Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

Open In Colab

Example 1

Let us consider the operator A=(100010002).A=\begin{pmatrix}1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 2 \end{pmatrix}. To verify that it is a Hermitian matrix, we can verify its conjugate transposeA†=A.A^{\dagger}=A. Its eigenvalues are {1,1,2},\left\{ 1,1,2\right\} ,and the corresponding eigenvectors (each column is an eigenvector) are (∣a1⟩,∣a2⟩,∣a3⟩)=(100010001).\left(\left|a_{1}\right\rangle ,\left|a_{2}\right\rangle ,\left|a_{3}\right\rangle \right)=\begin{pmatrix}1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 1 \end{pmatrix}. We have then that eigenvalue 1 is degenerate. We can reconstruct matrix AA by spectral decomposition: A=∑n=13an∣an⟩⟨an∣.A=\sum^{3}_{n=1}a_{n}\left|a_{n}\right\rangle \left\langle a_{n}\right|. We have now defined a Hermitian matrix B=(310140005).B=\begin{pmatrix}3 & 1 & 0\\ 1 & 4 & 0\\ 0 & 0 & 5 \end{pmatrix}. We can verify that they commute, that is, [A,B]=AB−BA=0\left[A,B\right]=AB-BA=0. In the simplest case, we then have a subspace of only one dimension for eigenvalue 2. We can see that B∣a3⟩=5∣a3⟩B\left|a_{3}\right\rangle =5\left|a_{3}\right\rangle .

Now for eigenvalue 1, we have a two-dimensional subspace formed by the vectors {∣a1⟩,∣a2⟩}\left\{ \left|a_{1}\right\rangle ,\left|a_{2}\right\rangle \right\} . All non-null vectors in this subspace continue to be eigenvectors of AA with the same eigenvalue 1, Any non-null vector that is a linear combination of them, ∣u⟩=a∣a1⟩+b∣a2⟩,\left|u\right\rangle =a\left|a_{1}\right\rangle +b\left|a_{2}\right\rangle , is also an eigenvector of AA. In fact,

A∣u⟩=A(a∣a1⟩+b∣a2⟩)=aA∣a1⟩+bA∣a2⟩=a(1⋅∣a1⟩)+b(1⋅∣a2⟩)=a∣a1⟩+b∣a2⟩=1⋅∣u⟩A\left|u\right\rangle =A\left(a\left|a_{1}\right\rangle +b\left|a_{2}\right\rangle \right)=aA\left|a_{1}\right\rangle +bA\left|a_{2}\right\rangle =a\left(1\cdot\left|a_{1}\right\rangle \right)+b\left(1\cdot\left|a_{2}\right\rangle \right)=a\left|a_{1}\right\rangle +b\left|a_{2}\right\rangle =1\cdot\left|u\right\rangle

Now applying BB to the first eigenvector, we can see that it is a linear combination of the other two vectors in the subspace and is not a BB eigenvector:

B∣a1⟩=(310140005)(100)=(310)=3∣a1⟩+∣a2⟩B\left|a_{1}\right\rangle =\begin{pmatrix}3 & 1 & 0\\ 1 & 4 & 0\\ 0 & 0 & 5 \end{pmatrix}\begin{pmatrix}1\\ 0\\ 0 \end{pmatrix}=\begin{pmatrix}3\\ 1\\ 0 \end{pmatrix}=3\left|a_{1}\right\rangle +\left|a_{2}\right\rangle

That is, even though these two eigenvectors associated with 1 are a basis for this subspace, they do not form a common eigenvector basis. But we can build one. Since all non-null vectors in this subspace are eigenvectors of AA, we only need to find those who are also eigenvectors of BB. In this two-dimensional space, BB’s action on the vectors of the space is given by B1=(3114).B_{1}=\begin{pmatrix}3 & 1\\ 1 & 4 \end{pmatrix}. Its eigenvalues are {2.38,4.62}\left\{ 2.38,4.62\right\} and the corresponding eigenvectors (each column is an eigenvector) are (−0.85−0.530.53−0.85).\begin{pmatrix}-0.85 & -0.53\\ 0.53 & -0.85 \end{pmatrix}.

Combining with the previous results, we can have a new basis given by the eigenvectors (where each column corresponds to a vector of the new basis)

(∣b1⟩,∣b2⟩,∣b3⟩)=(−0.85−0.5300.53−0.850001).\left(\left|b_{1}\right\rangle ,\left|b_{2}\right\rangle ,\left|b_{3}\right\rangle \right)=\begin{pmatrix}-0.85 & -0.53 & 0\\ 0.53 & -0.85 & 0\\ 0 & 0 & 1 \end{pmatrix}.

