ERRP: Entanglement Roto-Reflection Plane

• ERRP is the geometric object capturing directional two-qubit entanglement via the Pauli correlation tensor, complementing the scalar measure of concurrence.
• It separates entanglement into a scalar magnitude (given by concurrence) and a geometric orientation provided by an improper orthogonal map (roto-reflection).
• Its covariant structure under local unitaries ensures the plane’s orientation may shift, yet the invariant orientation-reversing signature (determinant -1) remains unchanged.

The Entanglement Roto-Reflection Plane (ERRP) is the geometric object that captures the directional form of pure two-qubit entanglement in the Pauli correlation tensor, complementary to the scalar entanglement measure given by concurrence. In "Roto-Reflection Geometry of Pure Two-Qubit Entanglement" (Filatov et al., 10 Jun 2026), maximally entangled states appear as improper orthogonal maps between two local Bloch spheres, and these maps are roto-reflections; for partially entangled pure states, the same roto-reflection geometry is recovered after separating the contraction associated with concurrence. In this formulation, concurrence quantifies the magnitude of entanglement, while the ERRP specifies how the entanglement is oriented in Bloch-space geometry.

1. Pauli correlation tensor and directional pairing

Any two-qubit state is expanded in the Pauli basis as

$$\rho=\frac{1}{4}\sum_{\mu,\nu=0}^3 R_{\mu\nu}\,\sigma_\mu\otimes\sigma_\nu, \qquad R_{\mu\nu}=\mathrm{Tr}\!\left(\rho\,\sigma_\mu\otimes\sigma_\nu\right),$$

with $\sigma_0=I$. The coefficient matrix is written as

$$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$

where $\mathbf a$ and $\mathbf b$ are the reduced Bloch vectors of the two qubits, and

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$

is the $3\times 3$ Pauli correlation tensor. For a pure state $|\psi\rangle$, this reduces to

$$T_{ij}=\langle\psi|\sigma_i\otimes\sigma_j|\psi\rangle.$$

The operational content of $T$ is that it organizes joint spin correlations for local measurement directions $\sigma_0=I$0 through

$\sigma_0=I$1

Accordingly, $\sigma_0=I$2 acts as a direction-pairing map between the two local Bloch spheres. The ERRP is defined within this tensorial picture: it is not an additional observable outside the Pauli decomposition, but the geometric structure extracted from the orientation-reversing part of $\sigma_0=I$3 itself (Filatov et al., 10 Jun 2026).

2. Maximally entangled states as improper orthogonal maps

For the Bell state

$\sigma_0=I$4

the correlation tensor is

$\sigma_0=I$5

The paper denotes this canonical matrix by

$\sigma_0=I$6

It satisfies

$\sigma_0=I$7

so it is an improper orthogonal transformation. Geometrically, $\sigma_0=I$8 is reflection in the $\sigma_0=I$9-plane.

More generally, every maximally entangled pure two-qubit state is locally equivalent to $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$0, so its tensor has the form

$$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$1

with $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$2 the local rotations induced by local unitaries. Since the outer factors are proper rotations,

$$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$3

Thus every maximally entangled pure state defines an improper orthogonal map between the two Bloch spheres. In three dimensions, any improper orthogonal map can be viewed as a roto-reflection: reflection in a plane followed by a rotation about the normal to that plane (Filatov et al., 10 Jun 2026).

For a maximally entangled state with $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$4, the ERRP is the plane perpendicular to the real eigendirection $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$5 satisfying

$$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$6

The vector $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$7 is therefore the normal to the reflection plane. The associated rotation angle $$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$8 is determined by

$$R= \begin{pmatrix} 1 & \mathbf b^T \ \mathbf a & T \end{pmatrix},$$9

The pair $\mathbf a$0, up to the equivalence

$\mathbf a$1

encodes the roto-reflection geometry.

A special limiting case occurs for $\mathbf a$2. Then every direction is an eigenvector with eigenvalue $\mathbf a$3, so the ERRP plane is not unique. The paper interprets this as the maximally symmetric limiting case: point inversion, equivalent to reflection in any plane followed by a $\mathbf a$4-rotation. This non-uniqueness is intrinsic to the symmetry of the state rather than a failure of the construction.

3. Partially entangled pure states and concurrence-dependent contraction

The same geometric structure persists away from maximal entanglement. For the Schmidt-form pure state

$\mathbf a$5

the concurrence is

$\mathbf a$6

The reduced Bloch vectors lie along $\mathbf a$7 and have magnitude $\mathbf a$8. The corresponding correlation tensor is

$\mathbf a$9

This tensor is not orthogonal unless $\mathbf b$0, but it factorizes as

$\mathbf b$1

Here $\mathbf b$2 is the same improper orthogonal reflection as in the maximally entangled case, while $\mathbf b$3 is a positive contraction that shrinks the plane transverse to the reduced Bloch-vector direction by the concurrence $\mathbf b$4, while leaving the longitudinal direction unchanged. The ERRP is therefore retained after removing the contraction associated with partial entanglement (Filatov et al., 10 Jun 2026).

