Quantum Mechanics The Theoretical Minimum Exercises Complete1

The exercises of „Quantum Mechanics, The Theoretical Minimum“
Preface
This file contains the exercises of „Quantum Mechanics, The Theoretical Minimum“ and is specific in
this respect. On the other hand, the topics generally deal with quantum mechanics and maybe are
helpful for reasons of training too.
I tried to make the exercises explicit so most of them can be tackled without specific knowledge of
the book itself. I also tried to write a kind of “deep dive” solutions that give you more information
than just the correct result.
Hope I can help you with learning quantum mechanics.
I would like to thank Kenneth Verbist for finding a bug (certainly not the only one …) and thereby
helping to improve this project.
Dieter Kriesell
page 1 of 105
Exercise 1.1
a) Using the axioms for inner products, prove {⟨𝐴| + ⟨𝐵|}|𝐶⟩ = ⟨𝐴|𝐶⟩ + ⟨𝐵|𝐶⟩.
b) Prove that ⟨𝐴|𝐴⟩ is a real number.
Exercise 1.2
Show that the inner product satisfies linearity and interchange.
Inner product:
𝑏1
𝑏
⟨𝐴|𝐵⟩ = (𝑎1∗ 𝑎2∗ … 𝑎𝑛∗ 1 ) ( 2 ) = 𝑎1∗ 𝑏1 + 𝑎2∗ 𝑏2 + ⋯ + 𝑎𝑛∗ 𝑏𝑛
…
𝑏𝑛
Linearity: ⟨𝐶|{|𝐴⟩ + |𝐵⟩} = ⟨𝐶|𝐴⟩ + ⟨𝐶|𝐵⟩
Interchange: ⟨𝐵|𝐴⟩ = ⟨𝐴|𝐵⟩∗
Exercise 2.1
Prove that the vector |𝑟⟩ is orthogonal to vector |𝑙⟩:
1 1
|𝑟⟩ = |𝑢⟩ + |𝑑⟩
√2 √2
1 1
|𝑙⟩ = |𝑢⟩ − |𝑑⟩
√2 √2
Exercise 2.2
Prove that |𝑖⟩ and |𝑜⟩ satisfy the conditions in Eqs. 2.7, 2.8, and 2.9.
Are they unique in that respect?
Eq. 2.7:
⟨𝑖|𝑜⟩ = 0
Eqs. 2.8:
1 1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ = ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ =
2 2
Eqs. 2.9
1 1
⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩ = ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩ =
2 2
1 𝑖 1 𝑖
|𝑖⟩ = |𝑢⟩ + |𝑑⟩ and ⟨𝑖| = ⟨𝑢| − ⟨𝑑| 2
√2 √2 √2 √
1 𝑖 1 𝑖
|𝑜⟩ = |𝑢⟩ − |𝑑⟩ and ⟨𝑜| = ⟨𝑢| + ⟨𝑑|
√2 √2 √2 √2
page 2 of 105
Exercise 2.3
Assume that |𝑖⟩ and |𝑜⟩ are given as:
|𝑖⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩ and ⟨𝑖| = ⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗
|𝑜⟩ = 𝛾|𝑢⟩ − 𝛿|𝑑⟩ and ⟨𝑜| = ⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗
a) Use Eqs. 2.8 to show that
1
𝛼 ∗𝛼 = 𝛽∗𝛽 = 𝛾 ∗𝛾 = 𝛿 ∗𝛿 =
2
b) Use the above result and Eqs. 2.9 to show that:
𝛼 ∗ 𝛽 + 𝛼𝛽 ∗ = 0
c) Show that 𝛼 ∗ 𝛽 must be pure imaginary.
Eqs. 2.8:
1 1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ = ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ =
2 2
Eqs. 2.9
1 1
⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩ = ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩ =
2 2
Exercise 3.1
Prove: If a vector space is N-dimensional, an orthonormal basis of N vectors can be constructed
from the eigenvectors of a Hermitian operator.
Exercise 3.2
Prove that 𝜎𝑧 of Eq. 3.16 is the unique solution to Eqs. 3.14 und 3.15.
Eq. 3.16
(𝜎 ) (𝜎𝑧 )12 1 0
𝜎𝑧 = ( 𝑧 11 )=( )
(𝜎𝑧 )21 (𝜎𝑧 )21 0 −1
Eq. 3.14
(𝜎 ) (𝜎𝑧 )12 1 1
( 𝑧 11 )( ) = ( )
(𝜎𝑧 )21 (𝜎𝑧 )21 0 0
Eq. 3.15
(𝜎 ) (𝜎𝑧 )12 0 0
( 𝑧 11 )( ) = −( )
(𝜎𝑧 )21 (𝜎𝑧 )21 1 1
Exercise 3.3
Calculate the eigenvectors and eigenvalues of 𝜎𝑛 .
Assume the eigenvector 𝜆1 has the form:
cos 𝛼
( )
sin 𝛼
𝛼 is an unknown parameter. Plug this vector into the eigenvalue equation and solve for 𝛼 in terms
of 𝜃.
Why did we use a single parameter 𝛼?
Notice that our suggested column vector must have unit length.
𝑐𝑜𝑠 𝜃 sin 𝜃
𝜎𝑛 = ( )
sin 𝜃 − cos 𝜃
To show:
𝑐𝑜𝑠 𝜃 sin 𝜃 cos 𝛼 cos 𝛼
( )( ) = 𝜆1 ( )
sin 𝜃 − cos 𝜃 sin 𝛼 sin 𝛼
page 3 of 105
Exercise 3.4
Let 𝑛𝑧 = 𝑐𝑜𝑠𝜃, 𝑛𝑥 = 𝑠𝑖𝑛𝜃𝑐𝑜𝑠𝜙, and 𝑛𝑦 = 𝑠𝑖𝑛𝜃𝑠𝑖𝑛𝜙.
Angles 𝜃 and 𝜙 are defined according to the usual conventions for spherical coordinates.
Compute the eigenvalues and eigenvectors for the matrix 𝜎𝑛 :
𝑛𝑧 (𝑛𝑥 − 𝑖𝑛𝑦 )
𝜎𝑛 = ( )
(𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧
Exercise 3.5
Suppose a spin is prepared so that 𝜎𝑚 = +1. The apparatus is then rotated to the 𝑛̂ direction and
𝜎𝑛 is measured. What is the probability that the result is +1?
Note that 𝜎𝑚 = 𝜎 ∙ 𝑚̂ , using the same convention we used for 𝜎𝑛 .
Exercise 4.1
Prove that if (the time operator) U is unitary, and if |𝐴⟩ and |𝐵⟩ are any two state-vectors, then the
inner product of 𝑈|𝐴⟩ and 𝑈|𝐵⟩ is the same as the inner product of |𝐴⟩ and |𝐵⟩. One could call this
the conservation of overlaps. It expresses the fact that the logical relation between states is
preserved with time.
U is unitary:
𝑈†𝑈 = 𝐼
𝑈|𝐴⟩ = ⟨𝐴|𝑈 †
Exercise 4.2
Prove that if 𝑀 and 𝐿 are both Hermitian, the (extended) commutator 𝑖[𝑀, 𝐿] is also Hermitian.
Note that the 𝑖 is important. The commutator is, by itself, not Hermitian.
Hermitian: the diagonal is pure real and: 𝑀 = 𝑀†
Exercise 4.3
With the definition of Poisson brackets check that the identification in Eq. 4.21 is dimensionally
consistent. Show that without the factor ℏ, it would not be.
Eq. 4.21
[𝐹, 𝐺] ↔ 𝑖ℏ{𝐹, 𝐺}
𝑘𝑔 ∙ 𝑚2
[ℏ] = 𝐽 ∙ 𝑠 = ∙𝑠
𝑠2
Exercise 4.4
Verify the commutation relations:
[𝜎𝑥 , 𝜎𝑦 ] = 2𝑖𝜎𝑧
[𝜎𝑦 , 𝜎𝑧 ] = 2𝑖𝜎𝑥
[𝜎𝑧 , 𝜎𝑥 ] = 2𝑖𝜎𝑦
0 1 0 −𝑖 1 0
𝜎𝑥 = ( ) , 𝜎𝑦 = ( ) , 𝜎𝑧 = ( )
1 0 𝑖 0 0 −1
page 4 of 105
Exercise 4.5
Take any unit 3-vector 𝑛⃗ and form the operator
ℏ𝜔
𝐻= 𝜎 ∙ 𝑛⃗
2
Find the energy eigenvalues and eigenvectors by solving the time-independent Schrödinger
equation. Recall 𝜎 ∙ 𝑛⃗ in component form:
𝜎𝑛 = 𝜎 ∙ 𝑛⃗ = ( )
Exercise 4.6
Carry out the Schrödinger Ket recipe for a single spin.
ℏ𝜔
The Hamiltonian is 𝐻 = 𝜎𝑧 and the final observable is 𝜎𝑥 .
2
The initial state is given as |𝑢⟩ (the state in which 𝜎𝑧 = ±1).
After time t, an experiment is done to measure 𝜎𝑦 .
What are the possible outcomes and what are the probabilities for those outcomes?
Exercise 5.1
Verify that any 2 × 2 Hermitian matrix L can be written as a sum of four terms,
𝐿 = 𝑎𝜎𝑥 + 𝑏𝜎𝑦 + 𝑐𝜎𝑧 + 𝑑𝐼
where a, b, c and d are real numbers.
The four Pauli matrices:
0 1 0 −𝑖 1 0 1 0
𝜎𝑥 = ( ), 𝜎𝑦 = ( ), 𝜎𝑧 = ( ), I= ( )
1 0 𝑖 0 0 −1 0 1
A general Hermitian matrix (𝑟, 𝑟) are real numbers:
𝑟 𝑤
( ∗ )
𝑤 𝑟′
Exercise 5.2
1) Show that (△ 𝐴)2 = 〈𝐴2̅ 〉 and (△ 𝐵)2 = 〈𝐵̅ 2 〉
2) Show that [𝐴̅, 𝐵̅ ] = [𝐴, 𝐵]
1
3) Using these relations, show that △ 𝐴 △ 𝐵 ≥ 2 |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|
The square of uncertainty (or standard deviation) of A, (△ 𝐴)2 :

(△ 𝐴)² = ∑ 𝑎̅2 𝑃(𝑎) = ∑(𝑎 − 〈𝐴〉)2 𝑃(𝐴)
𝑎 𝑎
Exercise 6.1
Prove that if 𝑃(𝑎, 𝑏) factorizes:
𝑃(𝑎, 𝑏) = 𝑃𝐴 (𝑎)𝑃𝑏 (𝑏)
then the correlation between a and b is zero:
〈𝜎𝐴 𝜎𝐵 〉 − 〈𝜎𝐴 〉〈𝜎𝐵 〉 = 0
Average:
〈𝜎𝐴 〉 = ∑ 𝑎𝑛 𝑃(𝑎𝑛 )
𝑛
〈𝜎𝐵 〉 = ∑ 𝑏𝑛 𝑃(𝑏𝑛 )
𝑛
〈𝜎𝐴 𝜎𝐵 〉 = ∑ ∑ 𝑎𝑛 𝑏𝑛 𝑃(𝑎𝑛 𝑏𝑛 )
𝑛 𝑛
page 5 of 105
Exercise 6.2
Show that if the two normalization conditions are satisfied:
𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 = 1
𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 = 1
then the state-vector is automatically normalized as well:
𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩
In other words, show that for this product state, normalizing the overall state-vector does not put
any additional constraints on the 𝛼′𝑠 and 𝛽′𝑠.
Exercise 6.3
Prove that the state |𝑠𝑖𝑛𝑔⟩ cannot be written as a product state.
1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
The shape of a product state:
Exercise 6.4
Use the matrix forms of 𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 and the column vectors for |𝑢⟩ and |𝑑⟩ to verify:
𝜎𝑧 |𝑢⟩ = |𝑢⟩ and 𝜎𝑧 |𝑑⟩ = −|𝑑⟩