We can test by applying AA to the vectors of the new basis:

A∣b1⟩=∣b1⟩(100010002)(−0.850.530)=(−0.850.530)A∣b2⟩=∣b2⟩(100010002)(−0.53−0.850)=(−0.53−0.850)A∣b3⟩=2∣b3⟩(100010002)(001)=2(001)\begin{align*} A\left|b_{1}\right\rangle & =\left|b_{1}\right\rangle & \begin{pmatrix}1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 2 \end{pmatrix}\begin{pmatrix}-0.85\\ 0.53\\ 0 \end{pmatrix} & =\begin{pmatrix}-0.85\\ 0.53\\ 0 \end{pmatrix}\\ A\left|b_{2}\right\rangle & =\left|b_{2}\right\rangle & \begin{pmatrix}1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 2 \end{pmatrix}\begin{pmatrix}-0.53\\ -0.85\\ 0 \end{pmatrix} & =\begin{pmatrix}-0.53\\ -0.85\\ 0 \end{pmatrix}\\ A\left|b_{3}\right\rangle & =2\left|b_{3}\right\rangle & \begin{pmatrix}1 & 0 & 0\\ 0 & 1 & 0\\ 0 & 0 & 2 \end{pmatrix}\begin{pmatrix}0\\ 0\\ 1 \end{pmatrix} & =2\begin{pmatrix}0\\ 0\\ 1 \end{pmatrix} \end{align*}

We can test by applying BB to the vectors of the new basis:

B∣b1⟩=2.38∣b1⟩(310140005)(−0.850.530)=2.38(−0.850.530)B∣b2⟩=4.62∣b2⟩(310140005)(−0.53−0.850)=4.62(−0.53−0.850)B∣b3⟩=5∣b3⟩(310140005)(001)=5(001)\begin{align*} B\left|b_{1}\right\rangle & =2.38\left|b_{1}\right\rangle & \begin{pmatrix}3 & 1 & 0\\ 1 & 4 & 0\\ 0 & 0 & 5 \end{pmatrix}\begin{pmatrix}-0.85\\ 0.53\\ 0 \end{pmatrix} & =2.38\begin{pmatrix}-0.85\\ 0.53\\ 0 \end{pmatrix}\\ B\left|b_{2}\right\rangle & =4.62\left|b_{2}\right\rangle & \begin{pmatrix}3 & 1 & 0\\ 1 & 4 & 0\\ 0 & 0 & 5 \end{pmatrix}\begin{pmatrix}-0.53\\ -0.85\\ 0 \end{pmatrix} & =4.62\begin{pmatrix}-0.53\\ -0.85\\ 0 \end{pmatrix}\\ B\left|b_{3}\right\rangle & =5\left|b_{3}\right\rangle & \begin{pmatrix}3 & 1 & 0\\ 1 & 4 & 0\\ 0 & 0 & 5 \end{pmatrix}\begin{pmatrix}0\\ 0\\ 1 \end{pmatrix} & =5\begin{pmatrix}0\\ 0\\ 1 \end{pmatrix} \end{align*}

Example 2 Suppose we have the operators A=(0110)A=\begin{pmatrix}0 & 1\\ 1 & 0 \end{pmatrix} and B=(100−1).B=\begin{pmatrix}1 & 0\\ 0 & -1 \end{pmatrix}. We can verify that they do not commute, since

[A,B]=AB−BA=(0−220)≠0.[A,B]=AB-BA=\begin{pmatrix}0 & -2\\ 2 & 0 \end{pmatrix}\neq0.

For the operator AA, the eigenvalues are {+1,−1}\left\{ +1,-1\right\} with corresponding normalized eigenvectors

∣a+1⟩=12(11),∣a−1⟩=12(−11).|a_{+1}\rangle=\frac{1}{\sqrt{2}}\begin{pmatrix}1\\ 1 \end{pmatrix},\qquad|a_{-1}\rangle=\frac{1}{\sqrt{2}}\begin{pmatrix}-1\\ 1 \end{pmatrix}.

Suppose, for example, that after measuring AA, we obtain the value a=+1a=+1. The system is then in the corresponding eigenstate, ∣ψ⟩=∣a+1⟩.|\psi\rangle=|a_{+1}\rangle. Now consider measuring BB. The eigenvalues of BB are also {+1,−1}\left\{ +1,-1\right\} with corresponding eigenvectors

∣b+1⟩=(10),∣b−1⟩=(01).|b_{+1}\rangle=\begin{pmatrix}1\\ 0 \end{pmatrix},\qquad|b_{-1}\rangle=\begin{pmatrix}0\\ 1 \end{pmatrix}.

The probability of obtaining a particular eigenvalue bnb_{n} is given by the squared magnitude of the inner product between the corresponding eigenvector and the state of the system: P(bn)=∣⟨bn∣ψ⟩∣2P(b_{n})=|\langle b_{n}|\psi\rangle|^{2}. Therefore,

P(b+1)=∣⟨b+1∣a+1⟩∣2=12P(b−1)=∣⟨b−1∣a−1⟩∣2=12\begin{align*} P(b_{+1})=|\langle b_{+1}|a_{+1}\rangle|^{2}=\frac{1}{2}\\ P(b_{-1})=|\langle b_{-1}|a_{-1}\rangle|^{2}=\frac{1}{2} \end{align*}

Thus, there is a 50%50\% probability of observing +1 and a 50%50\% probability of observing -1. In other words, since AA and BB do not commute and, in this example, do not share any eigenvectors, after measuring one of the observables, the outcome of a subsequent measurement of the other is intrinsically indeterminate within this mathematical formalism.

Codes