🔢 mathematics

Accessibility and Gorenstein injective envelopes

This paper establishes that the Gorenstein injective cotorsion pair in a Grothendieck category is complete if and only if the category admits a set of Tate trivial generators, a result derived from the accessibility of orthogonal classes that further guarantees the existence of Gorenstein injective envelopes and induces an injective abelian model structure.

Original authors: Sergio Estrada, James Gillespie

Published 2026-05-05

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sergio Estrada, James Gillespie

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a vast, infinite library called G. This isn't a library of books, but a library of mathematical objects (like shapes, numbers, or abstract structures) that follow specific rules. Mathematicians call this a "Grothendieck category."

The goal of this paper is to solve a specific problem within this library: How do we find the "best possible wrapper" for any object in the library?

In the world of standard math, we know how to wrap things in "injective envelopes" (think of it as putting a fragile object in the strongest, most protective bubble wrap possible). This has been known for a long time. However, the authors are interested in a newer, more complex type of wrapper called a "Gorenstein injective envelope." These are special wrappers that work for a more advanced branch of math called "Gorenstein homological algebra."

For a long time, mathematicians didn't know if every object in this giant library G could get one of these special wrappers. Sometimes, the library is too messy or lacks the right tools to guarantee a wrapper exists.

The Big Discovery: The "Perfect Fit" Rule

The authors, Sergio Estrada and James Gillespie, discovered a simple rule to determine when these special wrappers are guaranteed to exist.

They found that the library G must have a special set of "building blocks" (which they call generators). But not just any building blocks will do. These blocks must be "Tate trivial."

The Analogy:
Imagine you are trying to build a fortress (the wrapper) around a castle (the object).

The Old Way: You try to build the fortress using whatever materials you can find. Sometimes you run out of bricks, and the fortress collapses.
The New Rule: The authors say, "If you have a specific, pre-approved set of high-quality bricks (the Tate trivial generators) that are easy to work with, then you can always build a perfect fortress for any castle in the library."

If the library has these special bricks, then:

Completeness: Every object gets a wrapper. No one is left out.
Perfection: The system of wrappers is "perfect," meaning it works smoothly and predictably.
Model Structure: It creates a "map" (called a model structure) that helps mathematicians navigate the library, treating certain objects as if they don't exist (turning them into zero) to simplify complex problems.

The Secret Ingredient: "Accessibility"

How did they prove this? They used a concept called Accessibility.

Think of the library G as a massive, chaotic warehouse. You can't look at every single item at once. However, the authors realized that the "special wrappers" (the right side of their mathematical pair) are actually built from a manageable, finite set of smaller, simpler items.

The Metaphor: Imagine trying to describe a giant ocean. You can't list every drop of water. But if you realize that every drop of water is just a combination of a few specific types of molecules, you can describe the whole ocean by studying just those molecules.
The Paper's Claim: The authors proved that the class of objects needing these special wrappers is "accessible." This means they are all built from a small, manageable "set" of simpler objects. Because they are built from a set, we can use standard mathematical tools to prove that the wrappers exist.

Real-World Examples (in Math Land)

The paper shows that this rule applies to many important mathematical libraries:

Quasi-coherent sheaves on a scheme: This is a fancy way of describing geometric shapes defined by equations (like curves and surfaces). The authors show that if the shape is "quasi-compact and semi-separated" (a technical way of saying it's not too wild or infinite in a bad way), it has these special generators, and therefore, every object in it gets a Gorenstein injective envelope.
Ding Injectives and FPn-Injectives: These are other types of "special wrappers" mathematicians have been trying to find. The authors' method proves these exist too, without needing extra assumptions about the library.

What They Did Not Do

It is important to stick to what the paper actually says:

They did not invent new physical applications (like medical uses or engineering).
They did not claim this works for every possible mathematical library. They specifically identified the condition (having Tate trivial generators) where it works. They even gave an example (Neeman's example) of a library where this fails, proving that the condition is necessary.
They did not extend the results to future, unproven theories. They strictly proved the existence of these envelopes and the "perfect" nature of the system under the conditions they defined.

Summary

In short, Estrada and Gillespie solved a puzzle about "protective wrappers" in advanced math. They proved that if a mathematical library has a specific, manageable set of "building blocks" (Tate trivial generators), then every single object in that library is guaranteed to have a perfect, Gorenstein injective envelope. They used the idea of "accessibility" (breaking big problems down into manageable sets) to prove this, opening the door to understanding many complex mathematical structures that were previously too messy to handle.

Technical Summary: Accessibility and Gorenstein Injective Envelopes

Problem Statement
The paper addresses a fundamental open problem in Gorenstein homological algebra regarding the existence of Gorenstein injective envelopes in general Grothendieck categories. While the existence of Gorenstein injective envelopes was recently established for module categories ( $R$ -Mod) by Šaroch and Šťovíček [SS20], the extension to arbitrary Grothendieck categories $\mathcal{G}$ remained unresolved. Specifically, the authors investigate the conditions under which the class of Gorenstein injective objects, denoted $\mathcal{GI}$ , forms the right half of a perfect cotorsion pair $(^\perp\mathcal{GI}, \mathcal{GI})$ . A perfect cotorsion pair implies that $\mathcal{GI}$ is enveloping (every object has a Gorenstein injective envelope) and its left orthogonal class is covering. Previous results, such as those by Krause [Kra05], were limited to locally noetherian categories with compactly generated derived categories or categories possessing small projective generators. The central question is: under what conditions does this structure hold for general Grothendieck categories, potentially lacking enough projectives?

