Section 22.1

Chapter Problems

[13/100%]
Show
LaTeX
Problem 22.1.1

The four Dirac matrices are given by

γ1=(000−i00−i00i00i000),γ2=(000−100100100−1000), \gamma _{1} = \begin{pmatrix} 0 & 0 & 0 & -i \\ 0 & 0 & -i & 0 \\ 0 & i & 0 & 0 \\ i & 0 & 0 & 0 \end{pmatrix}, \quad \gamma _{2} = \begin{pmatrix} 0 & 0 & 0 & -1 \\ 0 & 0 & 1 & 0 \\ 0 & 1 & 0 & 0 \\ -1 & 0 & 0 & 0 \end{pmatrix}, γ3=(00−i0000ii0000−i00),γ4=(1000010000−10000−1) \gamma _{3} = \begin{pmatrix} 0 & 0 & -i & 0 \\ 0 & 0 & 0 & i \\ i & 0 & 0 & 0 \\ 0 & -i & 0 & 0 \end{pmatrix}, \quad \gamma _{4} = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \end{pmatrix}

and

γ5=γ1γ2γ3γ4=(00−10000−1−10000−100). \gamma _{5} = \gamma _{1}\gamma _{2}\gamma _{3}\gamma _{4} = \begin{pmatrix} 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \\ -1 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 \end{pmatrix}.

We define the 4×44 \times 4 matrices

σjk:=i2[γj,γk],j<k \sigma _{jk} := \frac{i}{2}[\gamma _{j}, \gamma _{k}], \qquad j < k

where j=1,2,3j = 1, 2, 3, k=2,3,4k = 2, 3, 4.

?
(i)

Calculate σ12\sigma_{12}, σ13\sigma_{13}, σ14\sigma_{14}, σ23\sigma_{23}, σ24\sigma_{24}, σ34\sigma_{34}.

(ii)

Do the 16 matrices

I4, γ1, γ2, γ3, γ4, γ5, γ5γ1, γ5γ2, γ5γ3, γ5γ4, σ12, σ13, σ14, σ23, σ24, σ34 I_{4}, \ \gamma _{1}, \ \gamma _{2}, \ \gamma _{3}, \ \gamma _{4}, \ \gamma _{5}, \ \gamma _{5}\gamma _{1}, \ \gamma _{5}\gamma _{2}, \ \gamma _{5}\gamma _{3}, \ \gamma _{5}\gamma _{4}, \ \sigma _{12}, \ \sigma _{13}, \ \sigma _{14}, \ \sigma _{23}, \ \sigma _{24}, \ \sigma _{34}

form a basis in the Hilbert space M4(C)M_{4}(\mathbb {C})? If so is the basis orthogonal?

Problem 22.1.2

Let PP be the n×nn \times n primary permutation matrix

P:=(010…0001…0⋮⋮⋮⋱⋮000…1100…0) P := \begin{pmatrix} 0 & 1 & 0 & \ldots & 0 \\ 0 & 0 & 1 & \ldots & 0 \\ \vdots & \vdots & \vdots & \ddots & \vdots \\ 0 & 0 & 0 & \ldots & 1 \\ 1 & 0 & 0 & \ldots & 0 \end{pmatrix}

and VV be the n×nn \times n unitary diagonal matrix (ζ∈C\zeta \in \mathbb {C})

V:=(10000ζ0000ζ20⋮⋮⋮⋮000…ζn−1) V := \begin{pmatrix} 1 & 0 & 0 & & 0 \\ 0 & \zeta & 0 & & 0 \\ 0 & 0 & \zeta ^{2} & & 0 \\ \vdots & \vdots & \vdots & & \vdots \\ 0 & 0 & 0 & \ldots & \zeta ^{n-1} \end{pmatrix}

where ζn=1\zeta^{n} = 1. Then the set of matrices

{PjVk:j,k=0,1,…,n−1} \left\{ P^{j}V^{k} : j, k = 0, 1, \ldots , n-1\right\}

provide a basis in the Hilbert space for all n×nn \times n matrices with the scalar product

⟨A,B⟩:=1ntr⁡(AB∗) \left\langle A, B \right\rangle := \frac{1}{n}\operatorname {tr}(AB^{*})

for n×nn \times n matrices AA and BB. Write down the basis for n=2n = 2.