For a general pure state, local rotations give

$\mathbf b$5

The paper writes this as a polar-like decomposition

$\mathbf b$6

with

$\mathbf b$7

where

$\mathbf b$8

It also presents the usual Euclidean polar decomposition

$\mathbf b$9

This decomposition yields the paper’s central separation: pure two-qubit entanglement consists of a scalar magnitude, given by concurrence $$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$0, together with a geometric orientation-reversing structure, given by the ERRP.

4. Geometric interpretation of the ERRP

The ERRP is the plane of reflection hidden inside the correlation tensor of a pure entangled state. It organizes the maximally correlated directions: the directions on one Bloch sphere that are mapped to the corresponding strongest-correlation directions on the other sphere. In the maximally entangled case, the tensor itself is already the full improper orthogonal map, so the ERRP is directly visible. In the partially entangled case, the same map survives after the concurrence-dependent contraction is factored out.

The geometric content can be visualized in two complementary ways described in the paper. In the triad picture, three paired directions are shown at once. In the single-arrow picture, one direction on Bob’s sphere is mapped by $$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$1 to the maximally correlated direction on Alice’s sphere. The ERRP is the structure that explains how these directions are arranged and why the correlation map has an orientation-reversing form (Filatov et al., 10 Jun 2026).

This geometric role makes the ERRP a covariant geometric complement to concurrence. Concurrence measures “how much” entanglement there is, while the ERRP tells “how it is oriented” in Bloch-space geometry. The construction therefore supplements, rather than replaces, scalar entanglement quantifiers.

5. Covariance, determinant, and invariant content

A key property of the ERRP is covariance under local unitaries. If

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$2

then the orthogonal part transforms as

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$3

The particular plane and angle therefore depend on the chosen local Bloch frames, but the invariant signature

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$4

does not. In this sense the ERRP is basis-covariant rather than absolute: local rotations move the plane, but do not destroy the roto-reflection structure.

For pure two-qubit states, the singular values of $$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$5 are

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$6

so

$$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$7

This yields a clean division of roles. The determinant of $$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$8 gives the scalar entanglement magnitude, while the improper orthogonal factor $$T_{ij}=\mathrm{Tr}\!\left(\rho\,\sigma_i\otimes\sigma_j\right), \qquad i,j\in\{x,y,z\},$$9 gives the geometric form of entanglement (Filatov et al., 10 Jun 2026).

Two common misreadings are thereby excluded by the formalism itself. First, the ERRP is not a scalar entanglement monotone; concurrence retains that role. Second, the ERRP is not a basis-independent plane embedded once and for all in physical space; its content is covariant under local frame changes, with the invariant feature being the orientation-reversing character of the orthogonal factor.

6. Scope of the concept and relation to adjacent geometries

The ERRP is defined for pure two-qubit entanglement through the Pauli correlation tensor. Two nearby literatures are conceptually adjacent but distinct. "ER = EPR is an operational theorem" (Fields et al., 2024) does not explicitly mention ERRP or any phrase like “Entanglement Roto-Reflection Plane.” What it provides instead is a two-agent LOCC formulation of ER = EPR, an observer-relative local topology result, and a purely topological identification between the endpoints of a quantum channel and a wormhole-like Einstein-Rosen bridge. The paper explicitly states that no embedding geometry, no plane structure, and no topological object with that name appears there. The connection to ERRP is therefore only indirect and conceptual.

A different geometric adjacency appears in "Geometry of entanglement witnesses for two qutrits" (Chruściński et al., 2011). That work also does not use the term ERRP, but it studies a two-qutrit family of positive maps and entanglement witnesses whose boundary is parameterized by proper rotations in $3\times 3$0, together with a second family built from improper rotations in $3\times 3$1. The paper itself states that if one interprets “ERRP” as a plane whose structure is organized by rotations and roto-reflections, then its geometry is essentially an ERRP-like picture; however, its actual result concerns the geometry of witnesses and positive maps, not the pure two-qubit Pauli correlation tensor.

A plausible implication is that orientation-reversing structures recur across several geometric descriptions of entanglement-related phenomena, but the objects involved are not interchangeable. In the strict sense established in (Filatov et al., 10 Jun 2026), the ERRP is the reflection plane underlying the improper orthogonal part of the pure two-qubit Pauli correlation tensor: maximally entangled states realize it as an exact roto-reflection between local Bloch spheres, and partially entangled pure states realize the same geometry after the concurrence-dependent contraction is factored out.