Physlib

PhyslibAlpha.CondensedMatter.TightBindingChain.Cube

The open tight binding cube

i. Overview

Three open tight binding chains `Tx`, `Ty`, `Tz` along the axes of a cube give the sites `(nx, ny, nz)` and the Hilbert space `𝓗[Fin Nx × Fin Ny × Fin Nz]`. Each observable of a chain acts on its own axis and leaves the other two in place. Observables of different axes commute: only the Hamiltonian and the position of the same axis fail to commute.

In the maximal current state of the cube, the product of the maximal current states of the three chains, each axis carries the constant `C_Nava` of its own chain. The energy–position uncertainty relation of an axis is an equality exactly when that axis has `2` or `3` sites, and it is strict from `4` sites on, on the three axes at once.

ii. Key results

- `alongX`, `alongY`, `alongZ` : an observable of a chain acting on its axis of the cube. - `bracket_alongX_alongY`, `bracket_alongX_alongZ`, `bracket_alongY_alongZ` : different axes commute. - `maxCurrentCubeState` : the maximal current state of the cube. - `CNava_alongX`, `CNava_alongY`, `CNava_alongZ` : each axis carries its own `C_Nava`. - `centeredGramDefect_alongX_eq_zero_iff` (and `Y`, `Z`) : saturation on an axis iff it has `2` or `3` sites. - `nava_robertson_schrodinger_cube` : from `4 × 4 × 4` on, strict on the three axes.

iii. Table of contents

  • A. Observables along the axes
  • B. Different axes commute
  • C. The maximal current state of the cube
  • D. The uncertainty relation along each axis

iv. References

* https://www.damtp.cam.ac.uk/user/tong/aqm/aqmtwo.pdf. [ref: tong_statistical_physics]

A. Observables along the axes

B. Different axes commute

C. The maximal current state of the cube

D. The uncertainty relation along each axis

44 declarations

abbrev

Hilbert space of the tight binding cube

Given three open tight-binding chains Tx,TyT_x, T_y, and TzT_z with Nx,NyN_x, N_y, and NzN_z sites respectively, the Hilbert space of the cube is the finite-dimensional complex Hilbert space H\mathcal{H} whose basis is indexed by the Cartesian product of the sites of each chain, denoted as Fin Nx×Fin Ny×Fin Nz\text{Fin } N_x \times \text{Fin } N_y \times \text{Fin } N_z. This space represents the quantum states of a system defined on the 3D grid of sites (nx,ny,nz)(n_x, n_y, n_z).

abbrev

Linear operators on the Hilbert space of the tight-binding cube

Given three open tight-binding chains TxT_x, TyT_y, and TzT_z, `CubeOperators` is the space of continuous linear operators acting on the Hilbert space of the cube Hcube\mathcal{H}_{\text{cube}}. Mathematically, it is the space of linear maps L(Hcube)\mathcal{L}(\mathcal{H}_{\text{cube}}) from the Hilbert space to itself over the complex numbers C\mathbb{C}, where Hcube\mathcal{H}_{\text{cube}} is the finite-dimensional Hilbert space representing the 3D grid of sites.

abbrev

Observable aa acting along the xx-axis of the cube

Given an observable aa (a self-adjoint operator) acting on the Hilbert space Hx\mathcal{H}_x of a tight-binding chain TxT_x, `alongX a` defines a corresponding observable on the Hilbert space of the cube Hcube≅Hx⊗Hy⊗Hz\mathcal{H}_{\text{cube}} \cong \mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. This observable acts as aa on the xx-axis and as the identity operator on the yy and zz axes, representing the tensor product operator a⊗Ia \otimes I.

abbrev

Observable bb acting along the yy-axis

Given an observable bb (a self-adjoint linear operator) acting on the Hilbert space Hy\mathcal{H}_y of a tight-binding chain TyT_y, `alongY b` is the corresponding observable on the cube Hilbert space Hcube≅Hx⊗Hy⊗Hz\mathcal{H}_{\text{cube}} \cong \mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. It represents the operator Ix⊗b⊗IzI_x \otimes b \otimes I_z, which acts as bb on the yy-axis and as the identity operator on the xx and zz axes.

abbrev

Observable along the zz-axis of the cube

Given an observable cc acting on the Hilbert space Hz\mathcal{H}_z of a tight-binding chain along the zz-axis, this definition constructs the corresponding observable on the 3D cube Hilbert space Hx⊗Hy⊗Hz\mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. Specifically, it represents the operator Ix⊗Iy⊗cI_x \otimes I_y \otimes c, where IxI_x and IyI_y are the identity operators on the xx and yy axes, respectively.

theorem

⟦alongX a,alongY b⟧=0\llbracket \text{alongX } a, \text{alongY } b \rrbracket = 0

In a tight-binding cube formed by three chains TxT_x, TyT_y, and TzT_z, let aa be an observable (a self-adjoint operator) acting on the Hilbert space Hx\mathcal{H}_x of the xx-axis chain, and let bb be an observable acting on the Hilbert space Hy\mathcal{H}_y of the yy-axis chain. Let alongX(a)\text{alongX}(a) and alongY(b)\text{alongY}(b) be the corresponding lifted observables acting on the cube's Hilbert space Hcube≅Hx⊗Hy⊗Hz\mathcal{H}_{\text{cube}} \cong \mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. Then the Lie bracket of these observables is zero: ⟦alongX(a),alongY(b)⟧=0, \llbracket \text{alongX}(a), \text{alongY}(b) \rrbracket = 0, where the Lie bracket for self-adjoint operators is defined as ⟦A,B⟧=−i2(AB−BA)\llbracket A, B \rrbracket = -\frac{i}{2}(AB - BA).

