Operator-Valued Toeplitz/Laurent Coupling

• Operator-valued Toeplitz/Laurent coupling is the framework linking Toeplitz operator systems with Laurent polynomial systems via duality, tensor products, and matrix order structures.
• The approach employs a single-matrix, Choi-like complete positivity criterion that verifies positivity in tensor products through a universal Toeplitz matrix.
• This framework reveals significant implications for nuclearity and the weak expectation property, enhancing our understanding of separability and entanglement in operator systems.

Operator-valued Toeplitz/Laurent coupling refers to the interplay between finite-dimensional operator systems constructed from Toeplitz matrices and those consisting of Laurent (trigonometric) polynomials, particularly in the context of tensor products, duality, and the analysis of separability and entanglement in the associated tensor cones. The formalism connects matrix order structures, tensor-categorical relationships, and criteria for complete positivity, with implications for nuclearity and the weak expectation property in operator system theory (Farenick, 2023).

1. Structure of Toeplitz and Laurent Operator Systems

The $n \times n$ Toeplitz operator system, denoted $\mathcal{T}_n$, is realized as a unital *-subspace of $M_n(\mathbb{C})$ generated by shifted versions of the unilateral shift matrix $S$, with basis elements $r_\ell = S^\ell$ for $\ell \geq 0$ and $r_\ell = (S^*)^{-\ell}$ for $\ell < 0$, spanning $$\mathcal{T}_n = \operatorname{span}\{r_{-n+1}, \ldots, r_0, \ldots, r_{n-1}\}$$. The inherited matrix-order structure from $M_n(\mathbb{C})$ gives $$\mathcal{T}_n^+ = \{T \in M_n(\mathbb{C}) \mid T = T^*, T_{i,j} = T_{i-j}, T \geq 0\}$$, with $1_n$ as the Archimedean order unit.

The degree-$d$ Fejér–Riesz (Laurent) operator system, denoted $\mathcal{F}_d$, is the unital *-subspace of $C(S^1)$ spanned by $\{x^k \mid -d+1 \leq k \leq d-1\}$, where $x^k$ are trigonometric monomials $z \mapsto z^k$ on the unit circle $S^1 \subset \mathbb{C}$. The order structure $$\mathcal{F}_d^+ = \{f \in \mathcal{F}_d \mid f(z) \geq 0 \ \forall z \in S^1\}$$ is induced from $C(S^1)$, with the constant function $1$ as its order unit (Farenick, 2023).

2. Duality Between Toeplitz and Fejér–Riesz Systems

The fundamental duality is encapsulated in a unital complete-order isomorphism $\Theta : \mathcal{T}_n \to (\mathcal{F}_n)^*$ given by

$\Theta(T)[f] = \sum_{k=-n+1}^{n-1} T_{-k} a_k$

for $T = [T_{i-j}]_{i,j=1}^n \in \mathcal{T}_n$ and $$f(z) = \sum_{k=-n+1}^{n-1} a_k x^k \in \mathcal{F}_n$$. This establishes a categorical duality: $$\mathcal{T}_n \cong (\mathcal{F}_n)^*,\quad \mathcal{F}_n \cong (\mathcal{T}_n)^*,$$ with dual pairing $\langle r_\ell, x^k \rangle = \delta_{\ell,k}$. This dual structure links the Toeplitz and Laurent systems in the operator system category (Farenick, 2023).

3. Tensor Products and Tensor Cones in Operator Systems

Given operator systems $R, T$, two canonical tensor-product orderings are central:

• The minimal tensor product $(R \otimes_{\min} T)^+$ is defined via positivity preservation under all completely positive unital maps into matrix algebras.
• The maximal tensor product $(R \otimes_{\max} T)^+$ is the minimal matrix cone making all product unital completely positive maps positive.

For any operator-system tensor product $\otimes_\alpha$ and finite-dimensional operator systems $R, T$, the inclusion chain

$R^+ \otimes_{\sep} T^+ \subset (R \otimes_{\alpha} T)^+ \subset R^+ \otimes_{\sep^*} T^+$

holds, where the usual separable cone $R^+ \otimes_{\sep} T^+$ consists of finite sums $\sum_i a_i \otimes b_i$ with $a_i \in R^+$, $b_i \in T^+$, and its dual is $R^+ \otimes_{\sep^*} T^+$. The operator-system cone $(R \otimes_{\alpha} T)^+$ is a "tensor cone" in the sense of Namioka–Phelps, interpolating between separable and dual-separable positivity (Farenick, 2023).

