Deformation Framework for C*-Algebras

• The framework deforms C*-algebras by twisting coactions with circle-valued Borel 2-cocycles, unifying prior deformation methods.
• It preserves crucial structural properties such as nuclearity and K-theory invariance, ensuring consistent behavior after deformation.
• The approach employs Landstad duality and twisted crossed products to accommodate deformations for both abelian and nonabelian groups.

A deformation framework for $C^*$-algebras provides a systematic machinery for constructing new $C^*$-algebras from a given $C^*$-algebra $A$ by introducing a deformation parameter, typically via coactions, actions, cocycles, twists, or more elaborate "fusion data". In the modern context, this framework is formulated for a $C^*$-algebra $A$ equipped with a (maximal, reduced, or exotic) coaction of a second-countable locally compact group $G$, together with a circle-valued Borel $2$-cocycle on $G$. This expansive approach subsumes earlier deformation methods based on group actions, groupoid twists, or Rieffel quantization, and demonstrates K-theory invariance, nuclearity preservation, and continuity in fields under the presence of appropriate hypotheses (Buss et al., 2023).

1. Coactions and Structural Setup

Let $G$ be a second-countable locally compact group, and let $A$ be a $C^*$-algebra. A (nondegenerate) coaction of $G$ on $A$ is an injective $*$-homomorphism

$\delta: A \to M(A \otimes C^*(G))$

satisfying coassociativity,

$$(\delta \otimes \operatorname{id}) \circ \delta = (\operatorname{id} \otimes \Delta_G) \circ \delta,$$

where $\Delta_G$ is the comultiplication on $C^*(G)$, and the "spectral subspace" density condition,

$$\operatorname{span}\{(1 \otimes C^*(G))\delta(A)\} = A \otimes C^*(G).$$

For a given coaction $\delta$, one forms the crossed product $B := A \rtimes_\delta G$, which carries a dual action $\beta: G \to \operatorname{Aut}(B)$. There is also a nondegenerate $\beta$-equivariant embedding $j_{C_0(G)}: C_0(G) \to M(B)$. The data $(B, \beta, j_{C_0(G)})$ is called a weak $G\times G$-algebra, and every such triple, under mild hypotheses, reconstructs $(A, \delta)$ via Landstad duality (Buss et al., 2023).

A circle-valued Borel $2$-cocycle is a Borel function $\omega: G \times G \to \mathbb{T}$ satisfying

$$\omega(s, e) = \omega(e, s) = 1,\;\;\; \omega(s, t)\omega(st, u) = \omega(s, tu)\omega(t, u)\;\;\forall\,s,t,u\in G.$$

2. Deformation Construction via Coactions and Cocycle Twist

The deformation procedure operates by twisting the dual action $\beta$ of $G$ on $B$ via the $2$-cocycle $\omega$. Specifically, one defines a unitary $1$-cocycle $U^\omega: G \to UM(B)$ by

$U^\omega(s) = j_{C_0(G)}(r \mapsto \omega(r, s)),$

and sets

$$\beta^\omega_s = \operatorname{Ad}(U^\omega(s)) \circ \beta_s,$$

yielding a new (twisted) dual action. The Landstad subalgebra for the twisted action is

$$A^\omega = \{ m \in M(B)^{\beta^\omega} : f m g \in B_c\;\;\forall\,f,g\in C_c(G) \},$$

with $B_c = C_c(G) \cdot B \cdot C_c(G)$. On dense subalgebras $A_c = C_c(G) \cdot A \cdot C_c(G)$, the product and involution are given by convolution twisted by $\omega$:

$$a \star_\omega b = \int_{G \times G} \omega(s,t)\,(\delta_s(a)\,\delta_t(b))\, ds\,dt,$$

where $$\delta_s(a) = (\operatorname{id} \otimes \operatorname{ev}_s)\,\delta(a)$$ (Buss et al., 2023).

The completion in an appropriate $C^*$-norm yields the deformed $C^*$-algebra $A^\omega$, with variants: maximal ($A^\omega_{\max}$), reduced ($A^\omega_r$), or any intermediate ("exotic") completion, depending on the coaction norm chosen.

3. Equivalence, Duality, and Structural Properties

Under Landstad duality, the deformed weak $G \times G$-algebra $(B, \beta^\omega, j)$ is isomorphic to $$(A^\omega \rtimes_{\delta^\omega} G, \widehat{\delta^\omega}, j_{C_0(G)})$$, where $\delta^\omega$ is the deformed coaction on $A^\omega$. For maximal and reduced coactions with continuous $\omega$, the deformation agrees with the frameworks of Kasprzak and Bhowmick–Neshveyev–Sangha. In the reduced (normal) case, $A^\omega_r$ is isomorphic to the Landstad algebra of the directly twisted crossed product $B \rtimes_{\beta, \omega} G$ (Buss et al., 2023, Bhowmick et al., 2012).

