---
title: Operator System Duality in Triangular Prisms
url: https://www.emergentmind.com/topics/operator-system-duality
type: topic
---

# Operator System Duality in Triangular Prisms

A noncommutative triangular prism is a mathematical structure arising at the interface of noncommutative operator systems and fusion category theory, characterized by the operator system generated inside the group C*-algebra $C^*(Z_3 * Z_2)$ and by the triangular prism equations that encode associativity constraints in categorical settings. This object generalizes the classical triangular prism, capturing the joint matrix range of certain generators and embodying extreme noncommutative geometric and categorical phenomena. It plays a pivotal role in dilation theory, convexity analysis, and categorification.

## 1. Operator System Construction and Canonical Generators

The noncommutative triangular prism operator system, denoted $S_{\mathrm{prism}}$, is constructed within the full group C*-algebra $C^*(Z_3 * Z_2)$, where $Z_3$ is the cyclic group of order $3$ (generator $w$) and $Z_2$ is the cyclic group of order $2$ (generator $v$). The free product group $G = Z_3 * Z_2$ underlies the structure. The operator system $S_{\mathrm{prism}}$ is the four-dimensional linear span
\[
S_{\mathrm{prism}} := \mathrm{span}\{ 1, w, w^2, v \} \subset C^*(G),
\]
equipped with the inherited *-vector-space structure, order unit, and operator system cones. The canonical generators satisfy
- $w^3 = 1$
- $v^2 = 1$
- $w^* = w^2$
- $v^* = v$

Alternatively, the selfadjoint coordinate generators are
\[
x = \Re(w) = \frac{w + w^2}{2}, \quad y = \Im(w) = \frac{w - w^2}{2i}, \quad z = v
\]
yielding $S_{\mathrm{prism}} = \mathrm{span}\{1, x, y, z\}$. The noncommutative triangular prism is identified with the joint matrix range of $(x, y, z)$ under completely positive unital maps from $S_{\mathrm{prism}}$ to matrices [2601.16902].

## 2. Dilation Theorems and Joint Unitary Dilations

The core structure of the noncommutative triangular prism is governed by a pairing of classical dilation theorems. For a Hilbert space $H$, operators $(T_1, T_2, T_3) \in B(H)^3$ are realized within $S_{\mathrm{prism}}$ if and only if there exist:
- A Hilbert space $K \supset H$
- Unitary $U \in B(K)$, $U^3 = 1$
- Selfadjoint unitary $V \in B(K)$, $V^2 = 1$
- Isometry $Z: H \to K$ such that
    - $Z^*UZ = T_1 + i T_2$
    - $Z^*U^2Z = T_1 - i T_2$
    - $Z^*VZ = T_3$

The block-matrix dilations
\[
U = \begin{pmatrix} Y & 0 \\ 0 & 1 \end{pmatrix}, \quad
V = \begin{pmatrix} B & \sqrt{I - B^2} \\ \sqrt{I - B^2} & -B \end{pmatrix}
\]
manifest Mirman's dilation for $T_1 + iT_2$ to a unitary of order $3$, and Halmos's dilation of the selfadjoint contraction $T_3$ to a symmetry [2601.16902]. The universal property of $C^*(Z_3 * Z_2)$ and Stinespring’s theorem guarantee the existence of such joint dilations, giving an explicit operator-theoretic realization of the noncommutative triangular prism.

## 3. Extreme Points, Exactness, and Duality Properties

The noncommutative geometric features of $S_{\mathrm{prism}}$ are pronounced:
- **Matrix-Extreme Points**: For each $n \in \mathbb{N}$, the irreducible *-representations $\pi_n: C^*(Z_3 * Z_2) \to M_n(\mathbb{C})$ yield noncommutative extreme points at level $n$ of the maximal matrix convexity set, $P(3)^{\mathrm{max}}$. Realization via finite groups such as PSL$_2(\mathbb{F}_q)$ is possible for all $n$.
- **Infinite-dimensional Extremes**: The subgroup $F_2$ within $Z_3 * Z_2$ ensures the presence of type II$_1$, type II$_\infty$, and type III factorial representations, yielding extreme points at level $\aleph_0$.
- **Exactness**: $C^*(Z_3 * Z_2)$ is not exact as a C*-algebra, implying that $S_{\mathrm{prism}}$ fails to be exact as an operator system. However, being OMAX, $S_{\mathrm{prism}}$ does satisfy the lifting property.
- **Dual Operator System**: The operator-system dual $S_{\mathrm{prism}}^\delta$ is completely order isomorphic to the OMIN system
\[
\left\{ (z_1, z_2, z_3, z_4, z_5) \in \mathbb{C}^5 : z_1 + z_2 + z_3 = z_4 + z_5 \right\}
\]
encoding the relation for the classical prism $\mathrm{conv}\{1, \omega, \omega^2\} \times [-1, 1]$ [2601.16902].