theorem

⟦alongX(a),alongZ(c)⟧=0\llbracket \text{alongX}(a), \text{alongZ}(c) \rrbracket = 0

Let Tx,Ty,T_x, T_y, and TzT_z be open tight-binding chains with corresponding Hilbert spaces Hx,Hy,\mathcal{H}_x, \mathcal{H}_y, and Hz\mathcal{H}_z. For any observable (self-adjoint operator) aa acting on Hx\mathcal{H}_x and any observable cc acting on Hz\mathcal{H}_z, their representations on the cube Hilbert space Hx⊗Hy⊗Hz\mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z, denoted by alongX(a)\text{alongX}(a) and alongZ(c)\text{alongZ}(c), commute such that their Lie bracket is zero: ⟦alongX(a),alongZ(c)⟧=0\llbracket \text{alongX}(a), \text{alongZ}(c) \rrbracket = 0 where the Lie bracket is defined as ⟦A^,B^⟧=−i2(A^B^−B^A^)\llbracket \hat{A}, \hat{B} \rrbracket = -\frac{i}{2}(\hat{A}\hat{B} - \hat{B}\hat{A}).

theorem

Observables along the yy and zz axes commute: ⟦alongY b,alongZ c⟧=0\llbracket \text{alongY } b, \text{alongZ } c \rrbracket = 0

Let Tx,Ty,TzT_x, T_y, T_z be three open tight-binding chains forming a 3D cube with the Hilbert space Hcube≅Hx⊗Hy⊗Hz\mathcal{H}_{\text{cube}} \cong \mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. For any observable bb acting on the Hilbert space Hy\mathcal{H}_y and any observable cc acting on the Hilbert space Hz\mathcal{H}_z, let by=Ix⊗b⊗Izb_y = I_x \otimes b \otimes I_z and cz=Ix⊗Iy⊗cc_z = I_x \otimes I_y \otimes c be the corresponding observables acting on the cube. Then their Lie bracket vanishes: ⟦by,cz⟧=0\llbracket b_y, c_z \rrbracket = 0 where the bracket is defined as ⟦A,B⟧=−i2(AB−BA)\llbracket A, B \rrbracket = -\frac{i}{2}(AB - BA).

definition

Maximal current state ∣Ψmax⟩|\Psi_{\text{max}}\rangle of the tight-binding cube

Given three open tight-binding chains Tx,TyT_x, T_y, and TzT_z with Nx,NyN_x, N_y, and NzN_z sites respectively, the maximal current state of the cube ∣Ψmax⟩|\Psi_{\text{max}}\rangle is the product vector of the 1D maximal current states ∣ψmax⟩|\psi_{\text{max}}\rangle of each individual chain. Specifically, ∣Ψmax⟩=∣ψmax,x⟩⊗(∣ψmax,y⟩⊗∣ψmax,z⟩)|\Psi_{\text{max}}\rangle = |\psi_{\text{max}, x}\rangle \otimes (|\psi_{\text{max}, y}\rangle \otimes |\psi_{\text{max}, z}\rangle), which resides in the cube Hilbert space H≅CNx×Ny×Nz\mathcal{H} \cong \mathbb{C}^{N_x \times N_y \times N_z}.

theorem

The maximal current state ∣Ψmax⟩|\Psi_{\text{max}}\rangle of the cube is normalized

Given three open tight-binding chains TxT_x, TyT_y, and TzT_z, the norm of the maximal current state of the cube ∣Ψmax⟩|\Psi_{\text{max}}\rangle is equal to 1, i.e., ∥∣Ψmax⟩∥=1\||\Psi_{\text{max}}\rangle\| = 1.

abbrev

Maximal current vector state of the tight-binding cube

Given three open tight-binding chains Tx,TyT_x, T_y, and TzT_z, the maximal current vector state is the quantum state (unital positive linear map) on the algebra of cube operators L(Hcube)\mathcal{L}(\mathcal{H}_{\text{cube}}) defined by the normalized maximal current state vector ∣Ψmax⟩|\Psi_{\text{max}}\rangle. For any operator A∈L(Hcube)A \in \mathcal{L}(\mathcal{H}_{\text{cube}}), the state maps AA to the expectation value ⟨Ψmax,AΨmax⟩\langle \Psi_{\text{max}}, A \Psi_{\text{max}} \rangle.

theorem

Variance of an xx-axis Observable in the Cube equals the 1D Chain Variance

Consider three open tight-binding chains TxT_x, TyT_y, and TzT_z. Let ωcube\omega_{\text{cube}} be the maximal current vector state of the 3D cube formed by these chains, and let ωx\omega_x be the maximal current vector state of the individual chain TxT_x. For any observable aa acting on the Hilbert space of TxT_x, let axa_x be the corresponding observable acting along the xx-axis of the cube (representing a⊗Iy⊗Iza \otimes I_y \otimes I_z). The variance of axa_x in the cube state ωcube\omega_{\text{cube}} is equal to the variance of aa in the chain state ωx\omega_x: Varωcube(ax)=Varωx(a)\text{Var}_{\omega_{\text{cube}}}(a_x) = \text{Var}_{\omega_x}(a)

theorem

The centered Gram defect along the xx-axis of the cube equals the chain's centered Gram defect

