Gorenstein Projective Precovers

• Gorenstein projective precovers are special morphisms from Gorenstein projective modules that provide a homological approximation via totally acyclic resolutions.
• They are constructed through cotorsion pair theory and reduction criteria that ensure finite Gorenstein projective dimensions in module categories.
• Their applications span Gorenstein rings, commutative noetherian rings, and non-coherent GF-closed rings, underlining the depth of modern homological methods.

A Gorenstein projective precover of a module provides a powerful homological approximation theory underlying relative homological algebra over associative rings. The existence, structure, and properties of such precovers are central to modern approaches in Gorenstein homological algebra, module theory, and representation theory. The systematic study of Gorenstein projective precovers merges classical projective approximation, cotorsion pairs, relative dimension theory, and modern methods from triangulated and model categories, with deep implications for the structure of modules, derived categories, and invariants of rings.

1. Definitions and Framework

Let $R$ be an associative ring with unit, and $R$–Mod the category of (left) $R$‐modules. A complex $P = \cdots \to P_1 \to P_0 \to P_{-1} \to \cdots$ of projective $R$‐modules is called totally acyclic if it is exact and, for every projective $Q$, the complex $\operatorname{Hom}_R(P, Q)$ is also exact. An $R$‐module $G$ is Gorenstein projective if there exists a totally acyclic complex of projectives $P$ such that $G \cong \ker(P_0 \to P_{-1})$ (Estrada et al., 2023).

Given a class $D$ of modules, a $D$-precover of $M$ is a morphism $\varphi: D \to M$ with $D \in D$ such that any map from any $D' \in D$ to $M$ factors through $\varphi$. If additionally $\ker \varphi \in D^{\perp_1}$, where $$D^{\perp_1} = \{ N \mid \operatorname{Ext}^1_R(D', N) = 0, \forall D' \in D \}$$, then $\varphi$ is a special $D$-precover. When $D$ is the class of Gorenstein projective modules (denoted $\operatorname{GP}$), the resulting map is a Gorenstein projective precover, or special Gorenstein projective precover if the kernel condition holds (Estrada et al., 2023, Estrada et al., 2016).

The Gorenstein projective dimension of a module $M$ is the infimum over all $n \geq 0$ such that there exists an exact sequence

$$0 \to G_n \to G_{n-1} \to \cdots \to G_0 \to M \to 0$$

with each $G_i$ Gorenstein projective; if no such $n$ exists, $Gpd_R(M)=\infty$.

2. Existence Criteria: Reduction and Finiteness Principles

The existence problem for Gorenstein projective precovers is subtle and depends on ring-theoretic finiteness conditions. If $R$ has finite Gorenstein global dimension, i.e., $\sup\{Gpd_R(M) : M \in R\text{–Mod}\}<\infty$, every module admits a special Gorenstein projective precover (Estrada et al., 2023). However, in the absence of global dimension finiteness, a key result (Estrada–Iacob reduction theorem) provides an accessible criterion:

• If there exists $n \geq 0$ such that every finitely presented $R$‐module has $Gpd_R(T) \leq n$, then $\operatorname{GP}$ is special precovering in $R$–Mod (Estrada et al., 2023).

This criterion reduces the global existence problem to a finiteness check on finitely presented modules and their syzygies. The construction passes through the hereditary cotorsion pair $(C,SFPI)$ where $SFPI$ is the class of strongly FP-injective modules (modules $M$ with $\operatorname{Ext}^i_R(T, M)=0$ for every finitely presented $T$ and $i\geq 1$) (Estrada et al., 2023, Becerril, 31 Jan 2026). Every module in $C$ is shown to have finite Gorenstein projective dimension by analyzing transfinite extensions and direct summands.

Under the additional property that every Gorenstein projective is Gorenstein flat and every Gorenstein flat has finite Gorenstein projective dimension, special Gorenstein projective precovers exist over any left GF-closed ring—a class encompassing Gorenstein rings, two-sided noetherian rings of finite Krull dimension, and right coherent left $n$–perfect rings (Estrada et al., 2016).

3. Construction Methods, Cotorsion Pairs, and Homological Approximations

The existence theorems for special precovers are established via cotorsion pair theory. Given a cotorsion pair $(\mathcal{A}, \mathcal{B})$ in an abelian category (with enough projectives), one produces for any module $M$ a short exact sequence

$0 \to B \to A \to M \to 0$

with $A \in \mathcal{A}$ and $B \in \mathcal{B}$, and $\mathcal{A}$-precovers correspond to such $A \to M$. Completeness and heredity of the cotorsion pair are crucial, and these are established under the finiteness and closure conditions described above (Estrada et al., 2016, Estrada et al., 2023, Becerril, 31 Jan 2026).