Methodology
The authors employ a methodology rooted in the theory of accessible and locally presentable categories, specifically leveraging the concept of accessibility to analyze cotorsion classes. The core technical innovation is the application of results from Adámek and Rosický [AR94] to homological algebra, a connection the authors note has been underutilized in the field.

The proof strategy proceeds as follows:

Accessibility of Right Cotorsion Classes: The authors establish that for any set of objects $S$ in an accessible additive category, the right Ext-orthogonal class $B := S^\perp$ is an accessibly embedded, accessible subcategory. This is formalized in Theorem 2.5, which proves that such classes are closed under sufficiently large direct limits and possess a set of presentable generators.
Model Structure Construction: Utilizing the accessibility of these classes, the authors construct cofibrantly generated abelian model structures on the category of chain complexes $\text{Ch}(\mathcal{G})$ . They define specific classes of "acyclic" complexes (e.g., $A$ -acyclic complexes of injectives) and show that these classes coincide with the fibrant objects of these model structures.
Tate Trivial Generators: The authors introduce the concept of "Tate trivial generators" (Definition 5.3). A set of generators $U$ for $\mathcal{G}$ is called $B$ -Tate trivial if every object in $U$ is "trivial" with respect to the Gorenstein B-injective model structure (i.e., $U \subseteq {}^\perp \mathcal{GIB}$ ).
Quillen Equivalence: The paper establishes a Quillen equivalence between the Gorenstein B-injective model structure on $\mathcal{G}$ and an "exact B-acyclic" model structure on $\text{Ch}(\mathcal{G})$ . This equivalence is mediated by the sphere functor $S_0(-)$ and the 0-cycle functor $Z_0(-)$ .

Key Contributions and Results
The paper provides a complete characterization of when Gorenstein injective cotorsion pairs are perfect in Grothendieck categories.

Main Theorem (Corollary 6.3 / Theorem 5.5): Let $\mathcal{G}$ $G$ be a Grothendieck category and $B := S^\perp$ $B := S^{⊥}$ for a set $S$ $S$ . The cotorsion pair $(^\perp \mathcal{GIB}, \mathcal{GIB})$ $(^{⊥} G I B, G I B)$ is perfect (and thus $\mathcal{GIB}$ $G I B$ is enveloping) if and only if $\mathcal{G}$ $G$ admits a set of $B$ $B$ -Tate trivial generators.
- In the specific case where $B$ is the class of all injectives, the pair $(^\perp \mathcal{GI}, \mathcal{GI})$ is perfect if and only if $\mathcal{G}$ admits a set of Tate trivial generators.
Locally Finite Dimensional Categories: The authors define a Grothendieck category as locally finite dimensional if it admits a generating set of objects with finite projective dimension. They prove (Lemma 4.4) that in such categories, the condition of having Tate trivial generators is automatically satisfied. Consequently, Gorenstein injective envelopes exist in all locally finite dimensional Grothendieck categories.
Generalization to Relative Gorenstein Objects: The framework is extended to "Gorenstein B-injective" objects, "Ding injective" objects, "Gorenstein AC-injective" objects, and "Gorenstein $FP_n$ -injective" objects. The paper proves that envelopes exist for all these classes in any Grothendieck category admitting a set of generators analogous to the Tate trivial condition (e.g., finite projective dimension).
Model Structure Properties: The resulting model structures are shown to be injective, cofibrantly generated, and "best possible" in the sense that they kill the class $B$ in the homotopy category and are minimal among injective model structures with this property.

Significance and Applications
The paper claims to resolve the existence of Gorenstein injective envelopes in a broad class of categories previously inaccessible to these methods, specifically those without enough projectives.

Quasi-Coherent Sheaves: A primary application is the category of quasi-coherent sheaves, $\text{Qco}(X)$ , on a quasi-compact and semi-separated scheme $X$ . The authors demonstrate that $\text{Qco}(X)$ is locally finite dimensional (admitting generators of finite projective dimension via very flat sheaves), thereby guaranteeing the existence of Gorenstein injective envelopes for all quasi-coherent sheaves on such schemes. This strengthens previous results that were limited to semi-separated Noetherian schemes.
Relative Homological Algebra: The results unify and generalize the existence of envelopes for Ding injectives and Gorenstein $FP_n$ -injectives, removing the necessity for the category to be $n$ -coherent or locally coherent, provided the generator condition is met.
Theoretical Unification: By identifying the accessibility of right cotorsion classes as the key mechanism, the paper provides a unified approach to constructing model structures and proving the existence of envelopes without relying on specific ring-theoretic properties or the existence of projective generators.

The authors note that while their results cover a vast array of examples, they do not claim universality for all Grothendieck categories; they cite Neeman's example [Nee14] of a locally noetherian category that does not admit a set of Tate trivial generators, indicating that the condition is non-trivial and necessary.

The Big Discovery: The "Perfect Fit" Rule

The Secret Ingredient: "Accessibility"

Real-World Examples (in Math Land)

What They Did Not Do

Summary

More like this