𝜎𝑥 |𝑢⟩ = |𝑑⟩ and 𝜎𝑥 |𝑑⟩ = |𝑢⟩
𝜎𝑦 |𝑢⟩ = 𝑖|𝑑⟩ and 𝜎𝑦 |𝑑⟩ = −𝑖|𝑢⟩
Then, use:
𝜏𝑧 |𝑢⟩ = |𝑢⟩ and 𝜏𝑧 |𝑑⟩ = −|𝑑⟩
𝜏𝑥 |𝑢⟩ = |𝑑⟩ and 𝜏𝑥 |𝑑⟩ = |𝑢⟩
𝜏𝑦 |𝑢⟩ = 𝑖|𝑑⟩ and 𝜏𝑦 |𝑑⟩ = −𝑖|𝑢⟩
to write the equations for all possible combinations of the tensor product states 𝜎𝑧 |𝑢𝑢⟩ = |𝑢𝑢⟩
etc.
Exercise 6.5
Prove the following theorem:
When any one of Alice’s and Bob’s spin operators acts on a product state, the result is still a
product state.
Show that in a product state, the expectation value of any component of 𝜎̅ or 𝜏̅ is the same as it
would be in the individual single-spin states.
Exercise 6.6
Assume Charlie has prepared the two spins in the singlet state. This time, Bob measures 𝜏𝑦 and
Alice measures 𝜎𝑥 . What is the expectation value of 𝜎𝑥 𝜏𝑦 ?
What does this say about the correlation between the two measurements?
1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
1
⟨𝑠𝑖𝑛𝑔| = (⟨𝑢𝑑| − ⟨𝑑𝑢|)
√2
page 6 of 105
Exercise 6.7
Next (after Charlie has had prepared the two spins in the singlet state), Charlie prepares the spins
in a different state, called |𝑇1 ⟩, where
1
|𝑇1 ⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩)
√2
T stands for triplet. The triplet states are completely different from the states in the coin and die
examples.
What are the expectation values of the operators 𝜎𝑧 𝜏𝑧 , 𝜎𝑥 𝜏𝑥 , and 𝜎𝑦 𝜏𝑦 ?
Notice what a difference a sign can make.
Exercise 6.8
Calculate the expectation values of the operators 𝜎𝑧 𝜏𝑧 , 𝜎𝑥 𝜏𝑥 , and 𝜎𝑦 𝜏𝑦 for the other two
entangled triplet states:
1
|𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)
√2
1
|𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩)
√2
(Addendum: at the end of this exercise we check the expectation values for the singlet state too.)
Exercise 6.9
Prove that the four vectors |𝑠𝑖𝑛𝑔⟩, |𝑇1 ⟩, |𝑇2 ⟩, |𝑇3 ⟩ are eigenvectors of 𝜎 ∙ 𝜏.
What are their eigenvalues?
1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
1 1 1
|𝑇1 ⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩) |𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩) |𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩)
√2 √2 √2
Exercise 6.10
A system of two spins has the Hamiltonian:
𝜔ℏ
ℋ= 𝜎∙𝜏
2
Question 1)
What are the possible energies of the system, and what are the eigenvectors of the Hamiltonian?
Question 2)
Suppose the system starts in the state |𝑢𝑢⟩.
What is the state at any later time?
Answer the same question for initial states of |𝑢𝑑⟩, |𝑑𝑢⟩, |𝑑𝑑.
Exercise 7.1
Write the tensor product 𝐼 ⨂ 𝜏𝑥 as a matrix, and apply that matrix to each of the |𝑢𝑢⟩, |𝑢𝑑⟩, |𝑑𝑢⟩
and |𝑑𝑑⟩ column vectors. Show that Alice’s half of the state-vector is unchanged in each case.
Recall that 𝐼 is the 2 × 2 unit matrix.
Exercise 7.2
Calculate the matrix elements of 𝜎𝑧 ⨂ 𝜏𝑥 by forming the inner product analog to:
⟨𝑢𝑢|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑑𝑑⟩
⟨𝑢𝑑|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑑𝑑⟩
𝜎𝑧 ⨂ 𝐼 = ( )
⟨𝑑𝑢|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑑𝑑⟩
⟨𝑑𝑑|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑑𝑑⟩
𝜎𝑧 operates to the left, 𝐼 to the right.
page 7 of 105
Exercise 7.3
Consider the equation:
(𝐴 ⨂ 𝐵)(𝑎 ⨂ 𝑏) = (𝐴𝑎 ⨂ 𝐵𝑏)
𝐴 and 𝐵 represent 2 × 2 matrices (or operators), and 𝐴 and 𝑏 represent 2 × 1 column vectors.
a) Rewrite the equation:

(𝐴 ⨂ 𝐵)(𝑎 ⨂ 𝑏) = (𝐴𝑎 ⨂ 𝐵𝑏)
in component form, replacing the symbols 𝐴, 𝐵, 𝑎, 𝑏 with the following matrices and column
vectors:
𝐴11 𝐵11 𝐴11 𝐵12 𝐴12 𝐵11 𝐴12 𝐵12

𝐴11 𝐵21 𝐴11 𝐵22 𝐴12 𝐵21 𝐴12 𝐵22
𝐴⨂𝐵 =( )
𝐴21 𝐵11 𝐴21 𝐵12 𝐴22 𝐵11 𝐴22 𝐵12
𝐴21 𝐵21 𝐴21 𝐵22 𝐴22 𝐵21 𝐴22 𝐵22
𝑎11 𝑏11
𝑎11 𝑏11 𝑎11 𝑏21
(𝑎 ) ⨂ ( ) = ( )
21 𝑏21 𝑎21 𝑏11
𝑎21 𝑏21
b) Perform the matrix multiplication 𝐴𝑎 and 𝐵𝑏 on the right-hand side. Verify that each result is a
4 × 1 matrix.
c) Expand all three Kronecker products.
d) Verify the row and column sizes of each Kronecker product:

• 𝐴⨂𝐵:4 × 4
• 𝑎⨂𝑏:4 × 1
• 𝐴𝑎 ⨂ 𝐵𝑏 : 4 × 1
e) Perform the matrix multiplication on the left-hand side, resulting in a 4 × 1 column vector. Each
row should be the sum of four separate terms
f) Finally, verify that the resulting column vectors on the left and right sides are identical.
Exercise 7.4
Calculate the density matrix for:
|Ψ⟩ = 𝛼|𝑢𝑢⟩ + 𝛽|𝑢𝑢⟩
Answer:
𝜓(𝑢) = 𝛼; 𝜓 ∗ (𝑢) = 𝛼 ∗
𝜓(𝑑) = 𝛽; 𝜓 ∗ (𝑑) = 𝛽 ∗
𝛼 ∗𝛼 𝛼 ∗𝛽
𝜌𝑎′𝑎 = ( )
𝛽∗𝛼 𝛽∗𝛽
Now try plugging some numbers for 𝛼 and 𝛽. Make sure they are normalized to 1. For example,
1 1
𝛼 = 2,𝛽 = 2
√ √
page 8 of 105
Exercise 7.5
a) Show:
𝑎 0 2 2
( ) = (𝑎 0)
0 𝑏 0 𝑏2
b) Now, suppose:
1⁄ 0
𝜌=( 3 )
0 2⁄3
Calculate:
𝜌2 , 𝑇𝑟(𝜌), 𝑇𝑟(𝜌2 )
c) If 𝜌 is a density matrix, does it represent a pure state or a mixed state?
Exercise 7.6
By the standard rules of probability, the probability for a:
𝑃(𝑎) = ∑ 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎, 𝑏)
𝑏
This is just a diagonal entry in the density matrix:
𝑃(𝑎) = 𝜌𝑎𝑎
Use 𝑃(𝑎) = 𝜌𝑎𝑎 to show that if 𝜌 is a density matrix, then:

𝑇𝑟(𝜌) = 1
Exercise 7.7
We have the density matrix:
1⁄ 0
𝜌=( 2 )
0 1⁄2
Calculate 𝜌2 . How does this result confirm that 𝜌 represents an entangled state?
Exercise 7.8
Consider the following states:
1
|𝜓1 ⟩ = (|𝑢𝑢⟩ + |𝑢𝑑⟩ + |𝑑𝑢⟩ + |𝑑𝑑⟩)
2
1
|𝜓2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)
√2
1
|𝜓3 ⟩ = (3|𝑢𝑢⟩ + 4|𝑑𝑑⟩)
5
For each one, calculate Alice’s density matrix and Bob’s density matrix. Check their properties.
Exercise 7.9
Given any Alice observable 𝐴 and Bo observable 𝐵, show that for a product state, the correlation
𝐶(𝐴, 𝐵) is zero.
Exercise 7.10
Given a measuring apparatus with the states |𝑏⟩ for initial blank state, |+1⟩ for “result of spin
measurement is up” and |−1⟩ for “result of spin measurement is down”.
Verify that the state-vector
αu |u, b⟩ + αd |d, b⟩
represents a completely unentangled state.
page 9 of 105
Exercise 7.11
Calculate Alice’s density matrix for 𝜎𝑧 for the “near-singlet” state.
Exercise 7.12
Verify the numerical values in each rap sheet.
This is a very specific exercise that refers to so called “rap sheets”. The solution here is for reasons
of completeness only. It is a superset of the solution to exercise 7.11.
Exercise 8.1
Prove that the position operator 𝑋 and the momentum operator 𝐷 are linear operators.
𝑿: 𝑿𝜑(𝑥) = 𝑥𝜑(𝑥)
𝑑𝜑(𝑥)
𝑫: 𝑫𝜑(𝑥) =
𝑑𝑥
Exercise 9.1
Applicate the Hamiltonian
ℏ2 𝜕 2 𝜓(𝑥)
− = 𝐸𝜓(𝑥)
2𝑚 𝜕𝑥 2
to the wave function:
𝑖𝑝
𝜓(𝑥) = 𝑒 ℏ 𝑥
Show that this wave function is a solution, if we set:
𝑝2
𝐸=
2𝑚
Exercise 9.2
Prove the following equation by expanding each side and comparing the results:
[𝑃2 , 𝑋] = 𝑃[𝑃, 𝑋] + [𝑃, 𝑋]𝑃
𝑃 is the momentum operator, 𝑋 is the position operator – both are matrices.
[𝑃, 𝑋] is the commutator relation: [𝑃, 𝑋] = 𝑃𝑋 − 𝑋𝑃
Exercise 9.3
Show that the right-hand side of
𝑑 𝑑
[𝑉(𝑥), 𝑃]𝜓(𝑥) = 𝑉(𝑥) (−𝑖ℏ ) 𝜓(𝑥) − (−𝑖ℏ ) 𝑉(𝑥)𝜓(𝑥)
𝑑𝑥 𝑑𝑥
simplifies to the right-hand side of:
𝑑𝑉(𝑥)
[𝑉(𝑥), 𝑃]𝜓(𝑥) = 𝑖ℏ
𝑑𝑥
Hint: First expand the second term by taking the derivative of the product.
Then look for cancellations.
Exercise 10.1
Find the second derivative of 𝑥:
𝑥 = 𝐴 ∙ 𝑐𝑜𝑠(𝜔𝑡) + 𝐵 ∙ 𝑠𝑖𝑛(𝜔𝑡)
Show thereby that it solves:
−𝜔2 𝑥 = 𝑥̈
page 10 of 105
Exercise 1.1
a) Using the axioms for inner products, prove {⟨𝐴| + ⟨𝐵|}|𝐶⟩ = ⟨𝐴|𝐶⟩ + ⟨𝐵|𝐶⟩.
b) Prove that ⟨𝐴|𝐴⟩ is a real number.
**********
a)
{⟨𝐴| + ⟨𝐵|}|𝐶⟩ =
[⟨𝐶|{|𝐴⟩ + |𝐵⟩}]∗ =
⟨𝐶|𝐴⟩∗ + ⟨𝐶|𝐵⟩∗ =
⟨𝐴|𝐶⟩∗∗ + ⟨𝐵|𝐶⟩∗∗ =
⟨𝐴|𝐶⟩ + ⟨𝐵|𝐶⟩
b)
𝑎1
𝑎
⟨𝐴|𝐴⟩ = (𝑎1∗ 𝑎2∗ … 𝑎𝑛∗ 1 ) ( …2 ) =
𝑎𝑛
𝑎1∗ 𝑎1 + 𝑎2∗ 𝑎2 + ⋯ + 𝑎𝑛∗ 𝑎𝑛 =

|𝑎1 |2 + |𝑎2 |2 + ⋯ + |𝑎𝑛 |2 ∈ ℝ
page 11 of 105
Exercise 1.2
Show that the inner product defined by Eq. 1.2 satisfies all the axioms of inner products.
Inner product:
𝑏1
𝑏
⟨𝐴|𝐵⟩ = (𝑎1∗ 𝑎2∗ … 𝑎𝑛∗ 1 ) ( 2 ) = 𝑎1∗ 𝑏1 + 𝑎2∗ 𝑏2 + ⋯ + 𝑎𝑛∗ 𝑏𝑛
…
𝑏𝑛
Linearity: ⟨𝐶|{|𝐴⟩ + |𝐵⟩} = ⟨𝐶|𝐴⟩ + ⟨𝐶|𝐵⟩
Interchange: ⟨𝐵|𝐴⟩ = ⟨𝐴|𝐵⟩∗
**********
Linearity:
⟨𝐶|{|𝐴⟩ + |𝐵⟩} =
𝑎1 + 𝑏1
𝑎 + 𝑏2
(𝑐1∗ 𝑐2∗ … 𝑐𝑛∗ 1 ) ( 2 )=
…
𝑎𝑛 + 𝑏𝑛
𝑐1∗ (𝑎1 + 𝑏1 ) + 𝑐2∗ (𝑎2 + 𝑏2 ) + ⋯ + 𝑐𝑛∗ (𝑎𝑛 + 𝑏𝑛 ) =
𝑐1∗ 𝑎1 + 𝑐1∗ 𝑏1 + 𝑐2∗ 𝑎2 + 𝑐2∗ 𝑏2 + ⋯ + 𝑐𝑛∗ 𝑎𝑛 + 𝑐𝑛∗ 𝑏𝑛 =
𝑐1∗ 𝑎1 + 𝑐2∗ 𝑎2 + ⋯ + 𝑐𝑛∗ 𝑎𝑛 + 𝑐1∗ 𝑏1 + 𝑐2∗ 𝑏2 + ⋯ + 𝑐𝑛∗ 𝑏𝑛 =
𝑎1 𝑏1
𝑎 2 𝑏
(𝑐1∗ 𝑐2∗ … 𝑐𝑛∗ 1 ) ( … ) + (𝑐1∗ 𝑐2∗ … 𝑐𝑛∗ 1 ) ( 2 ) =
…
𝑎𝑛 𝑏𝑛
⟨𝐶|𝐴⟩ + ⟨𝐶|𝐵⟩
Interchange:
𝑏1
𝑏
⟨𝐴|𝐵⟩∗ = [(𝑎1∗ 𝑎2∗ … 𝑎𝑛∗ 1 ) ( 2 )] =∗
…
𝑏𝑛
(𝑎1∗ 𝑏1 + 𝑎2∗ 𝑏2 + ⋯ + 𝑎𝑛∗ 𝑏𝑛 )∗ =
𝑎1∗∗ 𝑏1∗ + 𝑎2∗∗ 𝑏2∗ + ⋯ + 𝑎𝑛∗∗ 𝑏𝑛∗ =