For three open tight-binding chains TxT_x, TyT_y, and TzT_z, let ωcube\omega_{\text{cube}} be the maximal current vector state on the Hilbert space of the cube Hcube\mathcal{H}_{\text{cube}}, and let ωx\omega_x be the maximal current vector state on the Hilbert space Hx\mathcal{H}_x of the chain TxT_x. For any two observables a,ba, b acting on Hx\mathcal{H}_x, let axa_x and bxb_x be the corresponding observables acting along the xx-axis of the cube (defined by the operators a⊗Iy⊗Iza \otimes I_y \otimes I_z and b⊗Iy⊗Izb \otimes I_y \otimes I_z respectively). The centered Gram defect of axa_x and bxb_x with respect to the state ωcube\omega_{\text{cube}} is equal to the centered Gram defect of aa and bb with respect to the state ωx\omega_x: centeredGramDefect(ωcube,ax,bx)=centeredGramDefect(ωx,a,b)\text{centeredGramDefect}(\omega_{\text{cube}}, a_x, b_x) = \text{centeredGramDefect}(\omega_x, a, b) where the centered Gram defect for a state ω\omega and observables u,vu, v is defined as Varω(u)⋅Varω(v)−∣ω(ΔuΔv)∣2\text{Var}_\omega(u) \cdot \text{Var}_\omega(v) - |\omega(\Delta u \Delta v)|^2.

theorem

Varωcube, max(by′)=Varωy,max(b′)\text{Var}_{\omega_{\text{cube, max}}}(b'_y) = \text{Var}_{\omega_{y, \text{max}}}(b')

Let Tx,TyT_x, T_y, and TzT_z be three open tight-binding chains. Let ωcube, max\omega_{\text{cube, max}} be the maximal current state on the Hilbert space of the cube Hx⊗Hy⊗Hz\mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z, and let ωy,max\omega_{y, \text{max}} be the maximal current state on the Hilbert space Hy\mathcal{H}_y of the chain TyT_y. For any observable b′b' acting on the chain TyT_y, let by′b'_y denote the corresponding observable acting along the yy-axis of the cube (representing Ix⊗b′⊗IzI_x \otimes b' \otimes I_z). Then the variance of by′b'_y in the state ωcube, max\omega_{\text{cube, max}} is equal to the variance of b′b' in the state ωy,max\omega_{y, \text{max}}: Varωcube, max(by′)=Varωy,max(b′)\text{Var}_{\omega_{\text{cube, max}}}(b'_y) = \text{Var}_{\omega_{y, \text{max}}}(b')

theorem

The Centered Gram Defect of yy-axis Observables in the Cube Equals the Defect in the Chain State

Consider three open tight-binding chains Tx,TyT_x, T_y, and TzT_z. Let Ωmax\Omega_{\text{max}} be the maximal current vector state of the tight-binding cube, and let ωmax,y\omega_{\text{max}, y} be the maximal current vector state of the yy-axis chain TyT_y. For any two observables b′b' and c′c' (self-adjoint linear operators) acting on the Hilbert space of TyT_y, let B=Ix⊗b′⊗IzB = I_x \otimes b' \otimes I_z and C=Ix⊗c′⊗IzC = I_x \otimes c' \otimes I_z be the corresponding lifted observables acting on the cube's Hilbert space. Then the centered Gram defect of BB and CC with respect to Ωmax\Omega_{\text{max}} is equal to the centered Gram defect of b′b' and c′c' with respect to ωmax,y\omega_{\text{max}, y}: GramDefectΩmax(B,C)=GramDefectωmax,y(b′,c′)\text{GramDefect}_{\Omega_{\text{max}}}(B, C) = \text{GramDefect}_{\omega_{\text{max}, y}}(b', c') where the centered Gram defect for a state ω\omega and observables a,ba, b is defined as the remainder in the Robertson–Schrödinger uncertainty relation: Varω(a)⋅Varω(b)−∣ω((a−ω⟨a⟩1)(b−ω⟨b⟩1))∣2\text{Var}_\omega(a) \cdot \text{Var}_\omega(b) - |\omega((a - \omega\langle a \rangle 1)(b - \omega\langle b \rangle 1))|^2.

theorem

The variance of a zz-axis observable in the cube's maximal current state equals its variance in the zz-chain's maximal current state.

For three open tight-binding chains Tx,TyT_x, T_y, and TzT_z, let Ω\Omega be the maximal current vector state of the 3D cube and ωz\omega_z be the maximal current vector state of the zz-axis chain TzT_z. For any observable b′′b'' acting on the Hilbert space of TzT_z, let bz′′b''_z be the corresponding observable acting along the zz-axis of the cube (defined as Ix⊗Iy⊗b′′I_x \otimes I_y \otimes b''). Then the variance of bz′′b''_z in the state Ω\Omega is equal to the variance of b′′b'' in the state ωz\omega_z: VarΩ(bz′′)=Varωz(b′′)\text{Var}_{\Omega}(b''_z) = \text{Var}_{\omega_z}(b'')

theorem

Centered Gram Defect along the zz-axis of the Cube equals the Centered Gram Defect of the zz-Chain

Let Tx,TyT_x, T_y, and TzT_z be three open tight-binding chains. Let ωcube\omega_{\text{cube}} be the maximal current vector state of the tight-binding cube, and let ωz\omega_z be the maximal current vector state of the chain TzT_z. For any two observables b,cb, c acting on the Hilbert space of the zz-axis chain, let BB and CC be the corresponding observables acting on the cube's Hilbert space along the zz-axis (representing Ix⊗Iy⊗bI_x \otimes I_y \otimes b and Ix⊗Iy⊗cI_x \otimes I_y \otimes c respectively). The centered Gram defect of BB and CC in the state ωcube\omega_{\text{cube}} is equal to the centered Gram defect of bb and cc in the state ωz\omega_z: centeredGramDefect(ωcube,B,C)=centeredGramDefect(ωz,b,c)\text{centeredGramDefect}(\omega_{\text{cube}}, B, C) = \text{centeredGramDefect}(\omega_z, b, c) where the centered Gram defect for two observables x,yx, y in a state ω\omega is defined as Varω(x)⋅Varω(y)−∣ω((x−ω(x)1)(y−ω(y)1))∣2\text{Var}_\omega(x) \cdot \text{Var}_\omega(y) - |\omega((x - \omega(x)1)(y - \omega(y)1))|^2.