Specifically, for $R = T = \mathcal{T}_n$ or $\mathcal{F}_n$: $\mathcal{T}_n^+ \otimes_{\sep} \mathcal{T}_n^+ \subset (\mathcal{T}_n \otimes_{\min} \mathcal{T}_n)^+ \subset (\mathcal{T}_n \otimes_{\max} \mathcal{T}_n)^+ \subset \mathcal{T}_n^+ \otimes_{\sep^*} \mathcal{T}_n^+,$ with parallel statements for $\mathcal{F}_n$.

4. Operator-Valued Coupling and the Single-Matrix CP Criterion

For any operator system $M$, elements $$X = \sum_{\ell=-n+1}^{n-1} r_\ell \otimes M_\ell \in \mathcal{T}_n \otimes M$$ correspond to linear maps $\Phi_X : \mathcal{F}_n \to M$ via $\Phi_X(x^\ell) = M_\ell$.

The principal criterion (Proposition 3.2 in (Farenick, 2023)) establishes that: $$X \in (\mathcal{T}_n \otimes_{\min} M)^+ \iff \Phi_X \text{ is completely positive} \iff \Phi_X^{(n)}(\boldsymbol{T}_n) \geq 0 \qquad \text{in } M_n(M)$$ where $$\boldsymbol{T}_n = \sum_{\ell=-n+1}^{n-1} r_\ell \otimes x^{-\ell} \in \mathcal{T}_n \otimes \mathcal{F}_n$$ is the universal positive Toeplitz matrix. This is directly analogous to the Choi matrix criterion for complete positivity of maps on $M_n(\mathbb{C})$. Thus, complete positivity can be decided by a single $n \times n$ positivity check on $\Phi_X^{(n)}(\boldsymbol{T}_n)$ (Farenick, 2023).

5. Entanglement, Separability, and Explicit Coupling Elements

The connection between operator-system tensor cones and entanglement is exemplified by explicit elements:

• The "maximally entangled" element in $\mathcal{T}_n \otimes \mathcal{F}_n$ is $$T_n = \sum_{\ell=-n+1}^{n-1} r_\ell \otimes x^{-\ell}$$, positive in the maximal tensor product but not in the minimal or separable cone. $T_n$ generates an extremal ray of the maximal cone and is entangled (Theorem 1.11) (Farenick, 2023).
• The "classical" separable coupling is $$R_n = \sum_{\ell=-n+1}^{n-1} r_\ell \otimes r_\ell \in \mathcal{T}_n \otimes \mathcal{T}_n$$, which is positive and separable (Proposition 4.10).
• Tensor-cone equalities hold: Toeplitz $\otimes_{\max}$ Toeplitz has exactly the separable cone at its base, Laurent $\otimes_{\min}$ Laurent has the dual-separable cone, while Toeplitz $\otimes_{\min}$ Toeplitz and Laurent $\otimes_{\max}$ Laurent are strictly larger (Farenick, 2023).

6. Categorical Consequences: Nuclearity and WEP

If $C^*(R)$ is nuclear, as holds for both $\mathcal{T}_n$ and $\mathcal{F}_n$ (since $C^*(R) \cong C(S^1)$), a suite of categorical results applies (Theorem 6.11):

• $R \otimes_{\min} A = R \otimes_{\max} A$ for every unital $C^*$-algebra $A$.
• $R \otimes_{\min} I = R \otimes_{\max} I$ for any injective operator system $I$.
• $R$ fails the weak expectation property (WEP).

Both Toeplitz and Fejér–Riesz systems are thus $C^*$-nuclear, have unique tensorings with injectives, and do not have the WEP (Corollaries 6.13, 6.14, 6.19) (Farenick, 2023).

7. Significance and Context in Operator System Theory

Operator-valued Toeplitz/Laurent coupling elucidates separability, entanglement, and the structure of tensor cones within operator systems. The duality and tensor-categorical correspondences provide tools for analyzing positive maps, complete positivity, and operator system nuclearity. The single-matrix criterion extends the operational logic of the Choi isomorphism to this broader noncommutative and function-theoretic context. A plausible implication is that these results support new approaches to bipartite entanglement beyond conventional matrix analysis, with relevance for quantum information and operator algebra theory (Farenick, 2023).