𝑎1 𝑏1∗ + 𝑎2 𝑏2∗ + ⋯ + 𝑎𝑛 𝑏𝑛∗ =
𝑏1∗ 𝑎1 + 𝑏2∗ 𝑎2 + ⋯ + 𝑏𝑛∗ 𝑎𝑛 =
𝑎1
𝑎
(𝑏1∗ 𝑏2∗ … 𝑏𝑛∗ ) ( …2 ) =
𝑎𝑛
⟨𝐵|𝐴⟩
page 12 of 105
Exercise 2.1
Prove that the vector |𝑟⟩ is orthogonal to vector |𝑙⟩:

1 1
|𝑟⟩ = |𝑢⟩ + |𝑑⟩
√2 √2
1 1
|𝑙⟩ = |𝑢⟩ − |𝑑⟩
√2 √2
**********
If |𝑟⟩ is orthogonal to |𝑙⟩, then the scalar product ⟨𝑟|𝑙⟩ must be 0.
⟨𝑟|𝑙⟩ =
1 1 1 1
(⟨𝑢| + ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 1 1 1 1 1
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 1 1 1
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
2 2 2 2
1 1
−0+0− =0
2 2
page 13 of 105
Exercise 2.2
Prove that |𝑖⟩ and |𝑜⟩ satisfy all the conditions in Eqs. 2.7, 2.8, and 2.9.
Are they unique in that respect?
Eq. 2.7:
⟨𝑖|𝑜⟩ = 0
Eqs. 2.8:
1 1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ = ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ =
2 2
Eqs. 2.9
1 1
⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩ = ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩ =
2 2
1 𝑖 1 𝑖
|𝑖⟩ = |𝑢⟩ + |𝑑⟩ and ⟨𝑖| = ⟨𝑢| − ⟨𝑑|
√2 √2 √2 √2
1 𝑖 1 𝑖
|𝑜⟩ = |𝑢⟩ − 2 |𝑑⟩ and ⟨𝑜| = ⟨𝑢| + ⟨𝑑|
√2 √ √2 √2
**********
Eq. 2.7: ⟨𝑖|𝑜⟩ = 0
⟨𝑖|𝑜⟩ =
1 𝑖 1 𝑖
(⟨𝑢| − ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 𝑖 𝑖 1 𝑖 𝑖
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 𝑖 𝑖 1
∙1− ∙0− ∙0− ∙1=0
2 2 2 2
Eqs. 2.8 ⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩:

1 𝑖
⟨𝑜|𝑢⟩ = (⟨𝑢| + ⟨𝑑| ) (|𝑢⟩) =
√2 √2
1 𝑖
( ⟨𝑢|𝑢⟩ + ⟨𝑑|𝑢⟩) =
√2 √2
1 1
( + 0) =
√2 √2
page 14 of 105
1 𝑖
⟨𝑢|𝑜⟩ = (⟨𝑢|) ( |𝑢⟩ − |𝑑⟩) =
√2 √2
1 𝑖
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ =
√2 √2
1 1
−0=
√2 √2
1 1 1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = =
√2 √2 2
Eqs. 2.8 ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩:

1 𝑖
⟨𝑜|𝑑⟩ = (⟨𝑢| + ⟨𝑑| ) (|𝑑⟩) =
√2 √2
1 𝑖
( ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑑⟩) =
√2 √2
𝑖 𝑖
(0 + )=
√2 √2
1 𝑖
⟨𝑑|𝑜⟩ = (⟨𝑑|) ( |𝑢⟩ − |𝑑⟩) =
√2 √2
1 𝑖
⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2
𝑖 𝑖
0− =−
√2 √2
𝑖 −𝑖 1
⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ = =
√2 √2 2
Eqs. 2.8 ⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩:

1 𝑖
⟨𝑖|𝑢⟩ = (⟨𝑢| − ⟨𝑑| ) (|𝑢⟩) =
√2 √2
1 𝑖
( ⟨𝑢|𝑢⟩ − ⟨𝑑|𝑢⟩) =
√2 √2
1 1
( − 0) =
√2 √2
page 15 of 105
1 𝑖
⟨𝑢|𝑖⟩ = (⟨𝑢|) ( |𝑢⟩ + |𝑑⟩) =
√2 √2
1 𝑖
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ =
√2 √2
1 1
−0=
√2 √2
1 1 1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ = =
√2 √2 2
Eqs. 2.8 ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩:

1 𝑖
⟨𝑖|𝑑⟩ = (⟨𝑢| − ⟨𝑑| ) (|𝑑⟩) =
√2 √2
1 𝑖
( ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑑⟩) =
√2 √2
𝑖 −𝑖
(0 − )=
√2 √2
1 𝑖
⟨𝑑|𝑖⟩ = (⟨𝑑|) ( |𝑢⟩ + |𝑑⟩) =
√2 √2
1 𝑖
⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2
𝑖 −𝑖
0− =
√2 √2
−𝑖 −𝑖 1
⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ = =
√2 √2 2
Eqs. 2.9 ⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩:

1 𝑖 1 𝑖
|𝑖⟩ = |𝑢⟩ + |𝑑⟩ and ⟨𝑖| = ⟨𝑢| − ⟨𝑑|
√2 √2 √2 √2
1 𝑖 1 𝑖
|𝑜⟩ = |𝑢⟩ − |𝑑⟩ and ⟨𝑜| = ⟨𝑢| + ⟨𝑑|
√2 √2 √2 √2
1 1 1 1
|𝑟⟩ = 2
|𝑢⟩ + 2 |𝑑⟩ and ⟨𝑟| = ⟨𝑢| + ⟨𝑑|
√ √ √2 √2
1 1 1 1
|𝑙⟩ = |𝑢⟩ − 2 |𝑑⟩ and ⟨𝑙| = ⟨𝑢| − ⟨𝑑|
√2 √ √2 √2
page 16 of 105
Eqs. 2.9 ⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩:

1 𝑖 1 1
⟨𝑜|𝑟⟩ = (⟨𝑢| + ⟨𝑑| )( |𝑢⟩ + |𝑑⟩) =
√2 √2 √2 √2
1 1 1 1 𝑖 1 𝑖 1
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 1 𝑖 𝑖
∙1+ ∙0+ ∙0+ ∙1=
2 2 2 2
1+𝑖
2
1 1 1 𝑖
⟨𝑟|𝑜⟩ = (⟨𝑢| + ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 𝑖 1 1 1 𝑖
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 𝑖 1 𝑖
∙1− ∙0+ ∙0− ∙1=
2 2 2 2
1−𝑖
2
1 + 𝑖 1 − 𝑖 1 − (𝑖 ∙ 𝑖)2 2 1
⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩ = ∙ = = =
2 2 4 4 2
Eqs. 2.9 ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩:

1 𝑖 1 1
⟨𝑜|𝑙⟩ = (⟨𝑢| + ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 1 𝑖 1 𝑖 1
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 1 𝑖 𝑖
∙1− ∙0+ ∙0− ∙1=
2 2 2 2
1−𝑖
2
1 1 1 𝑖
⟨𝑙|𝑜⟩ = (⟨𝑢| − ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 𝑖 1 1 1 𝑖
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 𝑖 1 𝑖
∙1− ∙0+ ∙0+ ∙1=
2 2 2 2
1+𝑖
2
1 − 𝑖 1 + 𝑖 1 − (𝑖 ∙ 𝑖)2 2 1
⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩ = ∙ = = =
2 2 4 4 2
page 17 of 105
Eqs. 2.9 ⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩:

1 𝑖 1 1
⟨𝑖|𝑟⟩ = (⟨𝑢| − ⟨𝑑| )( |𝑢⟩ + |𝑑⟩) =
√2 √2 √2 √2
1 1 1 1 𝑖 1 𝑖 1
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 1 𝑖 𝑖
∙1+ ∙0− ∙0− ∙1=
2 2 2 2
1−𝑖
2
1 1 1 𝑖
⟨𝑟|𝑖⟩ = (⟨𝑢| + ⟨𝑑| )( |𝑢⟩ + |𝑑⟩) =
√2 √2 √2 √2
1 1 1 𝑖 1 1 1 𝑖
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 𝑖 1 𝑖
∙1+ ∙0+ ∙0+ ∙1=
2 2 2 2
1+𝑖
2
1 − 𝑖 1 + 𝑖 1 − (𝑖 ∙ 𝑖)2 2 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ∙ = = =
2 2 4 4 2
Eqs. 2.9 ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩:

1 𝑖 1 1
⟨𝑖|𝑙⟩ = (⟨𝑢| − ⟨𝑑| )( |𝑢⟩ − |𝑑⟩) =
√2 √2 √2 √2
1 1 1 1 𝑖 1 𝑖 1
⟨𝑢|𝑢⟩ − ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 1 𝑖 𝑖
∙1− ∙0− ∙0+ ∙1=
2 2 2 2
1+𝑖
2
1 1 1 𝑖
⟨𝑙|𝑖⟩ = (⟨𝑢| − ⟨𝑑| )( |𝑢⟩ + |𝑑⟩) =
√2 √2 √2 √2
1 1 1 𝑖 1 1 1 𝑖
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ − ⟨𝑑|𝑢⟩ − ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2 √2 √2 √2 √2
1 𝑖 1 𝑖
∙1+ ∙0− ∙0− ∙1=
2 2 2 2
1−𝑖
2
1 + 𝑖 1 − 𝑖 1 − (𝑖 ∙ 𝑖)2 2 1
⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩ = ∙ = = =
2 2 4 4 2
page 18 of 105
Are |𝑖⟩ and |𝑜⟩ unique in that respect?
I think that this question goes to the phase ambiguity. |𝑖⟩ and |𝑜⟩ can be multiplied by any z of the
form 𝑧 = 𝑒 𝑖𝜃 without disturbing the relationships 2.7, 2.8 and 2.9.
page 19 of 105
Exercise 2.3
Assume that |𝑖⟩ and |𝑜⟩ are given as:
|𝑖⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩ and ⟨𝑖| = ⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗
|𝑜⟩ = 𝛾|𝑢⟩ − 𝛿|𝑑⟩ and ⟨𝑜| = ⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗
a) Use Eqs. 2.8 to show that

1
𝛼 ∗𝛼 = 𝛽∗𝛽 = 𝛾 ∗𝛾 = 𝛿 ∗𝛿 =
2
b) Use the above result and Eqs. 2.9 to show that:
𝛼 ∗ 𝛽 + 𝛼𝛽 ∗ = 0
Eqs. 2.8:
1 1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ = ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ =
2 2
Eqs. 2.9
1 1
⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩ = ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩ =
2 2
1 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩ =
2 2
**********
1
a) Use Eqs. 2.8 to show that 𝛼 ∗ 𝛼 = 𝛽 ∗ 𝛽 = 𝛾 ∗ 𝛾 = 𝛿 ∗ 𝛿 = 2
1
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ =
2
⟨𝑜|𝑢⟩ = (⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗ )|𝑢⟩ =
𝛾 ∗ ⟨𝑢|𝑢⟩ − 𝛿 ∗ ⟨𝑑|𝑢⟩ =
𝛾∗ ∙ 1 − 𝛿∗ ∙ 0 = 𝛾∗
⟨𝑢|𝑜⟩ = ⟨𝑢|(𝛾|𝑢⟩ − 𝛿|𝑑⟩) =
𝛾⟨𝑢|𝑢⟩ − 𝛿⟨𝑢|𝑑⟩ =
𝛾∙1−𝛿∙0= 𝛾
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = 𝛾 ∗ 𝛾
1
According to equation 2.8 this gives 𝛾 ∗ 𝛾 = .
2
1
⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ =
2
page 20 of 105
⟨𝑜|𝑑⟩ = (⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗ )|𝑑⟩ =
𝛾 ∗ ⟨𝑢|𝑑⟩ − 𝛿 ∗ ⟨𝑑|𝑑⟩ =
𝛾 ∗ ∙ 0 − 𝛿 ∗ ∙ 1 = −𝛿 ∗
⟨𝑑|𝑜⟩ = ⟨𝑑|(𝛾|𝑢⟩ − 𝛿|𝑑⟩) =
𝛾⟨𝑑|𝑢⟩ − 𝛿⟨𝑑|𝑑⟩ =
𝛾 ∙ 0 − 𝛿 ∙ 1 = −𝛿
⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩ = 𝛿 ∗ 𝛿
1
According to equation 2.8 this gives 𝛿 ∗ 𝛿 = 2.
1
⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩ =
2
⟨𝑖|𝑢⟩ = (⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑢⟩ =
𝛼 ∗ ⟨𝑢|𝑢⟩ + 𝛽 ∗ ⟨𝑑|𝑢⟩ =
𝛼 ∗ ∙ 1 + 𝛽∗ ∙ 0 = 𝛼 ∗
⟨𝑢|𝑖⟩ = ⟨𝑢|(𝛼|𝑢⟩ + 𝛽|𝑑⟩) =
𝛼⟨𝑢|𝑢⟩ + 𝛽⟨𝑢|𝑑⟩ =
𝛼∙1+𝛽∙0=𝛼
⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩ = 𝛼 ∗ 𝛼
1
According to equation 2.8 this gives 𝛼 ∗ 𝛼 = 2.
1
⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ =
2
⟨𝑖|𝑑⟩ = (⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑑⟩ =
𝛼 ∗ ⟨𝑢|𝑑⟩ + 𝛽 ∗ ⟨𝑑|𝑑⟩ =
𝛼 ∗ ∙ 0 + 𝛽∗ ∙ 1 = 𝛽∗
⟨𝑑|𝑖⟩ = ⟨𝑑|(𝛼|𝑢⟩ + 𝛽|𝑑⟩) =
𝛼⟨𝑑|𝑢⟩ + 𝛽⟨𝑑|𝑑⟩ =
𝛼∙0+𝛽∙1= 𝛽
⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩ = 𝛽 ∗ 𝛽
1
According to equation 2.8 this gives 𝛽 ∗ 𝛽 = 2.
b) Use the above result and Eqs. 2.9 to show that
𝛼 ∗ 𝛽 + 𝛼𝛽 ∗ = 0
page 21 of 105
|𝑖⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩ and ⟨𝑖| = ⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗

1 1 1 1
|𝑟⟩ = 2
|𝑢⟩ + 2 |𝑑⟩ and ⟨𝑟| = ⟨𝑢| + ⟨𝑑|
√ √ √2 √2
1 1
⟨𝑖|𝑟⟩ = (⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ ) ( |𝑢⟩ + |𝑑⟩) =
√2 √2
𝛼∗ 𝛼∗ 𝛽∗ 𝛽∗
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2
𝛼∗ 𝛼∗ 𝛽∗ 𝛽∗
∙1+ ∙0+ ∙0+ ∙1=
√2 √2 √2 √2
𝛼∗ 𝛽∗
+
√2 √2
1 1
⟨𝑟|𝑖⟩ = ( |𝑢⟩ + |𝑑⟩) (⟨𝑢|𝛼 + ⟨𝑑|𝛽) =
√2 √2
𝛼 𝛼 𝛽 𝛽
⟨𝑢|𝑢⟩ + ⟨𝑢|𝑑⟩ + ⟨𝑑|𝑢⟩ + ⟨𝑑|𝑑⟩ =
√2 √2 √2 √2
𝛼 𝛼 𝛽 𝛽
∙1+ ∙0+ ∙0+ ∙1=
√2 √2 √2 √2
𝛼 𝛽
+
√2 √2
1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ =
2
𝛼∗ 𝛽∗ 𝛼 𝛽
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = ( + )( + )=
√2 √2 √2 √2
1 1
(𝛼 ∗ + 𝛽 ∗ ) (𝛼 + 𝛽) =
√2 √2
1 ∗
(𝛼 + 𝛽 ∗ )(𝛼 + 𝛽) =
2
1 ∗
(𝛼 𝛼 + 𝛼 ∗ 𝛽 + 𝛽 ∗ 𝛼 + 𝛽 ∗ 𝛽) =
2
1 1
⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩ = (𝛼 ∗ 𝛼 + 𝛽 ∗ 𝛽 + 𝛼 ∗ 𝛽 + 𝛽 ∗ 𝛼) =
2 2
page 22 of 105
1
From part a) we know that 𝛼 ∗ 𝛼 = 𝛽 ∗ 𝛽 =
2
1 1 1 1
( + + 𝛼 ∗ 𝛽 + 𝛽 ∗ 𝛼) =
2 2 2 2
1 1
+ + 𝛼 ∗𝛽 + 𝛽∗𝛼 = 1
2 2
1 + 𝛼 ∗𝛽 + 𝛽∗𝛼 = 1
𝛼 ∗𝛽 + 𝛽∗𝛼 = 0
Let 𝛼 = 𝑟 + 𝑖𝑠, 𝛼 ∗ = 𝑟 − 𝑖𝑠, 𝛽 = 𝑡 + 𝑖𝑢, 𝛽 ∗ = 𝑡 − 𝑖𝑢.
𝛼 ∗ 𝛽 + 𝛽 ∗ 𝛼 = 0 → (𝑟 − 𝑖𝑠)(𝑡 + 𝑖𝑢) + (𝑟 + 𝑖𝑠)(𝑡 − 𝑖𝑢) = 0

(𝑟𝑡 + 𝑖𝑟𝑢 − 𝑖𝑠𝑡 + 𝑠𝑢) + (𝑟𝑡 − 𝑖𝑟𝑢 + 𝑖𝑠𝑡 + 𝑠𝑢) = 0
2𝑟𝑡 + 2𝑠𝑢 = 0
𝑟𝑡 + 𝑠𝑢 = 0
𝑟𝑡 = −𝑠𝑢
𝑠𝑢
𝑟=−
𝑡
𝑠𝑢
𝛼 ∗ 𝛽 = (𝑟 − 𝑖𝑠)(𝑡 + 𝑖𝑢) = (− − 𝑖𝑠) (𝑡 + 𝑖𝑢) =
𝑡
𝑖𝑠𝑢2
−𝑠𝑢 − − 𝑖𝑠𝑡 + 𝑠𝑢 =
𝑡
𝑖𝑠𝑢2 𝑠𝑢2 𝑠𝑢2 + 𝑠𝑡 2
− − 𝑖𝑠𝑡 = −𝑖 ( + 𝑠𝑡) = −𝑖 ( )
𝑡 𝑡 𝑡
page 23 of 105
Exercise 3.1
Prove: If a vector space is N-dimensional, an orthonormal basis of N vectors can be constructed from
the eigenvectors of a Hermitian operator.
**********
The eigenvectors of a Hermitian operator form a complete set of linear independent vectors, out of
these can be constructed an orthogonal basis and, by dividing through their length an orthonormal
basis. This holds for finite dimensional vector spaces.
For the example of the R³ we want to show how to construct the change of basis vectors.
1 1 0
Let (0) , (1) 𝑎𝑛𝑑 (1) be a set of vectors B1, B2 and B3 that form a basis B of R³.
1 0 1
They are linear independent:
1 1 0 0
𝑎 (0) + 𝑏 (1) + 𝑐 (1) = (0)
1 0 1 0
1 1 00
0 1 1|0
1 0 10
We transform this:
1 1 00 1 0 −1 0 2 0 00 2 0 00 2 0 00
0 1 1|0 → 0 1 1 |0 → 0 1 1|0 → 0 1 1|0 → 0 1 0|0
1 0 10 1 0 1 0 1 0 10 0 0 10 0 0 10
which makes clear that the only solution to this is a = b = c = 0.
The linear independent vectors define a matrix P, a linear map:

1 1 0
𝑃 ≔ (0 1 1)
1 0 1
𝑥1
Let (𝑥2 ) be a vector with respect to this basis B: 𝑥 = 𝑥1 𝐵1 + 𝑥2 𝐵2 + 𝑥3 𝐵3 .
𝑥3
𝑥1 1 0 0
Then 𝑃 (𝑥2 ) give the coordinates of 𝑥 in the canonical basis E1, E2 and E3: (0) , (1) 𝑎𝑛𝑑 (0).
𝑥3 0 0 1
𝑥1 1 1 0 𝑥1 𝑥1 + 𝑥2
𝑥
𝑃 ( 2 ) = (0 1 𝑥
1) ( 2 ) = ( 2 + 𝑥3 )
𝑥
𝑥3 1 0 1 𝑥3 𝑥1 + 𝑥3
Check:
1 1 1 0 1 1
𝑃 (0) = (0 1 1) (0) = (0)
0 1 0 1 0 1
0 1 1 0 0 1
𝑃 (1) = (0 1 1) (1) = (1)
0 1 0 1 0 0
page 24 of 105
0 1 1 0 0 0
𝑃 (0) = (0 1 1) (0) = (1)
1 1 0 1 1 1
To get this the other way around we must find the inverse matrix 𝑃−1 .
1 1 0 1 0 0
(0 1 1) | ( 0 1 0) line 3 minus line 1
1 0 1 0 0 1
1 1 0 1 0 0
(0 1 1) | ( 0 1 0) line 3 plus line 2
0 −1 1 −1 0 1
1 1 0 1 0 0
(0 1 1) | ( 0 1 0) line 2 minus line 3/2
0 0 2 −1 1 1
1 1 1
1 0 0 −
2 2 2
(0 1 0) | 1 1 1 line 3 divided by 2
0 0 2 −
2 2 2
(−1 1 1 )
1 1 1
−
2 2 2
1 0 0 1 1 1
(0 1 0) | −
0 0 1 2 2 2
1 1 1
(− 2 2 2 )
Our inverse matrix 𝑃−1 :
1 1 −1 1
(1 1 −1)
2
−1 1 1
Applied to the linear independent vectors 𝐵1 , 𝐵2 , 𝐵3 this must give the canonical basis 𝐸1 , 𝐸2 , 𝐸3 .
Check:
1 1 1 −1 1 1 1 2 1
𝑃−1 (0) = ( 1 1 −1) (0) = (0) = (0)
2 2
1 −1 1 1 1 0 0
1 1 1 −1 1 1 1 0 0
𝑃−1 (1) = ( 1 1 −1) (1) = (2) = (1)
2 2
0 −1 1 1 0 0 0
0 1 1 −1 1 0 1 0 0
𝑃−1 (1) = ( 1 1 −1) (1) = (0) = (0)
2 2
1 −1 1 1 1 2 1
page 25 of 105
Exercise 3.2
Prove that 𝜎𝑧 of Eq. 3.16 is the unique solution to Eqs. 3.14 und 3.15.
Eq. 3.16
(𝜎 ) (𝜎𝑧 )12 1 0
𝜎𝑧 = ( 𝑧 11 )=( )
(𝜎𝑧 )21 (𝜎𝑧 )21 0 −1
Eq. 3.14
(𝜎 ) (𝜎𝑧 )12 1 1
( 𝑧 11 )( ) = ( )
(𝜎𝑧 )21 (𝜎𝑧 )21 0 0
Eq. 3.15
(𝜎 ) (𝜎𝑧 )12 0 0
( 𝑧 11 )( ) = −( )
(𝜎𝑧 )21 (𝜎𝑧 )21 1 1
**********
Eq. 3.14
𝑎 𝑏 1 𝑎
( ) ( ) = ( ) → 𝑎 = 1; 𝑐 = 0
𝑐 𝑑 0 𝑐
Eq. 3.15
𝑎 𝑏 0 𝑏
( ) ( ) = ( ) → 𝑏 = 0; 𝑑 = −1
𝑐 𝑑 1 𝑑
The result is the matrix
𝑎 𝑏 1 0
( )=( )
𝑐 𝑑 0 −1
Proof by contradiction eq. 3.14:

𝑎
let ( ) with 𝑎 ≠ 1 and 𝑏 ≠ 0
𝑏
1 0 𝑎 1
and ( )( ) = ( )
0 −1 𝑏 0
1 ∙ 𝑎 + 0 ∙ 𝑏 = 1 → 𝑎 = 1 𝑐𝑜𝑛𝑡𝑟𝑎𝑑𝑖𝑐𝑡𝑖𝑜𝑛
0 ∙ 𝑎 − 1 ∙ 𝑏 = 0 → 𝑏 = 0 𝑐𝑜𝑛𝑡𝑟.
The same holds for eq. 3.15:
𝑎
let ( ) with 𝑎 ≠ 0 and 𝑏 ≠ 1
𝑏
1 0 𝑎 0
and ( )( ) = ( )
0 −1 𝑏 −1
1 ∙ 𝑎 + 0 ∙ 𝑏 = 0 → 𝑎 = 0 𝑐𝑜𝑛𝑡𝑟.
0 ∙ 𝑎 − 1 ∙ 𝑏 = −1 → 𝑏 = 1 𝑐𝑜𝑛𝑡𝑟.
page 26 of 105
Exercise 3.3
Calculate the eigenvectors and eigenvalues of 𝜎𝑛 . Assume the eigenvector 𝜆1 has the form:
cos 𝛼
( )
sin 𝛼
𝛼 is an unknown parameter. Plug this vector into the eigenvalue equation and solve for 𝛼 in terms of
𝜃. Why did we use a single parameter 𝛼? Notice that our suggested column vector must have unit
length.
𝑐𝑜𝑠 𝜃 sin 𝜃
𝜎𝑛 = ( )
sin 𝜃 − cos 𝜃
To show:
𝑐𝑜𝑠 𝜃 sin 𝜃 cos 𝛼 cos 𝛼
( )( ) = 𝜆1 ( )
sin 𝜃 − cos 𝜃 sin 𝛼 sin 𝛼
**********
𝑐𝑜𝑠 𝜃 sin 𝜃 cos 𝛼 𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼
( )( )=( )
sin 𝜃 − cos 𝜃 sin 𝛼 sin 𝜃 ∙ cos 𝛼 − cos 𝜃 ∙ sin 𝛼
𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼 cos 𝛼
( ) = 𝜆1 ( )
sin 𝜃 ∙ cos 𝛼 − cos 𝜃 ∙ sin 𝛼 sin 𝛼
This must be valid for every coordinate, so we get two equations:
𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼 = 𝜆1 ∙ cos 𝛼

sin 𝜃 ∙ cos 𝛼 − cos 𝜃 ∙ sin 𝛼 = 𝜆1 ∙ sin 𝛼
Trigonometric identities:
𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼 = cos (𝜃 − 𝛼)

sin 𝜃 ∙ cos 𝛼 − cos 𝜃 ∙ sin 𝛼 = sin (𝜃 − 𝛼)
We have:
cos (𝜃 − 𝛼) = 𝜆1 ∙ cos 𝛼
sin (𝜃 − 𝛼) = 𝜆1 ∙ sin 𝛼
𝜃
Sin and cos are nonlinear functions, the only possible solution is: 𝜆1 = 1 and 𝛼 = 2:
𝜃 𝜃
cos cos
𝑐𝑜𝑠 𝜃 sin 𝜃 2) = ( 2)
( )(
sin 𝜃 − cos 𝜃 𝜃 𝜃
sin sin
2 2
For the second eigenvalue/eigenvector we use that both eigenvectors must be orthogonal.
cos 𝛼 𝑎
( )( ) = 0
sin 𝛼 𝑏
𝑎 ∙ cos 𝛼 + b ∙ sin 𝛼 = 0
Again, as sin and cos are nonlinear functions, possible solutions are
𝑎 = − sin 𝛼 ; 𝑏 = cos 𝛼
𝑎 = sin 𝛼 ; 𝑏 = − cos 𝛼
page 27 of 105
We try the first one: 𝑎 = − sin 𝛼 ; 𝑏 = cos 𝛼