theorem

The xx-axis of the Tight-Binding Cube carries the CNavaC_{\text{Nava}} Constant of its Chain

Consider a 3D tight-binding cube formed by three open chains Tx,TyT_x, T_y, and TzT_z with Nx,NyN_x, N_y, and NzN_z sites respectively. Let ωcube\omega_{\text{cube}} be the maximal current state of the cube. Let HxH_x be the Hamiltonian of the xx-axis chain acting on the cube (i.e., Hopen⊗I⊗IH_{\text{open}} \otimes I \otimes I) and XxX_x be the position operator of the xx-axis chain acting on the cube (i.e., X⊗I⊗IX \otimes I \otimes I). The following relationship holds: Varωcube(Hx)⋅Varωcube(Xx)∣axtxcos⁡(πNx+1)∣=CNava(Tx)\frac{\sqrt{\text{Var}_{\omega_{\text{cube}}}(H_x) \cdot \text{Var}_{\omega_{\text{cube}}}(X_x)}}{\left| a_x t_x \cos\left( \frac{\pi}{N_x + 1} \right) \right|} = C_{\text{Nava}}(T_x) where axa_x is the lattice spacing and txt_x is the hopping parameter of the chain TxT_x, and CNava(Tx)C_{\text{Nava}}(T_x) is the uncertainty constant inherent to the standalone xx-axis chain.

theorem

The yy-axis of the tight-binding cube carries the CNavaC_{\text{Nava}} constant.

Consider a tight-binding cube formed by three open chains Tx,TyT_x, T_y, and TzT_z. Let ωcube\omega_{\text{cube}} be the maximal current vector state of the cube. Let Hy,cubeH_{y, \text{cube}} and Xy,cubeX_{y, \text{cube}} denote the open Hamiltonian and the position operator of the yy-axis chain acting on the cube's Hilbert space (represented by `alongY Ty.openHamiltonianObservable` and `alongY Ty.positionObservable`). The following relationship holds: Varωcube(Hy,cube)⋅Varωcube(Xy,cube)∣aytycos⁡(πNy+1)∣=CNava(Ty)\frac{\sqrt{\text{Var}_{\omega_{\text{cube}}}(H_{y, \text{cube}}) \cdot \text{Var}_{\omega_{\text{cube}}}(X_{y, \text{cube}})}}{\left| a_y t_y \cos\left( \frac{\pi}{N_y + 1} \right) \right|} = C_{\text{Nava}}(T_y) where aya_y is the lattice spacing, tyt_y is the hopping parameter, NyN_y is the number of sites of the chain TyT_y, and CNava(Ty)C_{\text{Nava}}(T_y) is the uncertainty constant of the yy-axis chain.

theorem

The zz-axis of the tight-binding cube carries the constant CNavaC_{\text{Nava}} of its chain

For a system of three open tight-binding chains Tx,TyT_x, T_y, and TzT_z forming a 3D cube, let ωcube\omega_{\text{cube}} be the maximal current state of the cube. Let HzH_z and XzX_z be the observables corresponding to the open Hamiltonian and the position operator of the zz-axis chain acting on the cube's Hilbert space. Then the uncertainty relation for the zz-axis satisfies: Varωcube(Hz)⋅Varωcube(Xz)∣aztzcos⁡(πNz+1)∣=CNava,z\frac{\sqrt{\text{Var}_{\omega_{\text{cube}}}(H_z) \cdot \text{Var}_{\omega_{\text{cube}}}(X_z)}}{\left| a_z t_z \cos\left( \frac{\pi}{N_z+1} \right) \right|} = C_{\text{Nava}, z} where aza_z is the lattice spacing, tzt_z is the hopping parameter, NzN_z is the number of sites, and CNava,zC_{\text{Nava}, z} is the uncertainty constant of the independent zz-axis chain.

theorem

The centered Gram defect along the xx-axis vanishes iff Nx∈{2,3}N_x \in \{2, 3\}

Consider three open tight-binding chains Tx,Ty,T_x, T_y, and TzT_z forming a cubic lattice. Let NxN_x be the number of sites and tt be the hopping parameter for the chain along the xx-axis. Suppose t≠0t \neq 0 and Nx≥2N_x \geq 2. Let ω\omega be the maximal current vector state of the cube, and let HxH_x and XxX_x be the Hamiltonian and position observables acting along the xx-axis, respectively. The centered Gram defect between HxH_x and XxX_x in state ω\omega vanishes (meaning the Robertson–Schrödinger uncertainty relation becomes an equality) if and only if Nx=2N_x = 2 or Nx=3N_x = 3.

theorem

Saturation of the yy-axis Uncertainty Relation iff Ny=2N_y = 2 or Ny=3N_y = 3

Consider a tight-binding cube formed by three chains Tx,TyT_x, T_y, and TzT_z. Let NyN_y be the number of sites and tyt_y be the hopping parameter of the yy-axis chain TyT_y. Let ωcube\omega_{\text{cube}} be the maximal current vector state of the cube. Let HyH_y and XyX_y be the open Hamiltonian and the position operator acting along the yy-axis of the cube, respectively. For ty≠0t_y \neq 0 and Ny≥2N_y \geq 2, the centered Gram defect of HyH_y and XyX_y with respect to the state ωcube\omega_{\text{cube}} is zero if and only if Ny=2N_y = 2 or Ny=3N_y = 3: centeredGramDefectωcube(Hy,Xy)=0  ⟺  Ny=2 or Ny=3\text{centeredGramDefect}_{\omega_{\text{cube}}}(H_y, X_y) = 0 \iff N_y = 2 \text{ or } N_y = 3

theorem

Saturation of the Robertson–Schrödinger Uncertainty Relation along the zz-axis iff Nz∈{2,3}N_z \in \{2, 3\}