A representative proof proceeds as follows:

• First, construct resolutions for modules with finite Gorenstein projective dimension using projectives, then splice these with totally acyclic complexes to build special Gorenstein projective precovers for such modules.
• Then, for an arbitrary module, use the existence of Gorenstein flat covers in GF-closed settings, or coverings by strongly FP-injective modules, to reduce the existence of special precovers to the previously solved case via elementary diagram chasing and pullback arguments (Estrada et al., 2016, Estrada et al., 2023).

In summary, the machinery of hereditary cotorsion pairs, transfinite extensions, and closure under direct summands grounds the systematic construction of special Gorenstein projective precovers.

4. Applications and Examples Across Module and Ring Classes

Gorenstein projective precovers have significant applications:

• In Gorenstein rings (noetherian with finite injective dimension), $\operatorname{GP} = \operatorname{GF}$ and every module admits a Gorenstein projective precover; the cotorsion pair $(\operatorname{GP}, \operatorname{GP}^{\perp_1})$ is complete hereditary (Estrada et al., 2016, Estrada et al., 2023).
• For commutative noetherian rings of finite Krull dimension, existence is assured, recapturing results of Jørgensen, Murfet–Salarian, and extending to derived category approaches (Estrada et al., 2023).
• In right coherent left $n$–perfect rings, every Gorenstein flat module has finite Gorenstein projective dimension.

Exotic examples of non-coherent left GF-closed rings with the requisite properties further demonstrate the breadth: matrix ring constructions provide families where $R$ is left GF-closed, every Gorenstein projective is Gorenstein flat and has finite Gorenstein projective dimension, but $R$ is not right coherent (Estrada et al., 2016).

A summary of key situations is presented:

Ring Class	Notable Feature	Reference
$R$ of finite Gorenstein global dim.	GP special precovering by dimension bound	(Estrada et al., 2023)
Gorenstein, two-sided noetherian	GP = GF, cotorsion pair complete	(Estrada et al., 2016)
Commutative noetherian, Krull finite	Derived category approach, dualizing cplx	(Estrada et al., 2023)
Right coherent, left $n$–perfect	Gpd of flats bounded $\leq n$, precovers	(Estrada et al., 2016)
Non-coherent GF-closed rings	Matrix ring examples	(Estrada et al., 2016)

5. Connections, Generalizations, and Open Questions

The reduction theorem unifies disparate prior results and reduces the existence question to checking a finitely generated or finitely presented subcategory. The framework extends to Ding projective modules: if every finitely presented module has Ding projective dimension bounded by $n$, then the class of Ding projectives is special precovering (Estrada et al., 2023). Analogous results for semidualizing bimodules and Auslander/Bass classes have recently been obtained (Becerril, 31 Jan 2026).

However, the general question—whether special Gorenstein projective precovers exist over an arbitrary ring without the bounded syzygy hypothesis—remains open (Estrada et al., 2023). This is also linked to the Finitistic Dimension Conjecture: if the finitistic projective dimension is finite and every finitely generated module has finite Gorenstein projective dimension, then $\operatorname{GP}$ is special precovering (Estrada et al., 2023).

Recent advances characterize rings with $\operatorname{GP}$ special precovering via strongly FP-injective dimensions and the existence of suitable (projective) cotorsion pairs (Becerril, 31 Jan 2026).

6. Structural Properties and Closure Under Operations

The class of modules admitting special Gorenstein projective precovers forms a robust subcategory, closed under extensions and stable under direct summands with projectives (Zhao et al., 2017). If the class of projectives is a generator for the right 1-orthogonal of $\operatorname{GP}$, this subcategory is the minimal projectively resolving subcategory containing $\operatorname{GP}$ and its Ext-orthogonal, and every object therein admits a special Gorenstein projective precover with projective source (Zhao et al., 2017).

Special precovering is also shown to be compatible with module-theoretic constructs such as triangular matrix rings and comma categories; for instance, over the triangular matrix ring $$\Lambda = \begin{pmatrix} R & M \ 0 & S \end{pmatrix}$$, the class of Gorenstein projective modules is special precovering if and only if it is so over both $R$ and $S$ (Hu et al., 2019).

---

In summary, the existence and theory of Gorenstein projective precovers depend fundamentally on finiteness properties in the subcategory of finitely presented modules, reduction to complete hereditary cotorsion pairs, and diagram-chasing in extensions and syzygies. This theory bridges classical homological algebra and contemporary finiteness conjectures, and remains a principal tool in the current study of relative homological invariants in module and representation theory (Estrada et al., 2023, Estrada et al., 2016, Becerril, 31 Jan 2026).