𝑐𝑜𝑠 𝜃 sin 𝜃 − sin 𝛼 −𝑐𝑜𝑠 𝜃 ∙ sin 𝛼 + sin 𝜃 ∙ cos 𝛼
( )( )=( )
sin 𝜃 − cos 𝜃 cos 𝛼 − sin 𝜃 ∙ sin 𝛼 − cos 𝜃 ∙ cos 𝛼
−𝑐𝑜𝑠 𝜃 ∙ sin 𝛼 + sin 𝜃 ∙ cos 𝛼 − sin 𝛼
( ) = 𝜆2 ( )
− sin 𝜃 ∙ sin 𝛼 − cos 𝜃 ∙ cos 𝛼 cos 𝛼
Using the trigonometric identities again we get:
sin(𝜃 − 𝛼) = 𝜆2 ∙ (− sin 𝛼)
−cos (𝜃 − 𝛼) = 𝜆2 ∙ cos 𝛼
𝜃
with the solution: 𝜆2 = −1 and again 𝛼 = 2
Check the second solution 𝑎 = sin 𝛼 ; 𝑏 = − cos 𝛼:

𝑐𝑜𝑠 𝜃 sin 𝜃 sin 𝛼 𝑐𝑜𝑠 𝜃 ∙ sin 𝛼 − sin 𝜃 ∙ cos 𝛼
( )( )=( )
sin 𝜃 − cos 𝜃 −cos 𝛼 sin 𝜃 ∙ sin 𝛼 + cos 𝜃 ∙ cos 𝛼
𝑐𝑜𝑠 𝜃 ∙ sin 𝛼 − sin 𝜃 ∙ cos 𝛼 sin 𝛼
( ) = 𝜆2 ( )
sin 𝜃 ∙ sin 𝛼 + cos 𝜃 ∙ cos 𝛼 −cos 𝛼
Using the trigonometric identities again we get:
− sin(𝜃 − 𝛼) = 𝜆2 ∙ sin 𝛼
cos (𝜃 − 𝛼) = 𝜆2 ∙ (−cos 𝛼)
𝜃
with the same solution: 𝜆2 = −1 and 𝛼 = 2 .
Why did we use a single parameter 𝛼? Notice that our suggested column vector must have unit
length.
Working with polar coordinates in a plane we need two parameters to determine a vector. One
parameter is the length that is fixed to one, so we have as second parameter the angle.
page 28 of 105
Exercise 3.4
Let 𝑛𝑧 = 𝑐𝑜𝑠𝜃, 𝑛𝑥 = 𝑠𝑖𝑛𝜃𝑐𝑜𝑠𝜙, and 𝑛𝑦 = 𝑠𝑖𝑛𝜃𝑠𝑖𝑛𝜙.
Angles 𝜃 and 𝜙 are defined according to the usual conventions for spherical coordinates.
Compute the eigenvalues and eigenvectors for the matrix 𝜎𝑛 :

𝜎𝑛 = ( )
**********
With the above definitions this transforms to:

𝑐𝑜𝑠 𝜃 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙
𝜎𝑛 = ( )
sin 𝜃 𝑐𝑜𝑠𝜙 + sin 𝜃 𝑠𝑖𝑛𝜙 − cos 𝜃
The determinant of the matrix
(𝑐𝑜𝑠 𝜃) − 𝜆 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙
( )
sin 𝜃 𝑐𝑜𝑠𝜙 + sin 𝜃 𝑠𝑖𝑛𝜙 − (cos 𝜃) − 𝜆
must be zero in order the matrix to have eigenvectors.
(𝑐𝑜𝑠 𝜃) − 𝜆 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙
|( )| = 0
sin 𝜃 𝑐𝑜𝑠𝜙 + sin 𝜃 𝑠𝑖𝑛𝜙 − (cos 𝜃) − 𝜆
((𝑐𝑜𝑠 𝜃) − 𝜆)(− (cos 𝜃) − 𝜆) − (sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙)(sin 𝜃 𝑐𝑜𝑠𝜙 + sin 𝜃 𝑠𝑖𝑛𝜙) =
−𝑐𝑜𝑠 2 𝜃 + 𝜆2 − (𝑠𝑖𝑛2 𝜃𝑐𝑜𝑠 2 𝜙 + 𝑠𝑖𝑛2 𝜃𝑠𝑖𝑛2 𝜙) =
−𝑐𝑜𝑠 2 𝜃 + 𝜆2 − 𝑠𝑖𝑛2 𝜃(𝑐𝑜𝑠 2 𝜙 + 𝑠𝑖𝑛2 𝜙) =
−𝑐𝑜𝑠 2 𝜃 + 𝜆2 − 𝑠𝑖𝑛2 𝜃 =
𝜆2 − (𝑠𝑖𝑛2 𝜃 + 𝑐𝑜𝑠 2 𝜃) =
𝜆2 − 1
The characteristic polynomial is 𝜆2 − 1 = 0 with the solutions 𝜆 = ∓1.

cos 𝛼
As candidate for eigenvector we choose the same as in exercise 3.3: ( )
sin 𝛼
Eigenvalue 𝜆 = +1
𝑐𝑜𝑠 𝜃 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙 cos 𝛼 cos 𝛼
( )( )=( )
sin 𝜃 𝑐𝑜𝑠𝜙 + i sin 𝜃 𝑠𝑖𝑛𝜙 − cos 𝜃 sin 𝛼 sin 𝛼
leads to two equations:
𝑐𝑜𝑠 𝜃 cos 𝛼 + (sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙) sin 𝛼 = cos 𝛼

(sin 𝜃 𝑐𝑜𝑠𝜙 + isin 𝜃 𝑠𝑖𝑛𝜙) cos 𝛼 − cos 𝜃 sin 𝛼 = sin 𝛼
First equation:
𝑐𝑜𝑠 𝜃 cos 𝛼 + sin 𝛼 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝛼 sin 𝜃 𝑠𝑖𝑛𝜙 = cos 𝛼

Second equation:
page 29 of 105
sin 𝜃 𝑐𝑜𝑠𝜙 cos 𝛼 + 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙 cos 𝛼 − cos 𝜃 sin 𝛼 = sin 𝛼

We eliminate the imaginary part of the first equation:
sin 𝛼 sin 𝜃 𝑐𝑜𝑠𝜙 − 𝑖 sin 𝛼 sin 𝜃 𝑠𝑖𝑛𝜙 =

sin 𝛼 sin 𝜃 (𝑐𝑜𝑠𝜙 − 𝑖 𝑠𝑖𝑛𝜙) =
sin 𝛼 sin 𝜃
because
(𝑐𝑜𝑠𝜙 − 𝑖 𝑠𝑖𝑛𝜙) = 𝑒 −𝑖𝜙
and
|𝑒 −𝑖𝜙 | = 1
The same holds for the second equation:
sin 𝜃 𝑐𝑜𝑠𝜙 cos 𝛼 + 𝑖 sin 𝜃 𝑠𝑖𝑛𝜙 cos 𝛼 =

sin 𝜃 cos 𝛼 (𝑐𝑜𝑠𝜙 + 𝑖 𝑠𝑖𝑛𝜙) =
sin 𝜃 cos 𝛼
The two equations simplify:
First equation:
𝑐𝑜𝑠 𝜃 cos 𝛼 + sin 𝛼 sin 𝜃 = cos 𝛼

Second equation:
sin 𝜃 cos 𝛼 − cos 𝜃 sin 𝛼 = sin 𝛼

Some more trigonometric identities
1
𝑐𝑜𝑠 𝜃 cos 𝛼 = (cos (𝜃 − 𝛼) + 𝑐𝑜𝑠(𝜃 + 𝛼))
2
1
𝑠𝑖𝑛 𝜃 sin 𝛼 = (cos (𝜃 − 𝛼) − 𝑐𝑜𝑠(𝜃 + 𝛼))
2
applied to the first equation
𝑐𝑜𝑠 𝜃 cos 𝛼 + sin 𝛼 sin 𝜃 = cos (𝜃 − 𝛼)

we get
cos(𝜃 − 𝛼) = cos 𝛼
with the solution:
𝜃
𝛼=
2
More trigonometric identities:
1
𝑐𝑜𝑠 𝜃 𝑠𝑖𝑛 𝜃 = (sin (𝜃 − 𝛼) + 𝑠𝑖𝑛(𝜃 + 𝛼))
2
page 30 of 105
1
𝑐𝑜𝑠 𝜃 sin 𝛼 = (sin (𝛼 − 𝜃) + 𝑠𝑖𝑛(𝜃 + 𝛼))
2
With sin(𝑥) = −sin (−𝑥) we write
1
𝑐𝑜𝑠 𝜃 sin 𝛼 = − (sin (𝜃 − 𝛼) + 𝑠𝑖𝑛(𝜃 + 𝛼))
2
and apply to the second equation:
𝑐𝑜𝑠 𝜃 𝑠𝑖𝑛 𝜃 − 𝑐𝑜𝑠 𝜃 sin 𝛼 = sin (𝜃 − 𝛼)

We get
sin(𝜃 − 𝛼) = sin 𝛼
with the solution:
𝜃
𝛼=
2
The computation for the eigenvalue 𝜆 = +1 is omitted (see exercise 3.3).
page 31 of 105
Exercise 3.5
Suppose that a spin is prepared so that 𝜎𝑚 = +1. The apparatus is then rotated to an arbitrary 𝑛̂
direction and 𝜎𝑛 is measured. What is the probability that the result is +1?
Note that 𝜎𝑚 = 𝜎 ∙ 𝑚
̂ , using the same convention we used for 𝜎𝑛 .
**********
We rotate the coordinate system twice.
Rotation doesn’t change the length of vectors and preserve the relative dependencies between
vectors (see exercise 4.1. is a rotation matrix unitary?).
We rotate one time in a way that 𝑚

̂ will be the z-axis, a second time in a way that 𝑛̂ will be in the x-z-
plane.
This is the situation of exercise 3.3 with the solution
𝜆1 = 1
𝜃
𝑐𝑜𝑠
|𝜆1 ⟩ = ( 2)
𝜃
𝑠𝑖𝑛
2
𝜆2 = −1
𝜃
−𝑠𝑖𝑛
|𝜆2 ⟩ = ( 2)
𝜃
𝑐𝑜𝑠
2
with the probability
𝜃 2
𝑃(+1) = |⟨𝑢|𝜆1 ⟩|2 = (𝑐𝑜𝑠 )
2
page 32 of 105
Exercise 4.1
Prove that if (the time operator) U is unitary, and if |𝐴⟩ and |𝐵⟩ are any two state-vectors, then the
inner product of 𝑈|𝐴⟩ and 𝑈|𝐵⟩ is the same as the inner product of |𝐴⟩ and |𝐵⟩. One could call this
the conservation of overlaps. It expresses the fact that the logical relation between states is
preserved with time.
U is unitary:
𝑈†𝑈 = 𝐼
𝑈|𝐴⟩ = ⟨𝐴|𝑈 †
**********
⟨𝐴|𝐵⟩ = ⟨𝐴|𝐼|𝐵⟩ = ⟨𝐴|𝑈 † 𝑈|𝐵⟩
with ⟨𝐴|𝑈 † 𝑈|𝐵⟩ being the inner product of 𝑈|𝐴⟩ and 𝑈|𝐵⟩.
page 33 of 105
Exercise 4.2
Prove that if 𝑀 and 𝐿 are both Hermitian, the (extended) commutator 𝑖[𝑀, 𝐿] is also Hermitian.
Note that the 𝑖 is important. The commutator is, by itself, not Hermitian.
Hermitian: the diagonal is pure real and: 𝑀 = 𝑀†
**********
Proof for a 3 × 3-matrix:

