volume: Andrei A. Suslin sixtieth birthday, 33–117.
[12] , Weight structures vs. t-structures; weight filtrations, spectral se-
quences, and complexes (for motives and in general), J. K-Theory 6 (2010),
no. 3, 387–504.
[13] Sheila Brenner and Michael C. R. Butler, Generalizations of the Bernstein-
Gel′ fand-Ponomarev reflection functors, Representation theory, II (Proc.
Second Internat. Conf., Carleton Univ., Ottawa, Ont., 1979), Lecture Notes
in Math., vol. 832, Springer, Berlin, 1980, pp. 103–169.
[14] Tom Bridgeland, t-structures on some local Calabi-Yau varieties, J. Alge-
bra 289 (2005), no. 2, 453–483.
[15] , Stability conditions on triangulated categories, Ann. of Math. (2)
166 (2007), no. 2, 317–345.
[16] Thomas Brüstle and Dong Yang, Ordered exchange graphs, Proceedings of
the ICRA XV (Bielefeld), to appear, arXiv:1302.6045.
[17] Aslak Bakke Buan, Idun Reiten, and Hugh Thomas, From m-clusters to
m-noncrossing partitions via exceptional sequences, Math. Z. 271 (2012),
1117–1139.
[18] , Three kinds of mutation, J. Algebra 339 (2011), 97–113.
[19] Igor Burban and Yuriy Drozd, Derived categories for nodal rings and pro-
jective configurations, Noncommutative algebra and geometry, Lect. Notes

Documenta Mathematica 19 (2014) 403–438

 Silting Objects, Simple-Minded Collections, . . . 437

Pure Appl. Math., vol. 243, Chapman & Hall/CRC, Boca Raton, FL, 2006,
pp. 23–46.
[20] Sergey Fomin and Andrei Zelevinsky, Cluster algebras. I. Foundations, J.
Amer. Math. Soc. 15 (2002), no. 2, 497–529 (electronic).
[21] , Cluster algebras IV: Coefficients, Compos. Math. 143 (2007), 112–
164.
[22] Changjian Fu, Aisles, recollements and dg categories, Master Thesis,
Sichuan University, 2006 (Chinese).
[23] Dieter Happel, Triangulated categories in the representation theory of
finite-dimensional algebras, London Mathematical Society Lecture Note
Series, vol. 119, Cambridge University Press, Cambridge, 1988.
[24] Dieter Happel, Idun Reiten, and Sverre O. Smalø, Tilting in abelian cate-
gories and quasitilted algebras, Mem. Amer. Math. Soc. 120 (1996), no. 575,
viii+ 88.
[25] Thorsten Holm, Peter Jørgensen, and Dong Yang, Sparseness of t-
structures and negative Calabi-Yau dimension in triangulated categories
generated by a spherical object, Bull. Lond. Math. Soc. 45 (2013), 120–130.
[26] Mitsuo Hoshino, Yoshiaki Kato, and Jun-Ichi Miyachi, On t-structures
and torsion theories induced by compact objects, J. Pure Appl. Algebra
167 (2002), no. 1, 15–35.
[27] Peter Jørgensen and David Pauksztello, The co-stability manifold of a tri-
angulated category, arXiv:1109.4006.
[28] Martin Kalck and Dong Yang, Derived categories of graded gentle one-cycle
algebras, preprint (2013).
[29] Bernhard Keller, Deriving DG categories, Ann. Sci. École Norm. Sup. (4)
27 (1994), no. 1, 63–102.
[30] , On differential graded categories, International Congress of Math-
ematicians. Vol. II, Eur. Math. Soc., Zürich, 2006, pp. 151–190.
[31] Bernhard Keller and Pedro Nicolás, Cluster hearts and cluster tilting ob-
jects, in preparation.
[32] , Weight structures and simple dg modules for positive dg algebras,
Int Math Res Notices 2013 (2013), 1028–1078.
[33] Bernhard Keller and Dieter Vossieck, Aisles in derived categories, Bull.
Soc. Math. Belg. Sér. A 40 (1988), no. 2, 239–253.
[34] Bernhard Keller and Dong Yang, Derived equivalences from mutations of
quivers with potential, Adv. Math. 226 (2011), no. 3, 2118–2168.
[35] Alastair King and Yu Qiu, Oriented exchange graphs of acyclic Calabi-Yau
categories, arXiv:1109.2924.
[36] Steffen Koenig and Yuming Liu, Gluing of idempotents, radical embeddings
and two classes of stable equivalences, J. Algebra 319 (2008), no. 12, 5144–
5164.
[37] Steffen Koenig and Dong Yang, On tilting complexes providing de-
rived equivalences that send simple-minded objects to simple objects,
arXiv:1011.3938.

Documenta Mathematica 19 (2014) 403–438

438 Steffen Koenig, Dong Yang

[38] Maxim Kontsevich and Yan Soibelman, Stability structures, mo-

tivic Donaldson-Thomas invariants and cluster transformations,
arXiv:0811.2435.
[39] Octavio Mendoza, Edith C. Sáenz, Valente Santiago, and Marı́a José Souto
Salorio, Auslander-Buchweitz context and co-t-structures, Appl. Categor.
Struct. 21 (2013), 417–440.
[40] Tomoki Nakanishi and Andrei Zelevinsky, On tropical dualities in cluster
algebras, Algebraic groups and quantum groups, Contemp. Math., vol. 565,
Amer. Math. Soc., Providence, RI, 2012, pp. 217–226.
[41] David Pauksztello, Compact corigid objects in triangulated categories and
co-t-structures, Cent. Eur. J. Math. 6 (2008), no. 1, 25–42.
[42] Pierre-Guy Plamondon, Cluster characters for cluster categories with
infinite-dimensional morphism spaces, Adv. Math. 227 (2011), no. 1, 1–
39.
[43] Jeremy Rickard, Equivalences of derived categories for symmetric algebras,
J. Algebra 257 (2002), no. 2, 460–481.
[44] Jeremy Rickard and Raphael Rouquier, Stable categories and reconstruc-
tion, arXiv:1008.1976.
[45] Paul Seidel and Richard Thomas, Braid group actions on derived categories
of coherent sheaves, Duke Math. J. 108 (2001), no. 1, 37–108.
[46] Jorge Vitória, Silting objects on derived module categories, in preparation.
[47] Jiaqun Wei, Semi-tilting complexes, Israel Journal of Mathematics 194
(2013), 871–893.
[48] Jonathan Woolf, Stability conditions, torsion theories and tilting, J. Lond.
Math. Soc. (2) 82 (2010), no. 3, 663–682.

Steffen Koenig Dong Yang

Universität Stuttgart Department of Mathematics
Institut für Algebra Nanjing University
und Zahlentheorie Nanjing 210093
Pfaffenwaldring 57 P. R. China
D-70569 Stuttgart [email protected]
Germany
[email protected]
stuttgart.de

Documenta Mathematica 19 (2014) 403–438