On Τ-Tilting Finiteness of the Schur Algebra
Total Page:16
File Type:pdf, Size:1020Kb
On τ-tilting finiteness of the Schur algebra Qi Wang∗ Abstract We determined the τ-tilting finiteness of Schur algebras over an algebraically closed field of arbitrary characteristic, except for a few small cases. Keywords: Schur algebras, τ-tilting finite, representation-finiteness. 1 Introduction Throughout this paper, we always assume that A is a finite-dimensional basic algebra over an algebraically closed field F of characteristic p> 0. In particular, the representa- tion type of A is divided into representation-finite, (infinite-)tame and wild. In recent years, τ-tilting theory introduced by Adachi, Iyama and Reiten [AIR] has drawn more and more attention. Here, τ is the Auslander-Reiten translation. The crucial concept of τ-tilting theory is the notion of support τ-tilting modules (see Definition 2.1), which can be regarded as a generalization of classical tilting modules from the viewpoint of mutation, that is, any basic almost complete support τ-tilting module is a direct summand of exactly two basic support τ-tilting modules. Moreover, support τ-tilting modules are in bijection with many important objects in representation theory, such as two-term silting complexes, functorially finite torsion classes, left finite semibricks, etc. We refer to [AIR] and [Asa] for more details. Similar to the representation finiteness, a modern notion named τ-tilting finiteness is introduced by Demonet, Iyama and Jasso [DIJ]. We recall that an algebra A is said to be τ-tilting finite if it has only finitely many pairwise non-isomorphic basic support τ-tilting modules. (There are several equivalent conditions in Definition 2.2 and Proposition 2.3.) A typical example of τ-tilting finite algebras is the class of representation-finite algebras. arXiv:2010.05206v2 [math.RT] 9 May 2021 We remark that the τ-tilting finiteness for several classes of algebras is determined, such as algebras with radical square zero [Ada], preprojective algebras of Dynkin type [Mi], Brauer graph algebras [AAC], biserial algebras [Mo] and minimal wild two-point algebras [W1]. In addition, it has been proved in some cases that τ-tilting finiteness coincides with representation-finiteness, including cycle-finite algebras [MS], gentle algebras [Pl], cluster-tilted algebras [Zi], simply connected algebras [W2], quasi-tilted algebras, locally hereditary algebras, etc. [AHMW]. In this paper, we focus on (classical) Schur algebras, which play an important role in both Representation theory and Lie theory. This class of algebras arose in connection with ∗Q.W. is supported by the JSPS Grant-in-Aid for JSPS Fellows 20J10492. 1 the theory of polynomial representations over the general linear group GLn(F) and then received widespread attention. So far, the representation theory of Schur algebras has developed very well and many derivatives appeared, such as q-Schur algebras, infinitesimal Schur algebras, Borel-Schur algebras and so on. In particular, the representation type of Schur algebras is completely determined by various authors, including Erdmann [Er], Xi [Xi], Doty-Nakano [DN] and Doty-Erdmann-Martin-Nakano [DEMN]. We summarize these results in Proposition 3.1. Let n, r be two positive integers and V an n-dimensional vector space over F. We ⊗r denote by V the r-fold tensor product V ⊗F V ⊗F ···⊗F V . Then, the symmetric group ⊗r Gr has a natural action (by permutation) on V which makes it to be a module over ⊗r the group algebra FGr of Gr. Then, the endomorphism ring EndFGr (V ) is called the Schur algebra (see [Ma, Section 2]) and we denote it by S(n, r). Our initial motivation is to check whether τ-tilting finiteness and representation- finiteness coincide in the class of Schur algebras. It is clear that representation-finite Schur algebras are τ-tilting finite. Moreover, we have already determined the number of pairwise non-isomorphic basic support τ-tilting modules blockwise in Theorem 3.2. Therefore, we start with the tame cases. Theorem 1.1 (Theorem 3.4). If the Schur algebra S(n, r) is tame, it is τ-tilting finite. In order to prove the above result, we may show that any block algebra of tame Schur algebras is τ-tilting finite. Let S(n, r) be a tame Schur algebra and A := eS(n, r)e an indecomposable block algebra with a central idempotent e of S(n, r). It is known from [Er] and [DEMN, Section 5] that the block algebra A is Morita equivalent to one of F, A2, D3, D4, R4 and H4 (see Section 3 for the definitions). Then, we have Theorem 1.2 (Theorem 3.2 and Lemma 3.3). Let sτ-tilt A be the set of pairwise non- isomorphic basic support τ-tilting A-modules. Then, A A2 D3 D4 R4 H4 . #sτ-tilt A 6 28 114 88 96 Since the above results give a negative answer to our initial motivation, we next ask whether all wild Schur algebras are τ-tilting infinite. There are a few cases in (⋆) that we currently do not have a suitable method to deal with. Since the number of pairwise non-isomorphic basic support τ-tilting modules for them is huge, we cannot confirm the τ-tilting finiteness of these exceptions by direct calculation. So, we leave these cases here and expect to find a new reduction method (for example, Conjecture 4.13) to show the τ-tilting finiteness. p =2, n =2,r =8, 17, 19; p =2, n =3,r = 4; (⋆) p =2, n > 5,r = 5; p > 5, n =2,p2 6 r 6 p2 + p − 1. 2 Now, we are able to determine the τ-tilting finiteness for most wild Schur algebras. Theorem 1.3 (Table 1, Table 2 and Table 3). Let p be the characteristic of F. Except for the cases in (⋆), the wild Schur algebra S(n, r) is τ-tilting finite if and only if p =2, n =2,r =6, 13, 15 or n =3,r =5 or n =4,r =4. We have the following answers for our motivations. Corollary 1.4. Let p =3. Then, S(n, r) is τ-tilting infinite if and only if it is wild. Corollary 1.5. Let p > 5. Except for the cases in (⋆), S(n, r) is τ-tilting finite if and only if it is representation-finite. Actually, except for the cases in (⋆), we have observed from the proof process that the Schur algebra S(n, r) is τ-tilting infinite if and only if it has one of τ-tilting infinite quivers Q1, Q2 and Q3 (see Lemma 2.15) as a subquiver. This leads us to conjecture that all cases in (⋆) are τ-tilting finite, because all the cases in (⋆) do not have one of Q1, Q2 and Q3 as a subquiver. This paper is organized as follows. In Section 2, we first recall some basic materials on τ-tilting theory and the Schur algebra S(n, r). Then, we give several reduction theorems on the τ-tilting finiteness of Schur algebras, such that we only need to consider small n and r. At the end of the section, we explain our strategy to prove S(n, r) to be τ-tilting infinite. In Section 3, we show that all tame Schur algebras are τ-tilting finite. In Section 4, we determine the τ-tilting finiteness of most wild Schur algebras. Acknowledgements. This paper is inspired by another on-going joint work with Susumu Ariki, Ryoichi Kase, Kengo Miyamoto and Euiyong Park, where the authors are using τ-tilting theory to classify τ-tilting finite block algebras of Hecke algebras of classical type. I am very grateful to them, especially, Prof. Ariki, for giving me a lot of kind guidance in writing this paper. I am also very grateful to the anonymous referee for giving me many useful suggestions. 2 Preliminaries We recall that any finite-dimensional basic algebra A over an algebraically closed field F is isomorphic to a bound quiver algebra FQ/I, where FQ is the path algebra of the Gabriel quiver Q = QA of A and I is an admissible ideal of FQ. We refer to [ASS] for more background materials on quiver and representation theory of algebras. 2.1 τ-tilting theory In this subsection, we review some background of τ-tilting theory. However, we only review the background needed for this paper, so that many aspects related to τ-tilting theory are omitted. One may look at [AIR], [Asa] and [AIRRT] for more materials. 3 Let mod A be the category of finitely generated right A-modules and proj A the category of finitely generated projective right A-modules. For any M ∈ mod A, we denote by add(M) (respectively, Fac(M)) the full subcategory of mod A whose objects are direct summands (respectively, factor modules) of finite direct sums of copies of M. We denote by Aop the opposite algebra of A and by |M| the number of isomorphism classes of indecomposable direct summands of M. In particular, we often describe A- modules via their composition factors. For example, each simple A-module Si is written S1 1 as i, and then S2 ≃ 2 is an indecomposable A-module M with a unique simple submodule S2 such that M/S2 = S1. We recall that the Nakayama functor ν := D(−)∗ is induced by the dualities op ∗ op D := HomF(−, F): mod A ↔ mod A and (−) := HomA(−, A): proj A ↔ proj A . Then, for any M ∈ mod A with a minimal projective presentation f1 f0 P1 −→ P0 −→ M −→ 0, the Auslander-Reiten translation τM is defined by the exact sequence: νf1 0 −→ τM −→ νP1 −→ νP0. Definition 2.1.