𝑎 𝑏 𝑐 𝑔 ℎ 𝑢
𝑀 ≔ (𝑏 ∗ 𝑑 𝑒 ) and 𝐿 ≔ (ℎ∗ 𝑗 𝑘 ) with 𝑎, 𝑑, 𝑓, 𝑔, 𝑗, 𝑙 being real.
𝑐∗ 𝑒∗ 𝑓 𝑢∗ 𝑘∗ 𝑙
𝑎𝑔 + 𝑏ℎ∗ + 𝑐𝑢 𝑎ℎ + 𝑏𝑗 + 𝑐𝑘 ∗ 𝑎𝑖 + 𝑏𝑘 + 𝑐𝑙
𝑀𝐿 = ( 𝑏 ∗ 𝑔 + 𝑑ℎ∗ + 𝑒𝑢∗ 𝑏 ∗ ℎ + 𝑑𝑗 + 𝑒𝑘 ∗ 𝑏 ∗ 𝑢 + 𝑑𝑘 + 𝑒𝑙 )
𝑐 ∗ 𝑔 + 𝑒 ∗ ℎ∗ + 𝑓𝑢∗ 𝑐 ∗ ℎ + 𝑒 ∗ 𝑗 + 𝑓𝑘 ∗ 𝑐 ∗ 𝑢 + 𝑒 ∗ 𝑘 + 𝑓𝑙
𝑔𝑎 + ℎ𝑏 ∗ + 𝑢𝑐 ∗ 𝑔𝑏 + ℎ𝑑 + 𝑢𝑒 ∗ 𝑔𝑐 + ℎ𝑒 + 𝑢𝑓
𝐿𝑀 = ( ℎ∗ 𝑎 + 𝑗𝑏 ∗ + 𝑘𝑐 ∗ ℎ∗ 𝑏 + 𝑗𝑑 + 𝑘𝑒 ∗ ℎ∗ 𝑐 + 𝑗𝑒 + 𝑘𝑓 )
𝑢∗ 𝑎 + 𝑘 ∗ 𝑏 ∗ + 𝑙𝑐 ∗ 𝑢∗ 𝑏 + 𝑘 ∗ 𝑑 + 𝑙𝑒 ∗ 𝑢∗ 𝑐 + 𝑘 ∗ 𝑒 + 𝑙𝑓
𝑀𝐿 − 𝐿𝑀 =
𝑎𝑔 + 𝑏ℎ∗ + 𝑐𝑢 − 𝑔𝑎 − ℎ𝑏 ∗ − 𝑢𝑐 ∗ 𝑎ℎ + 𝑏𝑗 + 𝑐𝑘 ∗ − 𝑔𝑏 − ℎ𝑑 − 𝑢𝑒 ∗ 𝑎𝑖 + 𝑏𝑘 + 𝑐𝑙 − 𝑔𝑐 − ℎ𝑒 − 𝑢𝑓
( 𝑏 ∗ 𝑔 + 𝑑ℎ∗ + 𝑒𝑢∗ − ℎ∗ 𝑎 − 𝑗𝑏 ∗ − 𝑘𝑐 ∗ 𝑏 ∗ ℎ + 𝑑𝑗 + 𝑒𝑘 ∗ − ℎ∗ 𝑏 − 𝑗𝑑 − 𝑘𝑒 ∗ 𝑏 ∗ 𝑢 + 𝑑𝑘 + 𝑒𝑙 − ℎ∗ 𝑐 − 𝑗𝑒 − 𝑘𝑓 )
𝑐 ∗ 𝑔 + 𝑒 ∗ ℎ∗ + 𝑓𝑢∗ − 𝑢∗ 𝑎 − 𝑘 ∗ 𝑏 ∗ − 𝑙𝑐 ∗ 𝑐 ∗ ℎ + 𝑒 ∗ 𝑗 + 𝑓𝑘 ∗ − 𝑢∗ 𝑏 − 𝑘 ∗ 𝑑 − 𝑙𝑒 ∗ 𝑐 ∗ 𝑢 + 𝑒 ∗ 𝑘 + 𝑓𝑙 − 𝑢∗ 𝑐 − 𝑘 ∗ 𝑒 − 𝑙𝑓
… with some transformation work …

𝑏ℎ∗ − ℎ𝑏 ∗ + 𝑐 ∗ 𝑢 − 𝑢𝑐 ∗ ℎ(𝑎 − 𝑑) + 𝑏(𝑗 − 𝑔) + 𝑐𝑘 ∗ − 𝑢𝑒 ∗ 𝑢(𝑎 − 𝑓) + 𝑐(𝑙 − 𝑔) + 𝑏𝑘 − ℎ𝑒
∗ (𝑎
( −ℎ − 𝑑) − 𝑏 ∗ (𝑗 − 𝑔) − 𝑐 ∗ 𝑘 + 𝑢∗ 𝑒 𝑏 ∗ ℎ − ℎ∗ 𝑏 + 𝑒𝑘 ∗ − 𝑘𝑒 ∗ 𝑒(𝑙 − 𝑗) + 𝑘(𝑑 − 𝑓) + 𝑏 ∗ 𝑢 − ℎ∗ 𝑐 )
∗ (𝑎
−𝑢 − 𝑓) − 𝑐 ∗ (𝑙 − 𝑔) − 𝑏 ∗ 𝑘 ∗ + ℎ∗ 𝑒 ∗ 𝑒 − 𝑗) − 𝑘 ∗ (𝑑 − 𝑓) − 𝑏𝑢∗ + ℎ𝑐 ∗
∗ (𝑙
𝑐 ∗ 𝑢 − 𝑢∗𝑐 + 𝑒 ∗ 𝑘 − 𝑘 ∗𝑒
The entries (1,1), (2,2), (3,3) are completely imaginary, because the differences 𝑏ℎ∗ − ℎ𝑏 ∗ etc. are
imaginary and become thus become real if we multiplicate them by the imaginary unit 𝑖.
We check this with 𝑧1 ≔ 𝑥 + 𝑖𝑦, 𝑧2 ≔ 𝑢 + 𝑖𝑣, 𝑥, 𝑦, 𝑢, 𝑣 are real numbers.
𝑧1 ∙ 𝑧2 ∗ − 𝑧1 ∗ ∙ 𝑧2 =
(𝑥 + 𝑖𝑦)(𝑢 − 𝑖𝑣) − (𝑥 − 𝑦)(𝑢 + 𝑖𝑣) =
𝑥𝑢 − 𝑖𝑥𝑣 + 𝑖𝑦𝑢 + 𝑦𝑣 − (𝑥𝑢 + 𝑖𝑥𝑣 − 𝑖𝑦𝑢 + 𝑦𝑣) =

𝑥𝑢 − 𝑖𝑥𝑣 + 𝑖𝑦𝑢 + 𝑦𝑣 − 𝑥𝑢 − 𝑖𝑥𝑣 + 𝑖𝑦𝑢 − 𝑦𝑣) =
−2𝑖𝑥𝑣 + 2𝑖𝑦𝑢) =
2𝑖(𝑦𝑢 − 𝑥𝑣)
This is a complete imaginary number.
We multiplicate the matrix with the imaginary unit 𝑖:

𝑖(𝑏ℎ∗ − ℎ𝑏 ∗ + 𝑐𝑢 − 𝑢𝑐 ∗ ) 𝑖ℎ(𝑎 − 𝑑) + 𝑖𝑏(𝑗 − 𝑔) + 𝑖𝑐𝑘 ∗ − 𝑖𝑢𝑒 ∗ 𝑖𝑢(𝑎 − 𝑓) + 𝑖𝑐(𝑙 − 𝑔) + 𝑖𝑏𝑘 − 𝑖ℎ𝑒
( −𝑖ℎ∗ (𝑎 − 𝑑) − 𝑖𝑏 ∗ (𝑗 − 𝑔) − 𝑖𝑐 ∗ 𝑘 + 𝑖𝑢∗ 𝑒 𝑖(𝑏 ∗ ℎ − ℎ∗ 𝑏 + 𝑒𝑘 ∗ − 𝑘𝑒 ∗ ) 𝑖𝑒(𝑙 − 𝑗) + 𝑖𝑘(𝑑 − 𝑓) + 𝑖𝑏 ∗ 𝑢 − 𝑖ℎ∗ 𝑐 )
−𝑖𝑢∗ (𝑎 − 𝑓) − 𝑖𝑐 ∗ (𝑙 − 𝑔) − 𝑖𝑏 ∗ 𝑘 ∗ + 𝑖ℎ∗ 𝑒 ∗ 𝑖𝑒 − 𝑗) − 𝑖𝑘 ∗ (𝑑 − 𝑓) − 𝑖𝑏𝑢∗ + 𝑖ℎ𝑐 ∗
∗ (𝑙
𝑖(𝑐 ∗ 𝑢 − 𝑢∗ 𝑐 + 𝑒 ∗ 𝑘 − 𝑘 ∗ 𝑒)
page 34 of 105
Multiplication of the whole matrix with the imaginary unit makes the entries on the diagonal real.
We have to show that the entries away from the diagonal (1,2), (2,1), (1,3), (3,1), (2,3), (3,2) fulfil
the Hermitian criterion: (1,2)∗ = (2,1) etc.
(1,2)∗ = (2,1):
(1,2) = 𝑖ℎ(𝑎 − 𝑑) + 𝑖𝑏(𝑗 − 𝑔) + 𝑖𝑐𝑘 ∗ − 𝑖𝑢𝑒 ∗
Note that 𝑎, 𝑑, 𝑗, 𝑔 are real numbers.
(1,2)∗ = −𝑖ℎ∗ (𝑎 − 𝑑) − 𝑖𝑏 ∗ (𝑗 − 𝑔) − 𝑖𝑐 ∗ 𝑘 + 𝑖𝑢∗ 𝑒

We compare this with (2,1):
(2,1) = −𝑖ℎ∗ (𝑎 − 𝑑) − 𝑖𝑏 ∗ (𝑗 − 𝑔) − 𝑖𝑐 ∗ 𝑘 + 𝑖𝑢∗ 𝑒
This is correct. The same holds for (1,3)∗ = (3,1) and (2,3)∗ = (3,2).
page 35 of 105
Exercise 4.3
With the definition of Poisson brackets and check that the identification in Eq. 4.21 is dimensionally
consistent. Show that without the factor ℏ, it would not be.
Eq. 4.21
[𝐹, 𝐺] ↔ 𝑖ℏ{𝐹, 𝐺}
𝑘𝑔 ∙ 𝑚2
[ℏ] = 𝐽 ∙ 𝑠 = ∙𝑠
𝑠2
**********
(not a proof, more a reflection about …)
In Wikipedia we find
𝑠
𝑑𝑓 𝜕𝑓 𝜕𝐻 𝜕𝑓 𝜕𝐻 𝜕𝑓
𝑓̇ = = ∑( + )+
𝑑𝑡 𝜕𝑞𝑘 𝜕𝑝𝑘 𝜕𝑝𝑘 𝜕𝑞𝑘 𝜕𝑡
𝑘=1
𝑑𝑓 𝜕𝑓
𝑓̇ = = {𝑓, 𝐻} +
𝑑𝑡 𝜕𝑡
In “Classical Mechanics, The Theoretical Minimum” we find:
𝜕𝐹 𝜕𝐻 𝜕𝐹 𝜕𝐻
𝐹̇ = ∑ ( + )
𝜕𝑞𝑖 𝜕𝑝𝑖 𝜕𝑝𝑖 𝜕𝑞𝑖
𝑖
𝐹̇ = {𝐹, 𝐻}
We can conclude that 𝐹 has no explicit time-dependency and in fact it is defined as 𝐹(𝑞, 𝑝) leading to
𝜕𝐹
= 0.
𝜕𝑡
△𝐹
Following calculus rules 𝐹̇ must be something like △𝑡 , so
1
[𝐹]̇~
𝑠
and accordingly, the unit of {𝐹, 𝐻}.
On the other hand, we find in “Quantum Mechanics” (4.19):

𝑑𝐿 𝑖
𝐿̇ = = − [𝐿, 𝐻]
𝑑𝑡 ℏ
𝑘𝑔∙𝑚2
H is an energy, so the unit of 𝐻 is 𝑠2
(and I hope that the multiplication with L doesn’t change
𝑘𝑔∙𝑚2
that). Divided by the unit of ℏ: 𝑠
this results in:
𝑘𝑔 ∙ 𝑚2
[𝐻] 2 𝑘𝑔 ∙ 𝑚2 ∙ 𝑠 1
= 𝑠 2 = 2 2
=
[ℏ] 𝑘𝑔 ∙ 𝑚 𝑘𝑔 ∙ 𝑚 ∙ 𝑠 𝑠
𝑠
𝑑𝐿
giving the correct dimension for 𝑑𝑡 = 𝐿̇.
page 36 of 105
Exercise 4.4
Verify the commutation relations:
[𝜎𝑥 , 𝜎𝑦 ] = 2𝑖𝜎𝑧
[𝜎𝑦 , 𝜎𝑧 ] = 2𝑖𝜎𝑥
[𝜎𝑧 , 𝜎𝑥 ] = 2𝑖𝜎𝑦
0 1 0 −𝑖 1 0
𝜎𝑥 = ( ) , 𝜎𝑦 = ( ) , 𝜎𝑧 = ( )
1 0 𝑖 0 0 −1
**********
[𝜎𝑥 , 𝜎𝑦 ] = 𝜎𝑥 𝜎𝑦 − 𝜎𝑦 𝜎𝑥 =
0 1 0 −𝑖 0 −𝑖 0 1
( )( )−( )( )=
1 0 𝑖 0 𝑖 0 1 0
𝑖 0 −𝑖 0
( )−( )=
0 −𝑖 0 𝑖
2𝑖 0
( ) = 2𝑖𝜎𝑧
0 −2𝑖
[𝜎𝑦 , 𝜎𝑧 ] = 𝜎𝑦 𝜎𝑧 − 𝜎𝑧 𝜎𝑦 =
0 −𝑖 1 0 1 0 0 −𝑖
( )( )−( )( )=
𝑖 0 0 −1 0 −1 𝑖 0
0 𝑖 0 −𝑖
( )−( )=
𝑖 0 −𝑖 0
0 2𝑖
( ) = 2𝑖𝜎𝑥
2𝑖 0
[𝜎𝑧 , 𝜎𝑥 ] = 𝜎𝑧 𝜎𝑥 − 𝜎𝑥 𝜎𝑧 =
1 0 0 1 0 1 1 0
( )( )−( )( )=
0 −1 1 0 1 0 0 −1
0 1 0 −1
( )−( )=
−1 0 1 0
0 2
( ) = 2𝑖𝜎𝑦
−2 0
page 37 of 105
Exercise 4.5
Take any unit 3-vector 𝑛⃗ and form the operator