Let TxT_x, TyT_y, and TzT_z be three open tight-binding chains forming a 3D cube. Let NzN_z be the number of sites and tzt_z be the hopping parameter of the zz-axis chain. Suppose tz≠0t_z \neq 0 and Nz≥2N_z \geq 2. Let ω\omega be the maximal current vector state of the cube, and let HzH_z and XzX_z be the open Hamiltonian and position observables acting along the zz-axis, respectively. Then, the centered Gram defect of HzH_z and XzX_z in the state ω\omega is zero if and only if NzN_z is 2 or 3: Varω(Hz)⋅Varω(Xz)−∣ω(ΔHzΔXz)∣2=0  ⟺  Nz=2 or Nz=3\text{Var}_\omega(H_z) \cdot \text{Var}_\omega(X_z) - |\omega(\Delta H_z \Delta X_z)|^2 = 0 \iff N_z = 2 \text{ or } N_z = 3 where ΔHz=Hz−ω(Hz)I\Delta H_z = H_z - \omega(H_z)I and ΔXz=Xz−ω(Xz)I\Delta X_z = X_z - \omega(X_z)I are the centered observables.

definition

Conjugate pair (H,X)(H, X) of a tight-binding chain TT

For a tight-binding chain TT with associated Hilbert space H≅CN\mathcal{H} \cong \mathbb{C}^N, the **conjugate pair** is a function mapping the index set {0,1}\{0, 1\} to the space of observables (self-adjoint operators) in L(H)\mathcal{L}(\mathcal{H}). This pair consists of the open Hamiltonian HH and the position operator XX of the chain.

definition

Conjugate pairs (H,X)(H, X) for the axes of the cube

This function maps a pair of indices (j,k)∈{0,1}×{0,1,2}(j, k) \in \{0, 1\} \times \{0, 1, 2\} to an observable (a self-adjoint operator) acting on the Hilbert space of the tight-binding cube Hcube\mathcal{H}_{\text{cube}}. The index kk specifies the axis of the cube (x,y,x, y, or zz), and the index jj selects the corresponding member of the conjugate pair (H,X)(H, X)—the Hamiltonian and the position operator—of that axis. Each resulting observable acts on its specific axis while acting as the identity operator on the other two axes.

definition

Expectation value ω⟨⟦H,X⟧⟩\omega\langle \llbracket H, X \rrbracket \rangle for axis kk

Given a quantum state ω\omega on the algebra of operators of a tight-binding cube (formed by three chains Tx,TyT_x, T_y, and TzT_z) and an axis index k∈{0,1,2}k \in \{0, 1, 2\}, this definition calculates the expectation value ω⟨⟦Hk,Xk⟧⟩\omega\langle \llbracket H_k, X_k \rrbracket \rangle of the Lie bracket of the Hamiltonian HkH_k and position operator XkX_k corresponding to the kk-th axis. The bracket is defined as ⟦A,B⟧=−i2(AB−BA)\llbracket A, B \rrbracket = -\frac{i}{2}(AB - BA), which ensures that the commutator of two self-adjoint operators results in another self-adjoint operator (an observable).

theorem

The xx-axis conjugate pairs of the cube are the lifted conjugate pairs of the xx-axis chain.

For three open tight-binding chains Tx,Ty,T_x, T_y, and TzT_z forming a cube, and for each index j∈{0,1}j \in \{0, 1\} representing either the Hamiltonian (j=0j=0) or the position operator (j=1j=1), the jj-th member of the conjugate pair for the xx-axis of the cube is equal to the jj-th conjugate pair of the chain TxT_x acting along the xx-axis of the cube. Mathematically, for j∈{0,1}j \in \{0, 1\}: cubePairs(j,0)=alongX(Tx.conjugatePair(j))\text{cubePairs}(j, 0) = \text{alongX}(T_x.\text{conjugatePair}(j)) where the axis index 00 denotes the xx-axis, and alongX\text{alongX} lifts an operator from the xx-chain's Hilbert space to the cube's Hilbert space by acting as the identity on the yy and zz axes.

theorem

The yy-axis conjugate pairs of the cube are the TyT_y conjugate pairs acting along the yy-direction

For three open tight-binding chains Tx,TyT_x, T_y, and TzT_z, let cubePairs(j,k)\text{cubePairs}(j, k) be the jj-th observable in the conjugate pair (where j=0j=0 denotes the Hamiltonian and j=1j=1 denotes the position operator) for the kk-th axis of the cube. For j∈{0,1}j \in \{0, 1\}, the jj-th conjugate operator associated with the yy-axis (k=1k=1) is the jj-th conjugate operator of the chain TyT_y acting along the yy-direction of the cube: cubePairs(j,1)=alongY(conjugatePairTy(j))\text{cubePairs}(j, 1) = \text{alongY}(\text{conjugatePair}_{T_y}(j)) where alongY\text{alongY} is the operator that acts as its argument on the yy-axis Hilbert space and as the identity on the xx and zz axes.

theorem

The zz-axis conjugate operators of the cube cubePairs(j,2)\text{cubePairs}(j, 2) equal alongZ(conjugatePairTz(j))\text{alongZ}(\text{conjugatePair}_{T_z}(j))

