Abstract
This paper investigates how motivic Donaldson–Thomas (DT) invariants behave in families of noncommutative Calabi–Yau threefolds defined by quivers with homogeneous potentials. These families provide deformation quantizations of affine three-space, the resolved conifold, and the resolution of the transversal $A_n$-singularity. The authors find that while invariants are generically constant, they exhibit "jumps" at special values of the deformation parameter, such as roots of unity. Generating series are derived in closed form as plethystic exponentials of simple rational functions.
Introduction
In classical algebraic geometry, Donaldson–Thomas (DT) invariants are used to "count" sheaves on 3D Calabi–Yau varieties. These numerical counts are typically deformation invariant, meaning they remain stable under small perturbations of the space's geometry. However, modern mathematical physics and geometry increasingly focus on noncommutative spaces, where coordinates do not commute ($xy \neq yx$).
This research explores the behavior of motivic DT invariants—a more refined version of these counts that takes values in a "ring of motives"—when a standard 3D space is "quantized" into a noncommutative algebra. The central question is whether these motivic counts remain constant as the quantization parameter $q$ varies, or if the noncommutativity introduces novel phenomena.
Noncommutative Threefolds and Quivers
The authors represent noncommutative threefolds using quivers with potentials. A quiver is a directed graph where nodes represent spaces and arrows represent maps between them. A potential is a polynomial that defines the relations between these maps.
For example, quantum affine three-space can be represented by a single-vertex quiver with three loops. The noncommutativity is introduced by relations like $xy = qyx$, where $q$ is the quantization parameter.
Computational Methods: Dimensional Reduction
To compute these invariants, the authors utilize dimensional reduction. By identifying a "cut" in the quiver, they reduce the 3D problem of counting representations to a 2D problem. This allows the application of known motivic techniques to handle the complexity.
The results are expressed using plethystic exponentials (the "Exp" operator), which package complex enumerative data into generating series. This transformation reveals that the invariants, while complex in their raw form, simplify into combinations of rational functions.
Results for Specific Geometries
The study provides explicit, closed-form formulas for several fundamental noncommutative threefolds:
- Quantum Affine Three-Space: Theorem 3.1 provides the first universal DT series for this space, showing an additional term that appears only when $q$ is a primitive $r$-th root of unity.
- The Resolved Conifold: A classic example in string theory, represented by a multi-vertex quiver.
- Cyclic Quivers ($A_n$-singularities): Theorems 3.5 and 3.6 provide formulas that recover known results for the commutative case ($q=1$) while extending them to the generic quantized case.
The "Jump" Phenomenon
A key discovery is that motivic DT invariants are not strictly deformation invariant in the noncommutative setting. They are "generically" constant but "jump" at specific values of $q$, particularly at roots of unity.
This behavior is linked to the appearance of new high-dimensional "simple" (indivisible) representations of the algebra at these special values. When the quantized algebra becomes a finite module over its center, the enumerative properties change abruptly.
Conclusion and Future Directions
While the dimensional reduction technique is powerful, it is limited to cases where the potential is linear in at least one variable. This currently precludes a full proof for more general cases like Sklyanin algebras, where the potential is cubic. However, the authors propose bold conjectures for these algebras, suggesting that the simplicity of the plethystic exponential form is a universal feature. Future work will likely focus on multi-vertex quivers and the deeper physical implications of these "jumps" for BPS states in string theory.
Summary updated