ℏ𝜔
𝐻= 𝜎 ∙ 𝑛⃗
2
Find the energy eigenvalues and eigenvectors by solving the time-independent Schrödinger equation.
Recall 𝜎 ∙ 𝑛⃗ in component form:
𝜎𝑛 = 𝜎 ∙ 𝑛⃗ = ( )
**********
ℏ𝜔 ℏ𝜔 𝑛𝑧 (𝑛𝑥 − 𝑖𝑛𝑦 )
𝐻= 𝜎 ∙ 𝑛⃗ = ( )
2 2 (𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧
The time-independent Schrödinger equation (4.28)
𝐻|𝐸𝑗 ⟩ = 𝐸𝑗 |𝐸𝑗 ⟩
ℏ𝜔 𝑛𝑧 (𝑛𝑥 − 𝑖𝑛𝑦 )
( ) |𝐸𝑗 ⟩ = 𝐸𝑗 |𝐸𝑗 ⟩
2 (𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧
The characteristic polynomial of the matrix 𝜎𝑛 must be zero:
𝑛𝑧 − 𝜆 (𝑛𝑥 − 𝑖𝑛𝑦 )
|( )| =
(𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧 − 𝜆
(𝑛𝑧 − 𝜆)(−𝑛𝑧 − 𝜆) − (𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 ) =
𝜆2 − (𝑛𝑥2 + 𝑛𝑦2 + 𝑛𝑧2 ) = 0
𝜆2 = (𝑛𝑥2 + 𝑛𝑦2 + 𝑛𝑧2 )
𝜆 = ±√(𝑛𝑥2 + 𝑛𝑦2 + 𝑛𝑧2 )
Because n is a unit vector 𝜆 = ±1
The eigenvectors we get out of the equation
𝑛𝑧 − 𝜆 (𝑛𝑥 − 𝑖𝑛𝑦 ) 𝑎 0
( )( ) = ( )
(𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧 − 𝜆 𝑏 0
page 38 of 105
eigenvalue 𝜆 = +1
𝑛𝑧 − 1 (𝑛𝑥 − 𝑖𝑛𝑦 ) 𝑎 0
( )( ) = ( )
(𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧 − 1 𝑏 0
We get two independent equations:
𝐼: (𝑛𝑧 − 1) ∙ 𝑎 + (𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏 = 0
𝐼𝐼: (𝑛𝑥 + 𝑖𝑛𝑦 ) ∙ 𝑎 − (𝑛𝑧 + 1) ∙ 𝑏 = 0
Upper line:
(𝑛𝑧 − 1) ∙ 𝑎 = −(𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏
−(𝑛𝑧 − 1) ∙ 𝑎 = (𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏
(1 − 𝑛𝑧 ) ∙ 𝑎 = (𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏
(𝑛𝑥 − 𝑖𝑛𝑦 )
𝑎= ∙𝑏
(1 − 𝑛𝑧 )
The result inserted in the lower line:
(𝑛𝑥 + 𝑖𝑛𝑦 ) ∙ ∙ 𝑏 − (𝑛𝑧 + 1) ∙ 𝑏 = 0
(1 − 𝑛𝑧 )
(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 )

∙ 𝑏 − (𝑛𝑧 + 1) ∙ 𝑏 = 0
(1 − 𝑛𝑧 )
(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 )

𝑏( − (𝑛𝑧 + 1)) = 0
(1 − 𝑛𝑧 )
(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 ) − (𝑛𝑧 + 1)(1 − 𝑛𝑧 )

𝑏( )=0
(1 − 𝑛𝑧 )
𝑏 ((𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 ) − (𝑛𝑧 + 1)(1 − 𝑛𝑧 )) = 0
𝑏(𝑛𝑥2 + 𝑛𝑦2 + 𝑛𝑧2 − 1) = 0
𝑏∙0=0
Valid for all b. Back to the upper line:
𝑎= ∙𝑏
(1 − 𝑛𝑧 )
This defines the first eigenvector |𝜆1 ⟩:
|𝜆1 ⟩ = ( (1 − 𝑛𝑧 ) )
1
page 39 of 105
eigenvalue 𝜆 = −1
𝑛𝑧 + 1 (𝑛𝑥 − 𝑖𝑛𝑦 ) 𝑎 0
( )( ) = ( )
(𝑛𝑥 + 𝑖𝑛𝑦 ) −𝑛𝑧 + 1 𝑏 0
𝐼: (𝑛𝑧 + 1) ∙ 𝑎 + (𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏 = 0
𝐼𝐼: (𝑛𝑥 + 𝑖𝑛𝑦 ) ∙ 𝑎 − (𝑛𝑧 − 1) ∙ 𝑏 = 0
Upper line:
(𝑛𝑧 + 1) ∙ 𝑎 = −(𝑛𝑥 − 𝑖𝑛𝑦 ) ∙ 𝑏
−(𝑛𝑥 − 𝑖𝑛𝑦 )
𝑎= ∙𝑏
(𝑛𝑧 + 1)
The result inserted in the lower line:
(𝑛𝑥 + 𝑖𝑛𝑦 ) ∙ ∙ 𝑏 − (𝑛𝑧 − 1) ∙ 𝑏 = 0
(𝑛𝑧 + 1)
−(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 )

∙ 𝑏 − (𝑛𝑧 − 1) ∙ 𝑏 = 0
(𝑛𝑧 + 1)
−(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 )

𝑏( − (𝑛𝑧 − 1)) = 0
(𝑛𝑧 + 1)
−(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 ) − (𝑛𝑧 − 1)(𝑛𝑧 + 1)

𝑏( )=0
(𝑛𝑧 + 1)
𝑏 (−(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑛𝑥 − 𝑖𝑛𝑦 ) − (𝑛𝑧 − 1)(𝑛𝑧 + 1)) = 0
𝑏(−𝑛𝑥2 − 𝑛𝑦2 − 𝑛𝑧2 + 1) = 0
𝑏∙0=0
Valid for all b. Back to the upper line:
𝑎= ∙𝑏
(𝑛𝑧 + 1)
This defines the second eigenvector |𝜆2 ⟩:
|𝜆2 ⟩ = ( (𝑛𝑧 + 1) )
1
page 40 of 105
Additional we check whether the eigenvectors are orthogonal.
|𝜆1 ⟩ = ( (1 − 𝑛𝑧 ) )
1
−(𝑛𝑥 − 𝑖𝑛𝑦 ) (𝑖𝑛𝑦 − 𝑛𝑥 )

|𝜆2 ⟩ = ( (𝑛𝑧 + 1) ) = ( (𝑛𝑧 + 1) )
1 1
We need the bra ⟨𝜆1 | and conjugate it:
(𝑛𝑥 + 𝑖𝑛𝑦 )
⟨𝜆1 | = ( 1)
(1 − 𝑛𝑧 )
(𝑖𝑛𝑦 − 𝑛𝑥 )
|𝜆2 ⟩ = ( (𝑛𝑧 + 1) )
1
The scalar product:
⟨𝜆1|𝜆2 ⟩ =
(𝑛 + 𝑖𝑛𝑦 ) (𝑖𝑛𝑦 − 𝑛𝑥 )
( 𝑥 1 ) ( (𝑛𝑧 + 1) ) =
(1 − 𝑛𝑧 )
1
(𝑛𝑥 + 𝑖𝑛𝑦 )(𝑖𝑛𝑦 − 𝑛𝑥 )
+1=
(1 − 𝑛𝑧 )(𝑛𝑧 + 1)
−𝑛𝑥2 − 𝑛𝑦2
+1=
1 − 𝑛𝑧2
−𝑛𝑥2 − 𝑛𝑦2 + 1 − 𝑛𝑧2
=
1 − 𝑛𝑧2
1 − (𝑛𝑥2 + 𝑛𝑦2 + 𝑛𝑧2 )
=0
1 − 𝑛𝑧2
Both vectors are orthogonal to each other.
page 41 of 105
Exercise 4.6
Carry out the Schrödinger Ket recipe for a single spin.

ℏ𝜔
The Hamiltonian is 𝐻 = 𝜎
2 𝑧
and the final observable is 𝜎𝑥 .
The initial state is given as |𝑢⟩ (the state in which 𝜎𝑧 = ±1).
After time t, an experiment is done to measure 𝜎𝑦 .
What are the possible outcomes and what are the probabilities for those outcomes?
**********
ℏ𝜔
For easier calculation we omit the factor and work with 𝐻 = 𝜎𝑧 only. We will correct this in the
2
end.
Recipe step 1. Derive, look, guess, borrow or steal the Hamiltonian operator H
The Hamiltonian is given:

1 0
𝐻 = 𝜎𝑧 = ( )
0 −1
Recipe step 2. Prepare an initial state |Ψ(0)⟩
1
The initial state is given: |𝑢⟩ or ( )
0
Recipe step 3. Find the eigenvalues and eigenvectors of H by solving the time-independent
Schrödinger equation:
𝐻|𝐸𝑗 ⟩ = 𝐸𝑗 |𝐸𝑗 ⟩
The characteristic polynomial of the matrix 𝜎𝑧 must be zero:

1−𝜆 0
|( )| =
0 −1 − 𝜆
(1 − 𝜆)(−1 − 𝜆) − 0 =
𝜆2 − 1 = 0
𝜆2 = 1
𝜆 = ±1
The eigenvectors we get out of the equation
1−𝜆 0 𝑎 0
( )( ) = ( )
0 −1 − 𝜆 𝑏 0
eigenvalue 𝜆1 = +1
1−1 0 𝑎 0
( )( ) = ( )
0 −1 − 1 𝑏 0
𝐼: 0 ∙ 𝑎 + 0 ∙ 𝑏 = 0
𝐼𝐼: 0 ∙ 𝑎 − 2 ∙ 𝑏 = 0 → 𝑏 = 0
page 42 of 105
This defines the first eigenvector |𝜆1 ⟩:

1
|𝜆1 ⟩ = ( )
0
eigenvalue 𝜆2 = −1
1+1 0 𝑎 0
( )( ) = ( )
0 −1 + 1 𝑏 0
𝐼: 2 ∙ 𝑎 + 0 ∙ 𝑏 = 0 → 𝑎 = 0
𝐼𝐼: 0 ∙ 𝑎 + 0 ∙ 𝑏 = 0
This defines the second eigenvector |𝜆2 ⟩:
0
|𝜆2 ⟩ = ( )
1
Obviously both eigenvectors are orthogonal to each other:
⟨𝜆1 |𝜆2 ⟩ = (1 0) ∙ (0) = 0

1
Recipe step 4. Use the initial state-vector |Ψ(0)⟩, along with the eigenvectors |𝐸𝑗 ⟩ from step 3, to
calculate the initial coefficients 𝑎𝑗 (0):
𝑎𝑗 (0) = ⟨𝐸𝑗 |Ψ(0)⟩
As both eigenvectors are real, we can change |𝜆1 ⟩ to ⟨𝜆1 | and |𝜆2 ⟩ to ⟨𝜆2 |.
𝑎1 (0) for Eigenvector |𝜆1 ⟩
𝑎1 (0) = ⟨𝐸1 |Ψ(0)⟩ = ⟨𝜆1 |u⟩ =