## 4. Tensor Product Structure and Complete Positivity

Noncommutative tensor product phenomena are essential to $S_{\mathrm{prism}}$:
- The minimal, commuting, and maximal tensor products of $S_{\mathrm{prism}}$ are non-equivalent:
    \[
    S_{\mathrm{prism}} \otimes_{\mathrm{min}} S_{\mathrm{prism}} \neq S_{\mathrm{prism}} \otimes_c S_{\mathrm{prism}} \neq S_{\mathrm{prism}} \otimes_{\mathrm{max}} S_{\mathrm{prism}}
    \]
due to the embedding into $C^*(Z_3 * Z_2) \otimes_{\mathrm{min}} C^*(Z_3 * Z_2)$, which contains $F_2 \otimes F_2$.
- Every positive map out of or into $S_{\mathrm{prism}}$ is automatically completely positive, reflecting extreme noncommutative convexity [2601.16902].

## 5. Triangular Prism Equations in Fusion Category Theory

The triangular prism equations (TPE) provide a higher categorical analog of the associativity constraints for fusion categories. In a pivotal fusion category $\mathcal{C}$ over an algebraically closed field, TPE are formulated using nine objects $X_1,\ldots,X_9$ and six morphisms $\alpha_1,\ldots,\alpha_6$ with assigned bases in appropriate hom-spaces and automorphisms $\rho$, $\sigma$ tracking dualities and pivotal shifts. The central equation (Theorem 4.5) is:
\[
\sum_{\beta_0 \in B_0}
T\left( \rho^{-2}(\alpha_2), \rho(\alpha_3), \rho^{-1}(\alpha_1), \beta_0 \right)
T\left( \rho^{-1}(\alpha_5), \beta_0', \rho(\alpha_4), \rho^{-1}(\alpha_6) \right) =
\\
\sum_{X \in \mathcal{O}}
\sum_{\beta_i \in B_i}
\dim(X)
T\left(\rho^{-2}(\sigma(\beta_3')), \rho(\alpha_3), \rho^2(\beta_1), \rho^{-1}(\alpha_6)\right)
T\left(\rho^{-1}(\alpha_4), \alpha_1, \beta_1', \beta_2\right)
T\left(\rho^{-1}(\alpha_5), \alpha_2, \beta_2', \beta_3\right)
\]
where $T(-,-,-,-)$ evaluates a tetrahedral string diagram [2203.06522]. In the spherical case, these equations reduce to the standard pentagon equations via an explicit change of basis (Theorem 5.9).

## 6. Localization, Complexity, and Applications

TPE localization enables direct management of categorification complexity. By selecting a small variable set, writing subsystem equations, solving via Gröbner bases, and propagating solutions, the otherwise intractable system of pentagon or TPE can be reduced. This stratagem permits the exclusion of certain fusion ring categorifications and informs the classification of unitary 1-Frobenius simple integral fusion categories up to rank $8$ and FP-dimension $20{,}000$, with the only solutions being representation rings of $\mathrm{PSL}(2, q)$ for specified $q$ values.

## 7. Noncommutative Categorical and Geometric Implications

TPE are inherently noncommutative—no commutativity or braiding is assumed, and edges record evaluation orientation. The automorphisms $\rho$, $\sigma$ maintain correct label ordering in absence of spherical or symmetric structure, and the localization technique utilizes multiplication rules in the Grothendieck ring, which need not commute. These features make the triangular prism a unifying structure for the full noncommutative tensor-category pentagon system, local subsystems of F-symbols, refined indicator theorems, and obstruction criteria for categorification. In the commutative (spherical) case, the classical pentagon constraint is recovered via explicit basis change [2203.06522].

---

For detailed and rigorous proofs, see "Representations of noncommutative cubes and prisms" [2601.16902] and "Triangular Prism Equations and Categorification" [2203.06522].

Source: https://www.emergentmind.com/topics/operator-system-duality