?
Problem 22.1.3

Consider the two normalized vectors

v→(α)=(cos⁡(α)sin⁡(α)),u→(β)=(cos⁡(β)sin⁡(β)) \overrightarrow {v}(\alpha ) = \begin{pmatrix} \cos (\alpha ) \\ \sin (\alpha ) \end{pmatrix}, \quad \overrightarrow {u}(\beta ) = \begin{pmatrix} \cos (\beta ) \\ \sin (\beta ) \end{pmatrix}

in R2\mathbb {R}^{2}. Find the condition on α,β∈R\alpha , \beta \in \mathbb {R} such that

∣v→∗(α)u→(β)∣2=12. \left|\overrightarrow {v}^{*}(\alpha )\overrightarrow {u}(\beta )\right|^{2} = \frac{1}{2}.
?
Problem 22.1.4

Consider the four nonnormal 2×22 \times 2 matrices

Q=12(1100),R=12(001−1), Q = \frac{1}{\sqrt{2}}\begin{pmatrix} 1 & 1 \\ 0 & 0 \end{pmatrix}, \quad R = \frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 0 \\ 1 & -1 \end{pmatrix}, S=12(1−100),T=12(0011). S = \frac{1}{\sqrt{2}}\begin{pmatrix} 1 & -1 \\ 0 & 0 \end{pmatrix}, \quad T = \frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 0 \\ 1 & 1 \end{pmatrix}.

Show that they form an orthonormal basis in the Hilbert space of the 2×22 \times 2 matrices with the scalar product ⟨X,Y⟩=tr⁡(XY∗)\left\langle X, Y \right\rangle = \operatorname {tr}(XY^{*}).

?
Problem 22.1.5

Consider the Hilbert space Md(C)M_{d}(\mathbb {C}) of d×dd \times d matrices with scalar product ⟨A,B⟩:=tr⁡(AB∗)\left\langle A, B \right\rangle := \operatorname {tr}(AB^{*}), A,B∈Md(C)A, B \in M_{d}(\mathbb {C}). Consider an orthogonal basis of d2d^{2} d×dd \times d hermitian matrices B1,B2,…,Bd2B_{1}, B_{2}, \ldots , B_{d^{2}}, i.e.

⟨Bj,Bk⟩=tr⁡(BjBk)=dδjk \left\langle B_{j}, B_{k} \right\rangle = \operatorname {tr}(B_{j}B_{k}) = d\delta _{jk}

since Bk∗=BkB_{k}^{*} = B_{k} for a hermitian matrix. Let MM be a d×dd \times d hermitian matrix. Let

mj=tr⁡(BjM)j=1,…,d2. m_{j} = \operatorname {tr}(B_{j}M) \qquad j = 1, \ldots , d^{2}.

Given mjm_{j} and BjB_{j} (j=1,…,d2j = 1, \ldots , d^{2}). Find MM.

?
Problem 22.1.6
?
(i)

Consider the Hilbert space H\mathcal{H} of the 2×22 \times 2 matrices over the complex numbers. Show that the rescaled Pauli spin matrices μj=12σj\mu_{j} = \frac{1}{\sqrt{2}}\sigma_{j}, j=1,2,3j = 1, 2, 3

μ1=12(0110),μ2=12(0−ii0),μ3=12(100−1) \mu _{1} = \frac{1}{\sqrt{2}}\begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}, \quad \mu _{2} = \frac{1}{\sqrt{2}}\begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix}, \quad \mu _{3} = \frac{1}{\sqrt{2}}\begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix}

plus the rescaled 2×22 \times 2 identity matrix

μ0=12(1001) \mu _{0} = \frac{1}{\sqrt{2}}\begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix}

form an orthonormal basis in the Hilbert space H\mathcal{H}.

(ii)

Find an orthonormal basis given by hermitian matrices in the Hilbert space H\mathcal{H} of 4×44 \times 4 matrices over C\mathbb {C}.

Problem 22.1.7

Let AA, BB be hermitian matrices. Find the scalar product

⟨A⊗B,B⊗A⟩. \left\langle A \otimes B, B \otimes A \right\rangle .

