Quasilifting of hulls and depth of tensor product of modules
Sutapa Dey, Amit Tripathi · 2024
Read Quasilifting of hulls and depth of tensor product of modules by Sutapa Dey, Amit Tripathi (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-01392.txt.
Introduction
Sutapa Dey and Amit Tripathi study depth in combined algebraic structures in this two thousand twenty-four paper. Their objects are modules over local rings, the algebraist's microscope for local behavior. Depth measures how many clean dividing steps a module tolerates. Tensor products merge two modules, and their depth is hard to predict. The authors use a lifting construction from earlier work by Huneke and others.
They recover a known result for complete intersection rings by their route. One question from that earlier work gets a negative answer, with a matching lower bound. Results extend to a special class of rings named in the paper. Commutative algebra specialists are the audience. The paper is pure theory, aimed at sharpening long-standing conjectures.
Catalog record
| CREATOR | Sutapa Dey, Amit Tripathi |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 75 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-01392
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Quasilifting of hulls and depth of tensor product of modules Sutapa Dey, Amit Tripathi Abstract . We use a quasilift type construction to obtain some bounds on the depth of the tensor product of certain modules over a local 𝒯 ℰ \mathcal{TE} ring. We recover a result of Celikbas, Sadeghi, and Takahashi for local complete intersection rings. Some general conditions for lifting hulls and approximations are also studied. 2010 Mathematics Subject Classification 13D07, 13C14, 13C15 The aim of this paper is to study the connection between bounds on the depth of tensor products of modules and the vanishing of Tor modules. This area traces its origin to the seminal work of Auslander [ 2 ] , who established that for an unramified regular local ring R , if M\otimes_{R}N is torsion-free for some non-zero finitely generated modules M and N, then M and N are torsion-free, Tor_{i}^{R}(M,N)=0 for all i\geq 1, and further, the depth formula holds i.e. depth(M\otimes_{R}N)=depth(M)+depth(N)-depth(R). Later, Lichtenbaum [ 27 ] generalized this result to the case of all regular local rings. These results were further extended to the case of abstract hypersurface rings by Huneke and Wiegand [ 22 , 2.7] (see also [ 23 ] ). They proved that if M\otimes_{R}N is reflexive, and either of these modules has a constant rank, then Tor_{i}^{R}(M,N)=0 for i\geq 1 and at least one of the M and N is reflexive (see [ 14 , 1.3] for a proof of the last claim). In particular, the depth formula holds. These investigations were continued by Huneke, Jorgensen, and Wiegand [ 20 ] , who used quasilifts to prove certain Tor-vanishing results for codimension 2 and 3 cases under additional assumptions. Without any assumption on the modules or the ring, even if M\otimes_{R}N is totally reflexive, it is not enough to force Tor-independence; see [ 12 , 3.14] for a counterexample over a codimension 2 local complete intersection. In the same paper, Celikbas obtains some Tor vanishing criteria [ 12 , 1.1, 1.2] . Recall that a local complete intersection ring is a local ring R whose m -adic completion is of the form Q/(\underline{x}) where Q is a regular local ring and \underline{x} is a regular sequence. Celikbas, Sadeghi, and Takahashi proved the following Theorem 1 . [ 15 , 1.2] Let R be a local complete intersection ring and let M and N be nonzero finitely generated R -modules satisfying: (a) \lambda(Tor_{i}^{R}(M,N))<\infty for all i\geq 1 (where \lambda denotes the length), and (b) depth(M\otimes_{R}N)\geq cx_{R}(M,N)+1. Then depth(M\otimes_{R}N)\geq depth(M)+depth(N)-depth(R) and the equality holds if and only if the pair (M,N) are Tor-independent. Let {\mathcal{C}}{\mathcal{I}} denote the class of local complete intersection rings and \mathcal{GOR} denote the class of Gorenstein local rings. Following Avramov and Buchweitz [ 7 , 6.3] , we say that a ring R satisfies the property te if for any pair (M,N) of finite R -modules Tor_{i}^{R}(M,N)=0 for i\gg 0 implies Ext^{i}_{R}(M,N)=0 for i\gg 0. Let {\mathcal{T}}{\mathcal{E}} represents the class of local rings which satisfy the property {\bf te}. It follows by [ 7 , III] that we have an inclusion \mathcal{CI}\subseteq\mathcal{TE}\subseteq\mathcal{GOR}. Subsequent works by Huneke and Jorgensen [ 21 ] , and Jorgensen and Sega [ 25 ] show that both the inclusions are strict. More recently, Kimura, Lyle, Otake and Takahashi [ 26 ] have studied some properties of \mathcal{TE} rings and their related homological cousins. Our main result is, Theorem 2 . Let R be a local ring, and let M and N be non-zero finitely generated R -modules with CI\mbox{-}dim_{R}M<\infty . Assume that (a) depth(M\otimes_{R}N)\geq cx_{R}(M,N)+1. (b) Assume that one of the following holds (i) R\in\mathcal{TE} and M is locally free over the punctured spectrum of R. (ii) R\in\mathcal{CI} and \lambda(Tor_{i}^{R}(M,N))<\infty for i\geq 1. Then depth(M\otimes_{R}N)\geq depth(M)+depth(N)-depth(R). Moreover, the equality holds (i.e. the depth formula holds) if and only if the pair (M,N) is Tor-independent. The part (b) (b)(ii) of Theorem 2 recovers Theorem 1 cited above, while part (b) (b)(i) is new. The proof of Theorem 2 is based on a quasilift type construction of Huneke, Jorgensen, and Wiegand [ 20 ] , which attaches to a projective hull of a module M on R, a {\mathcal{G}} -approximation of M on Q where Q is a deformation of R (see Proposition 6 and Lemma 10 ). This allows one to ”lift” the question to a pair of Q -modules. We then apply [ 7 , III] and a result of Celikbas and Dao [ 13 , 2.4] to obtain complexity reduction for this pair. One of the motivations behind this work was to answer the [ 15 , Question 3.10] which asked if one can replace the hypothesis ” \lambda(Tor_{i}^{R}(M,N))<\infty for all i\geq 1 ” with ” \lambda(Tor_{i}^{R}(M,N))<\infty for all i\gg 0 ” ? We show in Example 3.2 that this is not true. However, with this hypothesis, we establish the following lower bound on the depth of tensor product: Theorem 3 . Let R be a local complete intersection ring and let M and N be nonzero finitely generated R -modules. Assume that (a) \lambda(Tor_{i}^{R}(M,N))<\infty for i\gg 0. (b) depth(M\otimes_{R}N)\geq cx_{R}(M,N)+1. Then depth(M\otimes_{R}N)\geq syz(M)+syz(N)-depth(R). Here by syz(X), we mean the largest integer k\leq\dim\ R such that X is a k ’th syzygy module (see section 1.2 ). Setting a bound on the depth of the tensor product, using syz(\cdot) instead of depth, comes at a cost, see inequality ( 1.4 ). The proof of Theorem 3 uses a Tor rigidity criteria of Celikbas, Sadeghi, and Takahashi [ 15 , 3.4] . The hypothesis (a) on the length of Tor in Theorem 3 is satisfied, for example, when R has an isolated singularity or if either of the modules has finite projective dimension over the punctured spectrum of R. We now describe the organization of this paper. Section 1 contains notations and some preliminary results. In section 2 , we discuss a few examples concerning \perp and \top operations (defined in section 1 ) and give construction for lifting of hulls and approximations. Section 3 contains proofs for the main results. 1. Preliminaries Throughout, all rings are commutative Noetherian rings and modules are assumed to be finitely generated. For any R -module X, we denote Hom_{R}(X,R) by X^{*}. 1.1. Approximations and hulls See [ 31 ] or [ 18 ] . Let {\mathcal{D}} and {\mathcal{E}} be any non-empty subsets of mod(R). We say that {\mathcal{D}} is closed under extension if Y\in{\mathcal{D}} whenever there is a short exact sequence of R -modules 0\rightarrow Y^{\prime}\rightarrow Y\rightarrow Y^{\prime\prime}\rightarrow 0 with Y^{\prime},Y^{\prime\prime}\in{\mathcal{D}}. Similarly, we say that {\mathcal{D}} is closed under syzygies if for every module X\in{\mathcal{D}}, every first syzygy \Omega^{1}_{R} of X is in {\mathcal{D}}. Set \displaystyle{\mathcal{D}}^{\perp}= \displaystyle\{Y\in mod(R)\ |\ Ext^{i}_{R}(X,Y)=0\text{ for all }X\in{\mathcal{D}}\text{ and all }i>0\} (1.1) \displaystyle{\mathcal{E}}^{\top}= \displaystyle\{X\in mod(R)\ |\ Ext^{i}_{R}(X,Y)=0\text{ for all }Y\in{\mathcal{E}}\text{ and all }i>0\}. (1.2) We say that a module M\in mod(R), admits a {\mathcal{D}} -approximation if there are modules X\in{\mathcal{D}} and Y\in{\mathcal{D}}^{\perp} satisfying a short exact sequence 0\rightarrow Y\rightarrow X\rightarrow M\rightarrow 0. We say that {\mathcal{E}} admits {\mathcal{D}} -approximation if every module M\in{\mathcal{E}} admits a {\mathcal{D}} -approximation. The definitions 1.1 and {\mathcal{D}} -approximation of a module are from [ 18 , 2.1] and are motivated by the notion