Lean, informalized

Nielsen & Chuang · Chapter 7, Quantum computers: physical realization · p. 328

Exercise 7.35 - The spherical average of the dipolar coupling of two spins is the zero matrix

The dipolar coupling averages to zero over all directions.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Exercise

Two particles of spin 1/21/2, with gyromagnetic ratios γ1\gamma_1 and γ2\gamma_2, sit a distance rr apart, and the unit vector nˆ\hat n points along the line from one to the other. Their magnetic dipoles interact through the dipolar coupling H1,2DH^D_{1,2}, an operator on the pair of spins that depends on nˆ\hat n. Exercise 7.35 asks to show that this coupling vanishes on average when the direction nˆ\hat n ranges over the whole sphere, every direction counted with equal weight.

The library states the exercise for the coupling written out in Pauli matrices, with the average over directions as an integral over the polar and azimuthal angles of nˆ\hat n.

Two Spins

The Pauli matrices XX, YY and ZZ are the complex 2×22\times 2 matrices X=(0110),Y=(0−ii0),Z=(100−1),X = \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}, \qquad Y = \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix}, \qquad Z = \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix},(1) also written σ1=X\sigma_1 = X, σ2=Y\sigma_2 = Y, σ3=Z\sigma_3 = Z, the three components of the Pauli vector σ→\vec\sigma. For a real vector n→=(n1,n2,n3)\vec n = (n_1, n_2, n_3) in three dimensions, n→⋅σ→\vec n \cdot \vec\sigma is the complex 2×22\times 2 matrix n→⋅σ→=n1X+n2Y+n3Z.\vec n \cdot \vec\sigma = n_1 X + n_2 Y + n_3 Z.(2)✦

The two spins are two qubits, and an operator on the pair is a complex 4×44\times 4 matrix acting on ℂ2⊗ℂ2\mathbb{C}^2 \otimes \mathbb{C}^2. In a tensor product A⊗BA \otimes B of 2×22\times 2 matrices, AA acts on spin 1 and BB on spin 2; σ→1\vec\sigma_1 and σ→2\vec\sigma_2 denote the Pauli vector acting on spin 1 and on spin 2.

The isotropic coupling σ→1⋅σ→2\vec\sigma_1 \cdot \vec\sigma_2 of the two spins is the operator σ→1⋅σ→2=∑a=13σa⊗σa=X⊗X+Y⊗Y+Z⊗Z\vec\sigma_1 \cdot \vec\sigma_2 = \sum_{a=1}^{3} \sigma_a \otimes \sigma_a = X \otimes X + Y \otimes Y + Z \otimes Z(3)✦ on the pair. For a real vector n→\vec n, the anisotropic coupling (σ→1⋅n→)(σ→2⋅n→)(\vec\sigma_1 \cdot \vec n)(\vec\sigma_2 \cdot \vec n) along n→\vec n is the tensor product of n→⋅σ→\vec n \cdot \vec\sigma with itself: (σ→1⋅n→)(σ→2⋅n→)=(n→⋅σ→)⊗(n→⋅σ→).(\vec\sigma_1 \cdot \vec n)(\vec\sigma_2 \cdot \vec n) = (\vec n \cdot \vec\sigma) \otimes (\vec n \cdot \vec\sigma).(4)✦

The dipolar bracket D(n→)D(\vec n) along a real vector n→\vec n is the operator on the pair D(n→)=σ→1⋅σ→2−3(σ→1⋅n→)(σ→2⋅n→).D(\vec n) = \vec\sigma_1 \cdot \vec\sigma_2 - 3\,(\vec\sigma_1 \cdot \vec n)(\vec\sigma_2 \cdot \vec n).(5)✦ For real numbers γ1\gamma_1, γ2\gamma_2, ℏ\hbar and rr, the dipolar coupling H1,2D(n→)H^D_{1,2}(\vec n) along n→\vec n is the real multiple of the bracket H1,2D(n→)=γ1γ2ℏ4r3D(n→).H^D_{1,2}(\vec n) = \frac{\gamma_1 \gamma_2 \hbar}{4 r^3}\, D(\vec n).(6)✦ Evaluated at the unit vector nˆ\hat n joining the spins, it is the coupling of the exercise.

Averaging Over Directions

The unit vector nˆ(θ,ϕ)\hat n(\theta,\phi) of polar angle θ\theta and azimuth ϕ\phi is the point of the unit sphere with Cartesian coordinates nˆ(θ,ϕ)=(sin⁡θcos⁡ϕ,sin⁡θsin⁡ϕ,cos⁡θ).\hat n(\theta,\phi) = (\sin\theta\cos\phi,\ \sin\theta\sin\phi,\ \cos\theta).(7)✦

The spherical average ⟨f⟩\langle f \rangle of a complex function ff of a real vector in three dimensions is the mean of ff over the unit sphere, ⟨f⟩=14π∫0π(∫02πf(nˆ(θ,ϕ))dϕ)sin⁡θdθ.\langle f \rangle = \frac{1}{4\pi} \int_0^{\pi} \left( \int_0^{2\pi} f\bigl(\hat n(\theta,\phi)\bigr)\, d\phi \right) \sin\theta \, d\theta .(8)✦ The spherical average ⟨F⟩\langle F \rangle of a function FF that assigns a complex matrix to every real vector in three dimensions is the matrix of the same shape whose entry in row jj and column kk is the spherical average of the corresponding entry of FF: ⟨F⟩jk=⟨n→↦F(n→)jk⟩.\langle F \rangle_{jk} = \bigl\langle \vec n \mapsto F(\vec n)_{jk} \bigr\rangle .(9)

The Claim

The dipolar coupling is such a function, n→↦H1,2D(n→)\vec n \mapsto H^D_{1,2}(\vec n), with values in the 4×44\times 4 matrices on the pair, and its spherical average is the average over the directions nˆ\hat n that the exercise asks for.

Theorem 1. Let γ1\gamma_1, γ2\gamma_2, ℏ\hbar and rr be real numbers. Then the spherical average of the dipolar coupling H1,2DH^D_{1,2} over the directions nˆ\hat n is the zero operator on the two spins: ⟨H1,2D⟩=0.\bigl\langle H^D_{1,2} \bigr\rangle = 0 .(10)✦

What Is Checked

The theorem is stated over the dipolar coupling (6), built from the Pauli matrices (1) through the couplings (3) and (4), and over the spherical average of a matrix (9), with the integral (8) over the angles of (7).

The library proves of these constructions that each of the sixteen entries of H1,2D(nˆ(θ,ϕ))H^D_{1,2}\bigl(\hat n(\theta,\phi)\bigr), a function of the two angles, integrates to zero over the sphere.