Is ⟨A⊗B,B⊗A⟩≥0\left\langle A \otimes B, B \otimes A \right\rangle \geq 0. Prove or disprove.

?
Problem 22.1.8

Consider the Hilbert space of the n×nn \times n matrices over C\mathbb {C}. Let AA, BB be two n×nn \times n matrices over C\mathbb {C}. Assume that ⟨A,B⟩=0\left\langle A, B \right\rangle = 0. Calculate

⟨A⊗A,B⊗B⟩,⟨A⊗B,A⊗B⟩,⟨A⊗B,B⊗A⟩. \left\langle A \otimes A, B \otimes B \right\rangle , \qquad \left\langle A \otimes B, A \otimes B \right\rangle , \qquad \left\langle A \otimes B, B \otimes A \right\rangle .
?
Problem 22.1.9

Consider the Hilbert space H=C4\mathcal{H} = \mathbb {C}^{4} and the hermitian matrices

H=(2−10−1−12−100−12−1−10−12) H = \begin{pmatrix} 2 & -1 & 0 & -1 \\ -1 & 2 & -1 & 0 \\ 0 & -1 & 2 & -1 \\ -1 & 0 & -1 & 2 \end{pmatrix} A=(1000000000−100000),B=(000001000000000−1). A = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & 0 \end{pmatrix}, \qquad B = \begin{pmatrix} 0 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & -1 \end{pmatrix}.

We consider the 16×1616 \times 16 matrix

K=12(H⊗I4+I4⊗H)−λA⊗A−μB⊗B K = \frac{1}{2}(H \otimes I_{4} + I_{4} \otimes H) - \lambda A \otimes A - \mu B \otimes B

acting in the Hilbert space C16\mathbb {C}^{16}, where λ,μ≥0\lambda , \mu \geq 0.

?
(i)

Consider the permutation matrix

R=(0100001000011000). R = \begin{pmatrix} 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \\ 1 & 0 & 0 & 0 \end{pmatrix}.

Thus the inverse of RR exists. Calculate (R⊗R)−1K(R⊗R)(R \otimes R)^{-1}K(R \otimes R) and the commutator [R⊗R,K][R \otimes R, K]. Discuss.

(ii)

Consider the permutation matrices

S=(0010010010000001),T=(1000000100100100). S = \begin{pmatrix} 0 & 0 & 1 & 0 \\ 0 & 1 & 0 & 0 \\ 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 1 \end{pmatrix}, \qquad T = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 1 \\ 0 & 0 & 1 & 0 \\ 0 & 1 & 0 & 0 \end{pmatrix}.

Find the commutators [S⊗S,K][S \otimes S, K] and [T⊗T,K][T \otimes T, K]. Discuss.

Problem 22.1.10

Consider the Hilbert space of 2×22 \times 2 matrices. Let α,β∈R\alpha , \beta \in \mathbb {R} and

A(α)=12(cos⁡(α)sin⁡(α)sin⁡(α)−cos⁡(α)),B(β)=12(cos⁡(β)sin⁡(β)sin⁡(β)−cos⁡(β)). A(\alpha ) = \frac{1}{\sqrt{2}}\begin{pmatrix} \cos (\alpha ) & \sin (\alpha ) \\ \sin (\alpha ) & -\cos (\alpha ) \end{pmatrix}, \qquad B(\beta ) = \frac{1}{\sqrt{2}}\begin{pmatrix} \cos (\beta ) & \sin (\beta ) \\ \sin (\beta ) & -\cos (\beta ) \end{pmatrix}.

Find the conditions on α\alpha and β\beta such that ⟨A(α),B(β)⟩=12\left\langle A(\alpha ), B(\beta ) \right\rangle = \frac{1}{2}. The matrices A(α)A(\alpha ), B(β)B(\beta ) contain 12σ3\frac{1}{\sqrt{2}}\sigma_{3} with α=0\alpha = 0 and 12σ1\frac{1}{\sqrt{2}}\sigma_{1} with α=π/2\alpha = \pi /2.

Study the case

⟨A(α)⊗A(α)⟩=⟨B(β)⊗B(β)⟩=14. \left\langle A(\alpha ) \otimes A(\alpha )\right\rangle = \left\langle B(\beta ) \otimes B(\beta )\right\rangle = \frac{1}{4}.