For a tight-binding cube composed of three chains Tx,TyT_x, T_y, and TzT_z, let cubePairs(j,k)\text{cubePairs}(j, k) be the operator on the cube's Hilbert space corresponding to the jj-th conjugate variable (where j=0j=0 is the Hamiltonian and j=1j=1 is the position operator) along the kk-th axis (where k=2k=2 denotes the zz-axis). For any j∈{0,1}j \in \{0, 1\}, it holds that cubePairs(j,2)=alongZ(conjugatePairTz(j))\text{cubePairs}(j, 2) = \text{alongZ}(\text{conjugatePair}_{T_z}(j)) where conjugatePairTz(j)\text{conjugatePair}_{T_z}(j) is the jj-th conjugate operator of the individual chain TzT_z, and alongZ\text{alongZ} is the operation that embeds an operator cc acting on the zz-axis Hilbert space into the full cube Hilbert space Hx⊗Hy⊗Hz\mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z as the tensor product Ix⊗Iy⊗cI_x \otimes I_y \otimes c.

theorem

Observables on different axes of the cube commute

Let Tx,TyT_x, T_y, and TzT_z be three open tight-binding chains. Let pp and qq be indices in {0,1}×{0,1,2}\{0, 1\} \times \{0, 1, 2\} that specify an observable type (Hamiltonian HH or position XX) and an axis of the cube (x,yx, y, or zz). If pp and qq refer to different axes (i.e., p2≠q2p_2 \neq q_2), then the Lie bracket of the corresponding observables on the cube's Hilbert space is zero: ⟦cubePairs(p),cubePairs(q)⟧=0\llbracket \text{cubePairs}(p), \text{cubePairs}(q) \rrbracket = 0 This implies that observables associated with different spatial axes commute.

theorem

The Bracket Matrix of Tight-Binding Cube Observables is Block Diagonal

For any quantum state ω\omega on the operator algebra of a tight-binding cube (composed of three chains Tx,Ty,TzT_x, T_y, T_z), the 6×66 \times 6 bracket matrix Ω\Omega of the conjugate pairs of observables (Hk,Xk)(H_k, X_k) for each axis k∈{x,y,z}k \in \{x, y, z\} is block diagonal. Specifically, the matrix consists of three 2×22 \times 2 antisymmetric blocks: Ω=diag((0ω⟨⟦Hx,Xx⟧⟩−ω⟨⟦Hx,Xx⟧⟩0),(0ω⟨⟦Hy,Xy⟧⟩−ω⟨⟦Hy,Xy⟧⟩0),(0ω⟨⟦Hz,Xz⟧⟩−ω⟨⟦Hz,Xz⟧⟩0))\Omega = \text{diag}\left( \begin{pmatrix} 0 & \omega\langle \llbracket H_x, X_x \rrbracket \rangle \\ -\omega\langle \llbracket H_x, X_x \rrbracket \rangle & 0 \end{pmatrix}, \begin{pmatrix} 0 & \omega\langle \llbracket H_y, X_y \rrbracket \rangle \\ -\omega\langle \llbracket H_y, X_y \rrbracket \rangle & 0 \end{pmatrix}, \begin{pmatrix} 0 & \omega\langle \llbracket H_z, X_z \rrbracket \rangle \\ -\omega\langle \llbracket H_z, X_z \rrbracket \rangle & 0 \end{pmatrix} \right) where ⟦A,B⟧=−i2(AB−BA)\llbracket A, B \rrbracket = -\frac{i}{2}(AB - BA) is the Lie bracket, and ω⟨⟦Hk,Xk⟧⟩\omega\langle \llbracket H_k, X_k \rrbracket \rangle is the expectation value of the bracket of the Hamiltonian and position operator along axis kk.

theorem

Robertson's Determinant Inequality for the Tight-Binding Cube: det⁡Σ≥(∏ω⟨⟦Hk,Xk⟧⟩)2\det \Sigma \ge (\prod \omega\langle \llbracket H_k, X_k \rrbracket \rangle)^2

Consider a quantum system on a tight-binding cube formed by three chains Tx,TyT_x, T_y, and TzT_z. For any quantum state ω\omega, let HkH_k and XkX_k be the Hamiltonian and position observables corresponding to the kk-th axis for k∈{0,1,2}k \in \{0, 1, 2\}. Let Σ\Sigma be the 6×66 \times 6 covariance matrix of these six observables in the state ω\omega. Then the determinant of the covariance matrix is bounded below by the squared product of the expectation values of the brackets ⟦Hk,Xk⟧=−i2(HkXk−XkHk)\llbracket H_k, X_k \rrbracket = -\frac{i}{2}(H_k X_k - X_k H_k): (∏k=02ω⟨⟦Hk,Xk⟧⟩)2≤det⁡Σ\left( \prod_{k=0}^2 \omega\langle \llbracket H_k, X_k \rrbracket \rangle \right)^2 \le \det \Sigma

theorem

CovΩmax(aX,bX)=Covωmaxx(a,b)\text{Cov}_{\Omega_{\text{max}}}(a_X, b_X) = \text{Cov}_{\omega_{\text{max}}^x}(a, b)

Consider three open tight-binding chains Tx,TyT_x, T_y, and TzT_z. Let Ωmax\Omega_{\text{max}} be the maximal current vector state of the tight-binding cube, and let ωmaxx\omega_{\text{max}}^x be the maximal current vector state of the chain TxT_x. For any two observables aa and bb acting on the Hilbert space of TxT_x, let aXa_X and bXb_X be the corresponding observables acting along the xx-axis of the cube (i.e., acting as aa and bb on the xx-axis and as the identity on the yy and zz axes). Then the covariance of aXa_X and bXb_X in the state Ωmax\Omega_{\text{max}} is equal to the covariance of aa and bb in the state ωmaxx\omega_{\text{max}}^x: CovΩmax(aX,bX)=Covωmaxx(a,b)\text{Cov}_{\Omega_{\text{max}}}(a_X, b_X) = \text{Cov}_{\omega_{\text{max}}^x}(a, b)

theorem

CovΩmax(b′(y),c′(y))=Covωy(b′,c′)\text{Cov}_{\Omega_{\text{max}}}(b'^{(y)}, c'^{(y)}) = \text{Cov}_{\omega_y}(b', c')