If $G$ is abelian, the Fourier transform identifies these constructions with Rieffel–Kasprzak deformation theories. For discrete $G$, the framework reproduces and extends twisted Fell bundle and graph algebra deformations (Raeburn, 2016, Buss et al., 2024). In the presence of a representation group $Z \to H \to G$ (e.g. when $X = \widehat{Z} \simeq H^2(G, \mathbb{T})$ is nontrivial), continuous families of deformations assemble into $C_0(X)$-algebras forming $C^*$-bundles over $X$.

4. K-Theory, Nuclearity, and Continuity of Deformation

The deformation framework preserves significant structural and homological properties:

• If the action $\beta$ is amenable (for example $G$ is amenable), all completions coincide and nuclearity is preserved: $A^\omega$ is nuclear iff $A$ is nuclear.
• For continuous families of cocycles $\{\omega_x\}_{x \in X}$ parametrized by a locally compact space $X$, the family $\{A^{\omega_x}\}_{x\in X}$ forms an upper-semicontinuous—and, under exactness, continuous—field of $C^*$-algebras (Buss et al., 2023, Steeger et al., 2021, Belmonte et al., 2011, Raeburn, 2016).
• If $G$ satisfies the Baum–Connes conjecture with coefficients, and $\omega_0,\omega_1$ are homotopic as $2$-cocycles, $K_*(A^{\omega_0}) \cong K_*(A^{\omega_1})$ for all crossed-product functors. If $G$ is also $K$-amenable, isomorphism extends to all intermediate completions. In the strong Baum–Connes case, fiber evaluation maps in such continuous fields are KK-equivalences (Buss et al., 2023, Bhowmick et al., 2012, Yamashita, 2011).

5. Examples and Unification of Deformation Paradigms

This deformation framework encompasses and clarifies a breadth of existing constructions:

• For $G=\mathbb{R}^n$ and $\delta$ from a continuous action, the theory reproduces Rieffel's strict deformation by skew-form matrices.
• For $G$ abelian, with $\delta$ the dual coaction, the framework yields the familiar noncommutative torus, with deformation parameter induced by the cocycle (Buss et al., 2023, Bhowmick et al., 2012, Buss et al., 4 Jul 2025).
• For $G$ non-abelian but possessing a representation group, e.g., $G = PSL(2,\mathbb{R})$, the family of deformations indexed by $\mathbb{T}$ yields a continuous $C^*$-bundle, with all fibers KK-equivalent even when $G$ is nonamenable.
• For discrete groups and their Fell bundles, direct deformation at the bundle level and at the coaction level are canonically equivalent, and the theory unifies graph algebra twistings (Raeburn, 2016, Buss et al., 2024).
• In the setting of locally compact quantum groups, spectral fusion deformations parametrized by fusion data extend the above constructions to more general contexts, and capture Drinfeld and non-group-theoretic deformations (Sangha, 13 Jan 2026).

6. Connections with Quantum Groups and Further Generalizations

The framework extends to deformations by unitary $2$-cocycles on the duals of locally compact quantum groups, yielding new deformed $C^*$-algebras $A_\Omega$ with well-developed Morita stability, crossed-product duality, and regularity theorems (Neshveyev et al., 2013). For quantum group coactions, spectral fusion deformations allow for associators and higher $3$-cocycle invariants, producing genuinely new algebraic structures that lie outside the reach of crossed-product or classical dual cocycle methods (Sangha, 13 Jan 2026).

For strict deformation quantization in the sense of Rieffel, deformation can also be realized as a functor on continuous fields of $C^*$-algebras, associating to Poisson vector bundles continuous bundles of deformed $C^*$-algebras equipped with the fiberwise Weyl–Moyal product (Forger et al., 2014, Steeger et al., 2021).

7. Synthesis and Structural Table

Below is a summary of the principal deformation mechanisms unified by the coaction framework:

Deformation Data	Construction Method	Example Cases
Group coaction + 2-cocycle	Landstad duality, twisted action	Rieffel deformation, noncommutative tori (Buss et al., 2023)
Fell bundle + cocycle	Twisted fiberwise multiplication	Twisted (k-)graph algebras (Raeburn, 2016, Buss et al., 2024)
Continuous field	$C_0(X)$-algebra over parameter	Strict quantization bundles (Steeger et al., 2021, Belmonte et al., 2011)
Quantum group coaction + fusion data	Spectral fusion algebraic rules	Drinfeld/Connes-Landi/Moyal type (Sangha, 13 Jan 2026)

The deformation framework for $C^*$-algebras via coactions, as developed by Buss–Echterhoff and extended by subsequent authors, provides a robust, unifying, and highly flexible operator algebraic infrastructure supporting deformations by group-theoretic, cohomological, and categorical data, with profound implications for $K$-theory, noncommutative geometry, and representation theory (Buss et al., 2023, Buss et al., 4 Jul 2025, Bhowmick et al., 2012, Raeburn, 2016, Sangha, 13 Jan 2026).