Extend the matrix A(α)A(\alpha ) to

A(α,ϕ)=12(cos⁡(α)eiϕsin⁡(α)e−iϕsin⁡(α)−cos⁡(α)). A(\alpha , \phi ) = \frac{1}{\sqrt{2}}\begin{pmatrix} \cos (\alpha ) & e^{i\phi }\sin (\alpha ) \\ e^{-i\phi }\sin (\alpha ) & -\cos (\alpha ) \end{pmatrix}.
?
Problem 22.1.11

Consider the Pauli spin matrices σ3\sigma_{3}, σ1\sigma_{1}, σ2\sigma_{2}. Show that the normalized eigenvectors of σ3\sigma_{3}, σ1\sigma_{1}, σ2\sigma_{2} provide a set of mutually unbiased bases.

?
Problem 22.1.12

Consider the three 4×44 \times 4 matrices σ3⊗σ3\sigma_{3} \otimes \sigma_{3}, σ1⊗σ1\sigma_{1} \otimes \sigma_{1}, σ2⊗σ2\sigma_{2} \otimes \sigma_{2}. Find mutually unbiased bases for these matrices applying the result from the previous problem.

Extend the results from problem 11 and problem 12 the case in the Hilbert space C8\mathbb {C}^{8} with σ3⊗σ3⊗σ3\sigma_{3} \otimes \sigma_{3} \otimes \sigma_{3}, σ1⊗σ1⊗σ1\sigma_{1} \otimes \sigma_{1} \otimes \sigma_{1}, σ2⊗σ2⊗σ2\sigma_{2} \otimes \sigma_{2} \otimes \sigma_{2}. Extend to nn Kronecker products of the Pauli spin matrices.

?
Problem 22.1.13

Let d≥2d \geq 2. Consider the Hilbert space Cd\mathbb {C}^{d} and e→0,e→1,…,e→d−1\overrightarrow {e}_{0}, \overrightarrow {e}_{1}, \ldots , \overrightarrow {e}_{d-1} be the standard basis

e→0=(100⋮0),e→1=(010⋮0),…,e→d−1=(00⋮01). \overrightarrow {e}_{0} = \begin{pmatrix} 1 \\ 0 \\ 0 \\ \vdots \\ 0 \end{pmatrix}, \quad \overrightarrow {e}_{1} = \begin{pmatrix} 0 \\ 1 \\ 0 \\ \vdots \\ 0 \end{pmatrix}, \quad \ldots , \quad \overrightarrow {e}_{d-1} = \begin{pmatrix} 0 \\ 0 \\ \vdots \\ 0 \\ 1 \end{pmatrix}.

Let ωd:=ei2π/d\omega_{d} := e^{i2\pi /d}. Then we can form new orthonormal bases via

v→m;b≡∣m;b⟩=1d∑n=0d−1ωbn(n−1)/2−nme→n,b,m=0,1,…,d−1 \overrightarrow {v}_{m;b} \equiv \left|m;b\right\rangle = \frac{1}{\sqrt{d}}\sum _{n=0}^{d-1} \omega ^{bn(n-1)/2-nm}\overrightarrow {e}_{n}, \qquad b, m = 0, 1, \ldots , d-1

where the dd labelled by the bb are the bases and mm labels the state within a basis. Thus we find mutually unbiased bases.

?
(i)

Consider d=2d = 2 and b=0b = 0. Find the basis ∣0;0⟩\left|0;0\right\rangle, ∣1;0⟩\left|1;0\right\rangle.

(ii)

Consider d=3d = 3 and b=0b = 0. Find the basis ∣1;0⟩\left|1;0\right\rangle, ∣2;0⟩\left|2;0\right\rangle, ∣3;0⟩\left|3;0\right\rangle.

(iii)

Consider d=4d = 4 and b=0b = 0. Find the basis ∣0;0⟩\left|0;0\right\rangle, ∣1;0⟩\left|1;0\right\rangle, ∣2;0⟩\left|2;0\right\rangle, ∣3;0⟩\left|3;0\right\rangle. Find out whether the states can be written as Kronecker product of two vectors in C2\mathbb {C}^{2}, i.e. whether the states are entangled or not?