Consider a system of three open tight-binding chains Tx,TyT_x, T_y, and TzT_z forming a 3D cube. Let Ωmax\Omega_{\text{max}} be the maximal current vector state of the cube, and let ωy\omega_y be the maximal current vector state of the individual chain TyT_y. For any two observables b′b' and c′c' acting on the Hilbert space of the chain TyT_y, let b′(y)b'^{(y)} and c′(y)c'^{(y)} be the corresponding observables acting along the yy-axis of the cube (defined as Ix⊗b′⊗IzI_x \otimes b' \otimes I_z and Ix⊗c′⊗IzI_x \otimes c' \otimes I_z). Then the covariance of these yy-axis observables in the cube state Ωmax\Omega_{\text{max}} is equal to their covariance in the chain state ωy\omega_y: CovΩmax(b′(y),c′(y))=Covωy(b′,c′)\text{Cov}_{\Omega_{\text{max}}}(b'^{(y)}, c'^{(y)}) = \text{Cov}_{\omega_y}(b', c')

theorem

The covariance of zz-axis observables in the cube state equals their 1D covariance.

Let Tx,TyT_x, T_y, and TzT_z be three open tight-binding chains. Let Ωcube\Omega_{\text{cube}} be the maximal current vector state of the 3D cube formed by these chains. For any two observables b′′b'' and c′′c'' acting on the Hilbert space of the zz-axis chain TzT_z, let balongZ′′b''_{\text{alongZ}} and calongZ′′c''_{\text{alongZ}} denote their corresponding lifted observables on the cube's Hilbert space (representing Ix⊗Iy⊗b′′I_x \otimes I_y \otimes b'' and Ix⊗Iy⊗c′′I_x \otimes I_y \otimes c'' respectively). Then the covariance of these observables in the cube state Ωcube\Omega_{\text{cube}} is equal to the covariance of the original observables in the 1D maximal current vector state ωz\omega_{z} of the zz-axis chain: CovΩcube(balongZ′′,calongZ′′)=Covωz(b′′,c′′)\text{Cov}_{\Omega_{\text{cube}}}(b''_{\text{alongZ}}, c''_{\text{alongZ}}) = \text{Cov}_{\omega_{z}}(b'', c'')

theorem

Zero Covariance between xx and yy Observables in the Maximal Current Cube State

Let Tx,TyT_x, T_y, and TzT_z be three open tight-binding chains forming a cube. Let ω\omega be the maximal current vector state of this cube. For any observable aa acting on the chain TxT_x and any observable bb acting on the chain TyT_y, let AA and BB be the corresponding observables acting along the xx and yy axes of the cube (representing a⊗Iy⊗Iza \otimes I_y \otimes I_z and Ix⊗b⊗IzI_x \otimes b \otimes I_z respectively). Then the covariance of AA and BB in the state ω\omega is zero: Covω(A,B)=0\text{Cov}_\omega(A, B) = 0 This implies that in the maximal current state of the cube, the xx and yy axes are uncorrelated.

theorem

Zero Covariance between xx and zz Observables in the Maximal Current Cube State

In the maximal current state ω\omega of a tight-binding cube, the covariance between an observable aa acting along the xx-axis and an observable b′′b'' acting along the zz-axis is zero: Covω(alongX a,alongZ b′′)=0\text{Cov}_\omega(\text{alongX } a, \text{alongZ } b'') = 0 This implies that, in this state, measurements along the xx and zz axes are uncorrelated.

theorem

Covω(By,Cz)=0\text{Cov}_\omega(B_y, C_z) = 0 in the maximal current state of the cube

Let Tx,Ty,T_x, T_y, and TzT_z be three open tight-binding chains forming a cube with Hilbert space Hcube≅Hx⊗Hy⊗Hz\mathcal{H}_{\text{cube}} \cong \mathcal{H}_x \otimes \mathcal{H}_y \otimes \mathcal{H}_z. Let ω\omega be the maximal current vector state of the cube. For any observable b′b' acting on the yy-axis chain and any observable b′′b'' acting on the zz-axis chain, let ByB_y and CzC_z be the corresponding observables acting along the yy and zz axes of the cube, respectively. Then the covariance of these observables in the state ω\omega is zero: Covω(By,Cz)=0\text{Cov}_\omega(B_y, C_z) = 0 This indicates that the yy and zz axes are uncorrelated in the maximal current state.

theorem

The covariance matrix of the maximal current state of the tight-binding cube is block diagonal

For three open tight-binding chains Tx,TyT_x, T_y, and TzT_z forming a 3D cube, let ωcube\omega_{\text{cube}} be the maximal current state of the system. Let the set of six observables be the conjugate pairs (Hk,Xk)(H_k, X_k) for each axis k∈{x,y,z}k \in \{x, y, z\}, where HkH_k is the Hamiltonian and XkX_k is the position operator along that axis. Then the 6×66 \times 6 covariance matrix Σ\Sigma of these observables in the state ωcube\omega_{\text{cube}} is block diagonal: Σ=(Σx000Σy000Σz)\Sigma = \begin{pmatrix} \Sigma_x & 0 & 0 \\ 0 & \Sigma_y & 0 \\ 0 & 0 & \Sigma_z \end{pmatrix} where each Σk\Sigma_k is the 2×22 \times 2 covariance matrix of the conjugate pair (H,X)(H, X) of the individual chain TkT_k evaluated in its own maximal current state ωk\omega_k.

