On differential smoothness of certain Artin-Schelter regular algebras of dimension 5

This article investigates the differential smoothness of various five-dimensional Artin-Schelter regular algebras. By analyzing the relationship between the number of generators and the Gelfand-Kirillov dimension, we provide structural obstructions to differential smoothness in specific algebraic families. In particular, we prove that certain two- and four-generator AS-regular algebras of global dimension five fail to admit a differential calculus, while a five-generator graded Clifford algebra provides a contrasting positive example.

💡 Research Summary

The paper investigates differential smoothness for a variety of five‑dimensional Artin‑Schelter (AS) regular algebras. Differential smoothness, as introduced by Brzeziński and Sitarz, requires the existence of an n‑dimensional connected integrable differential calculus (Ω(A), d) on an affine algebra A whose Gelfand‑Kirillov dimension (GKdim) equals n. Such a calculus must admit a volume form ω that generates the top‑degree module Ωⁿ(A) as a free right (and left) A‑module, and there must exist a compatible complex of integral forms (I(A), ∇) linked by a Hodge‑star‑type isomorphism.

The authors first review the classification of AS‑regular algebras in low dimensions (1–4) and note that the five‑dimensional case remains largely open. They recall known families: two‑generator algebras classified via Z₂‑gradings (Zhou‑Lu), four‑generator algebras arising from deformations of universal enveloping algebras of graded Lie algebras (Li‑Wang), and a five‑generator graded Clifford algebra.

The central technical contribution is Theorem 3.3, which states that any AS‑regular algebra whose number of generators is strictly smaller than its GKdim cannot be differentially smooth. The proof hinges on the density condition for a differential calculus: the top‑degree module Ωⁿ(A) must be generated by a single volume form. If the number of generators is insufficient, the graded components of Ω(A) cannot achieve the required rank, preventing the construction of a free right A‑module generator and consequently obstructing the existence of a Hodge‑star isomorphism. This yields a structural obstruction that applies uniformly to all five‑dimensional AS‑regular algebras with fewer generators than GKdim.

Applying this obstruction, the paper examines two‑generator AS‑regular algebras from the family X (Zhou‑Lu). Although these algebras have GKdim ≥ 4, Theorem 3.3 shows they cannot support a five‑dimensional integrable calculus; thus they are not differentially smooth. Similarly, the four‑generator algebras of Li‑Wang, defined by quadratic relations with non‑zero parameters satisfying abcd = 1, also fail the generator‑versus‑dimension test and are shown to lack differential smoothness.

In contrast, the authors present a positive example: a five‑generator graded Clifford algebra C. This algebra satisfies global dimension 5, GKdim 5, and enjoys Koszul, strongly Noetherian, Auslander‑regular, and Cohen‑Macaulay properties. The authors explicitly construct a differential calculus (Ω(C), d) where Ω⁵(C) is freely generated by a volume form ω, and they identify an algebra automorphism ν such that the right‑twisted module νI⁰(C) is isomorphic to Ω⁵(C). They then define a divergence map ∇ on the integral forms and verify the required commutative diagrams, establishing that (Ω(C), d) is integrable. Consequently, C is differentially smooth, providing a concrete counterpoint to the negative results.

The paper concludes by emphasizing two lessons. First, a mismatch between the number of generators and GKdim serves as a universal obstruction to differential smoothness in five‑dimensional AS‑regular algebras. Second, when the number of generators equals GKdim and the algebra possesses additional symmetries (as in graded Clifford algebras), a full integrable calculus can be built, confirming differential smoothness. These findings enrich the noncommutative geometry toolkit, offering a clear algebraic criterion for “smoothness” in the challenging setting of dimension five and suggesting directions for future classification efforts.