theorem

Determinant of the Covariance Matrix of (H,X)(H, X) in the Maximal Current State Equals CNava2(atcos⁡πN+1)2C_{\text{Nava}}^2 (a t \cos \frac{\pi}{N+1})^2

For an open tight-binding chain TT with NN sites, lattice spacing aa, and hopping parameter t≠0t \neq 0, let N≥2N \geq 2. Let Σ\Sigma be the covariance matrix of the conjugate pair (H,X)(H, X), consisting of the Hamiltonian and the position operator, evaluated in the maximal current state ωmax\omega_{\text{max}}. The determinant of Σ\Sigma is given by det⁡Σ=CNava2(atcos⁡πN+1)2\det \Sigma = C_{\text{Nava}}^2 \left( a t \cos \frac{\pi}{N + 1} \right)^2 where CNavaC_{\text{Nava}} is the uncertainty constant of the chain.

theorem

Square of the Product of Axis Brackets in the Maximal Current State of a Cube Equals the Square of the Product of atcos⁡(π/(N+1))a t \cos(\pi / (N + 1)) terms.

Consider a 3D tight-binding cube formed by three open chains Tx,TyT_x, T_y, and TzT_z. Let ω\omega be the maximal current vector state of the cube. For each axis k∈{x,y,z}k \in \{x, y, z\}, let aka_k, tkt_k, and NkN_k denote the lattice spacing, hopping parameter, and number of sites of the corresponding chain. Let Bk=ω⟨⟦Hk,Xk⟧⟩B_k = \omega\langle \llbracket H_k, X_k \rrbracket \rangle be the expectation value of the bracket of the Hamiltonian HkH_k and position operator XkX_k for the kk-th axis, defined as ⟦Hk,Xk⟧=−i2(HkXk−XkHk)\llbracket H_k, X_k \rrbracket = -\frac{i}{2}(H_k X_k - X_k H_k). Then the square of the product of these expectation values is given by: (Bx⋅By⋅Bz)2=(axtxcos⁡πNx+1⋅aytycos⁡πNy+1⋅aztzcos⁡πNz+1)2(B_x \cdot B_y \cdot B_z)^2 = \left( a_x t_x \cos \frac{\pi}{N_x + 1} \cdot a_y t_y \cos \frac{\pi}{N_y + 1} \cdot a_z t_z \cos \frac{\pi}{N_z + 1} \right)^2

theorem

Determinant of the Covariance Matrix for the Maximal Current State of a Tight-Binding Cube equals (CNava,xCNava,yCNava,z)2∣det⁡Ω∣(C_{\text{Nava}, x} C_{\text{Nava}, y} C_{\text{Nava}, z})^2 |\det \Omega|

Consider a 3D tight-binding cube formed by three open chains Tx,Ty,TzT_x, T_y, T_z with hopping parameters tx,ty,tz≠0t_x, t_y, t_z \neq 0 and at least two sites along each axis (Nx,Ny,Nz≥2N_x, N_y, N_z \ge 2). Let ω\omega be the maximal current vector state of the cube, and let Σ\Sigma be the 6×66 \times 6 covariance matrix of the observables (Hx,Xx,Hy,Xy,Hz,Xz)(H_x, X_x, H_y, X_y, H_z, X_z) in the state ω\omega. Then the determinant of the covariance matrix is given by det⁡Σ=(CNava,xCNava,yCNava,z)2(∏k∈{x,y,z}ω⟨⟦Hk,Xk⟧⟩)2\det \Sigma = (C_{\text{Nava}, x} C_{\text{Nava}, y} C_{\text{Nava}, z})^2 \left( \prod_{k \in \{x, y, z\}} \omega\langle \llbracket H_k, X_k \rrbracket \rangle \right)^2 where CNava,kC_{\text{Nava}, k} is the uncertainty constant of the kk-th chain and ω⟨⟦Hk,Xk⟧⟩\omega\langle \llbracket H_k, X_k \rrbracket \rangle is the expectation value of the Lie bracket ⟦Hk,Xk⟧=−i2[Hk,Xk]\llbracket H_k, X_k \rrbracket = -\frac{i}{2}[H_k, X_k] for the kk-th axis.

theorem

Strict Robertson inequality ∣det⁡Ω∣<det⁡Σ|\det \Omega| < \det \Sigma for the maximal current state of a tight-binding cube (N≥4N \ge 4)

Consider a 3D tight-binding cube formed by three open chains Tx,Ty,TzT_x, T_y, T_z with hopping parameters tx,ty,tz≠0t_x, t_y, t_z \neq 0 and Nx,Ny,NzN_x, N_y, N_z sites, respectively. Let ω\omega be the maximal current vector state of the cube. Let {aj}j=16\{a_j\}_{j=1}^6 be the set of six observables representing the Hamiltonian and position operators for each axis, i.e., (Hx,Xx,Hy,Xy,Hz,Xz)(H_x, X_x, H_y, X_y, H_z, X_z). If the number of sites along each axis is at least four (Nx,Ny,Nz≥4N_x, N_y, N_z \geq 4), then Robertson's uncertainty relation is strict: ∣det⁡Ω∣<det⁡Σ|\det \Omega| < \det \Sigma where Ω\Omega is the 6×66 \times 6 bracket matrix with entries Ωjk=ω(−i2[aj,ak])\Omega_{jk} = \omega( - \frac{i}{2} [a_j, a_k] ) and Σ\Sigma is the 6×66 \times 6 covariance matrix with entries Σjk=Covω(aj,ak)\Sigma_{jk} = \text{Cov}_\omega(a_j, a_k).