1
(1 0) ( ) = 1
0
𝑎2 (0) for Eigenvector |𝜆2 ⟩
𝑎2 (0) = ⟨𝐸2 |Ψ(0)⟩ = ⟨𝜆2 |u⟩ =

1
(0 1) ( ) = 0
0
Recipe step 5. Rewrite |Ψ(0)⟩ in terms of the eigenvectors |𝐸𝑗 ⟩ and the initial coefficients 𝑎𝑗 (0):
|Ψ(0)⟩ = ∑ 𝑎𝑗 (0)|𝐸𝑗 ⟩
𝑗
1
|Ψ(0)⟩ = 1 ∙ |𝜆1 ⟩ + 0 ∙ |𝜆2 ⟩ = ( ) = |𝑢⟩
0
Recipe step 6. In the above equation, replace each 𝑎𝑗 (0) with 𝑎𝑗 (𝑡) to capture it’s time-dependence.
As a result, |Ψ(0)⟩ becomes |Ψ(𝑡)⟩:
|Ψ(𝑡)⟩ = ∑ 𝑎𝑗 (𝑡)|𝐸𝑗 ⟩
𝑗
1 0
|Ψ(𝑡)⟩ = 𝑎1 (𝑡) ∙ |𝜆1 ⟩ + 𝑎2 (𝑡) ∙ |𝜆2 ⟩ = 𝑎1 (𝑡) ( ) + 𝑎2 (𝑡) ( ) = 𝑎1 (𝑡)|𝑢⟩ + 𝑎1 (𝑡)|𝑑⟩
0 1
page 43 of 105
𝑖
Recipe step 7. Using Eq. 4.30, replace each 𝑎𝑗 (𝑡) with 𝑎𝑗 (0)𝑒 −ℏ𝐸𝑗𝑡 :
𝑖 𝑖
− 𝐸1 𝑡 1 − 𝐸 𝑡 0
|Ψ(𝑡)⟩ = 𝑎1 (𝑡) ∙ |𝜆1 ⟩ + 𝑎2 (𝑡) ∙ |𝜆2 ⟩ = 𝑎1 (0)𝑒 ℏ ( ) + 𝑎2 (0)𝑒 ℏ 2 ( ) =
0 1
𝑖 𝑖 𝑖 𝑖
1 0 1
1 ∙ 𝑒 −ℏ𝑡 ( ) + 0 ∙ 𝑒 −ℏ𝑡 ( ) = 𝑒 −ℏ𝑡 ( ) = 𝑒 −ℏ𝑡 |𝑢⟩
0 1 0
ℏ𝜔
Finally, we multiply the factor 2
and get:
𝑖ℏ𝜔 𝑖𝜔
|Ψ(𝑡)⟩ = 𝑒 − 2ℏ 𝑡 |𝑢⟩ = 𝑒 − 2 𝑡 |𝑢⟩
Question: After time t, an experiment is done to measure 𝜎𝑦 . What are the possible outcomes and
what are the probabilities for those outcomes?
We omit the time-changing phase and follow “3.7 Reaping the Results” modified in a way that:
𝜋 𝜋
𝑛𝑧 = 𝑐𝑜𝑠 2 , 𝑛𝑦 = 𝑠𝑖𝑛 2 , 𝑛𝑥 = 0
This gives
𝜋 𝜋
𝑐𝑜𝑠 −𝑖 ∙ 𝑠𝑖𝑛
2 2 0 −𝑖
𝜎𝑛 = ( 𝜋 𝜋 ) = ( 𝑖 0 ) ≔ 𝜎𝑦
𝑖 ∙ 𝑠𝑖𝑛 −𝑐𝑜𝑠
2 2
Calculating the eigenvectors of 𝜎𝑦 :
0 −𝑖 𝑎 𝑎
( )( ) = 𝜆( )
𝑖 0 𝑏 𝑏
𝐼: −𝑖𝑏 = 𝜆𝑎
𝐼𝐼: 𝑖𝑎 = 𝜆𝑏
Solving this pair of equations leads to eigenvalues:
𝜆 = ±1
The corresponding eigenvectors are:
𝜆 = +1:
1
( )
𝑖
𝜆 = −1:
1
( )
−𝑖
Normalization:
1
|( )| = √2
𝑖
1
|( )| = √2
−𝑖
The normalized eigenvectors are:
page 44 of 105
1 1
( )
√2 𝑖
1 1
( )
√2 −𝑖
We calculate the probability of observing 𝜎𝑦 = +1:

2
1 1
𝑃(+1) = |⟨𝑢|𝜆1 ⟩|2 = | ((1 0) ( ))| =
√2 𝑖
2 2
1 1 1
| ∙ (1 + 0 ∙ 𝑖)| = | | =
√2 √2 2
Analog the probability of observing 𝜎𝑦 = −1:
2
1 1
𝑃(−1) = |⟨𝑢|𝜆2⟩|2 = | ((1 0) ( ))| =
√2 −𝑖
2
1 1
| ∙ (1 − 0 ∙ 𝑖)| =
√2 2
The possible outcomes of the measurement are ±1 with probability each 50%.
𝑖𝜔
We ignore the phase-factor 𝑒 − 2 𝑡 .
page 45 of 105
Exercise 5.1
Verify that any 2 × 2 Hermitian matrix L can be written as a sum of four terms,
𝐿 = 𝑎𝜎𝑥 + 𝑏𝜎𝑦 + 𝑐𝜎𝑧 + 𝑑𝐼
where a, b, c and d are real numbers.
The four Pauli matrices:

0 1 0 −𝑖 1 0 1 0
𝜎𝑥 = ( ), 𝜎 = ( ), 𝜎𝑧 = ( ), I= ( )
1 0 𝑦 𝑖 0 0 −1 0 1
A general Hermitian matrix (𝑟, 𝑟 ′ ) are real numbers:
𝑟 𝑤
( ∗ )
𝑤 𝑟′
**********
Verification:
𝐿 = 𝑎𝜎𝑥 + 𝑏𝜎𝑦 + 𝑐𝜎𝑧 + 𝑑𝐼 =

0 1 0 −𝑖 1 0 1 0
𝑎( )+𝑏( )+𝑐( )+𝑑( )=
1 0 𝑖 0 0 −1 0 1
0 𝑎 0 −𝑖𝑏 𝑐 0 𝑑 0
( )+( )+( )+( )=
𝑎 0 𝑖𝑏 0 0 −𝑐 0 𝑑
𝑐+𝑑 𝑎 − 𝑖𝑏
( )
𝑎 + 𝑖𝑏 𝑑−𝑐
Obviously, it’s correct that 𝑤 ∗ and 𝑤 are complex conjugated: 𝑎 + 𝑖𝑏 and 𝑎 − 𝑖𝑏. Further 𝑟 and 𝑟′ are
real numbers. As Hermitian matrices are not necessarily unitary, the coefficients a, b, c and d are free
variables so we can express any 𝑟 and 𝑟′: 𝑟 = 𝑐 + 𝑑, 𝑟 ′ = 𝑑 − 𝑐.
page 46 of 105
Exercise 5.2
1) Show that (△ 𝐴)2 = 〈𝐴2̅ 〉 and (△ 𝐵)2 = 〈𝐵̅2 〉

2) Show that [𝐴̅, 𝐵̅] = [𝐴, 𝐵]
1
3) Using these relations, show that △ 𝐴 △ 𝐵 ≥ 2 |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|
The square of uncertainty (or standard deviation) of A, (△ 𝐴)2 :
(△ 𝐴)² = ∑ 𝑎̅2 𝑃(𝑎) = ∑(𝑎 − 〈𝐴〉)2 𝑃(𝐴)

𝑎 𝑎
**********
1) Show that (△ 𝐴)2 = 〈𝐴̅2 〉
First (△ 𝐴)2 :
(△ 𝐴)2 = ∑(𝑎 − 〈𝐴〉)2 𝑃(𝑎) =

𝑎
∑(𝑎2 − 2𝑎〈𝐴〉 + 〈𝐴〉2 )𝑃(𝑎) =

𝑎
∑ 𝑎2 𝑃(𝑎) − 2〈𝐴〉 ∑ 𝑎𝑃(𝑎) + 〈𝐴〉2 ∑ 𝑃(𝑎) =

𝑎 𝑎 𝑎
∑ 𝑎2 𝑃(𝑎) − 2〈𝐴〉〈𝐴〉 + 〈𝐴〉2 =

𝑎
∑ 𝑎2 𝑃(𝑎) − 〈𝐴〉2 =
𝑎
〈𝐴2 〉 − 〈𝐴〉2
We get (△ 𝐴)2 = 〈𝐴2 〉 − 〈𝐴〉2 .
On the other hand:
〈𝐴̅2 〉 = 〈(𝐴 − 〈𝐴〉)2 〉 =
〈𝐴2 − 2𝐴〈𝐴〉 + 〈𝐴〉2 〉 =
〈𝐴2 〉 − 2〈𝐴〉〈〈𝐴〉〉 + 〈〈𝐴〉〈𝐴〉〉 =
〈𝐴2 〉 − 2〈𝐴〉〈𝐴〉 + 〈𝐴〉〈𝐴〉 =

〈𝐴2 〉 − 2〈𝐴〉2 + 〈𝐴〉2 =
〈𝐴2 〉 − 〈𝐴〉2
We get: 〈𝐴2̅ 〉 = 〈𝐴2 〉 − 〈𝐴〉2 .
We can conclude (△ 𝐴)2 = 〈𝐴̅2 〉. The same holds for (△ 𝐵)2 = 〈𝐵̅2 〉.
page 47 of 105
2. Show that [𝐴̅, 𝐵̅] = [𝐴, 𝐵]
[𝐴̅, 𝐵̅] = 𝐴̅𝐵̅ − 𝐵̅𝐴̅ = (𝐴 − 〈𝐴〉)(𝐵 − 〈𝐵〉) − (𝐵 − 〈𝐵〉)(𝐴 − 〈𝐴〉) =
𝐴𝐵 − 𝐴〈𝐵〉 − 〈𝐴〉𝐵 + 〈𝐴〉〈𝐵〉 − (𝐵𝐴 − 𝐵〈𝐴〉 − 〈𝐵〉𝐴 + 〈𝐵〉〈𝐴〉) =

𝐴𝐵 − 𝐴〈𝐵〉 − 〈𝐴〉𝐵 + 〈𝐴〉〈𝐵〉 + 𝐵〈𝐴〉 + 〈𝐵〉𝐴 − 〈𝐵〉〈𝐴〉 =
𝐴𝐵 − 𝐵𝐴 − 𝐴〈𝐵〉 + 〈𝐵〉𝐴 − 〈𝐴〉𝐵 + 𝐵〈𝐴〉 + 〈𝐴〉〈𝐵〉 − 〈𝐵〉〈𝐴〉 =
𝐴𝐵 − 𝐵𝐴 − 𝐴〈𝐵〉 + 𝐴〈𝐵〉 − 𝐵〈𝐴〉 + 𝐵〈𝐴〉 + 〈𝐴〉〈𝐵〉 − 〈𝐴〉〈𝐵〉 =
𝐴𝐵 − 𝐵𝐴 = [𝐴, 𝐵]
1
3. Using these relations, show that △ 𝐴 △ 𝐵 ≥ 2 |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|
The Cauchy-Schwarz inequality:
2|𝑋||𝑌| ≥ |⟨𝑋|𝑌⟩ + ⟨𝑌|𝑋⟩|
Let |Ψ⟩ be any ket and let 𝐴 and 𝐵 be any two observables (∈ ℝ). The associated variables 𝐴̅ and 𝐵̅
are built in a way that the expectation values of 𝐴̅ and 𝐵̅ are zero. We define |X⟩ and |Y⟩ as follows:
|X⟩ = 𝐴̅|Ψ⟩
̅∗ = ⟨Ψ|𝐴̅
⟨X| = ⟨Ψ|A
|Y⟩ = 𝑖𝐵̅|Ψ⟩
⟨Y| = ⟨Ψ|(−𝑖𝐵̅∗ )
Notice the 𝑖 in the definition of 𝑌.
Now, substitute into the Cauchy-Schwarz inequality to get
2√〈𝐴̅2 〉〈𝐵̅2 〉 ≥ |⟨Ψ|𝑖𝐴̅𝐵̅|Ψ⟩ − ⟨Ψ|𝑖𝐵̅𝐴̅|Ψ⟩| =
2√〈𝐴̅2 〉〈𝐵̅2 〉 ≥ |𝑖(⟨Ψ|𝐴̅𝐵̅|Ψ⟩ − ⟨Ψ|𝐵̅𝐴̅|Ψ⟩)| =
2√〈𝐴̅2 〉〈𝐵̅2 〉 ≥ |⟨Ψ|𝐴̅𝐵̅|Ψ⟩ − ⟨Ψ|𝐵̅𝐴̅|Ψ⟩| =

The minus sign is due to the factor of 𝑖 in the definition of |Y⟩. Using the definition of a commutator,
we find that
2√〈𝐴̅2 〉〈𝐵̅2 〉 ≥ |⟨Ψ|[𝐴̅, 𝐵̅]|Ψ⟩|
We use (△ 𝐴)2 = 〈𝐴̅2 〉 and (△ 𝐵)2 = 〈𝐵̅2 〉 and replace the left side of the inequality:
2√(△ 𝐴)2 (△ 𝐵)2 ≥ |⟨Ψ|[𝐴̅, 𝐵̅]|Ψ⟩|
We use [𝐴̅, 𝐵̅] = [𝐴, 𝐵] and replace the right side of the inequality:
2√(△ 𝐴)2 (△ 𝐵)2 ≥ |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|
page 48 of 105
We simplify the left side:
2(△ 𝐴)(△ 𝐵) ≥ |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|

We get the result:
1
(△ 𝐴)(△ 𝐵) ≥ |⟨Ψ|[𝐴, 𝐵]|Ψ⟩|
2
page 49 of 105
Exercise 6.1
Prove that if 𝑃(𝑎, 𝑏) factorizes: 𝑃(𝑎, 𝑏) = 𝑃𝐴 (𝑎)𝑃𝑏 (𝑏)
then the correlation between a and b is zero:
〈𝜎𝐴 𝜎𝐵 〉 − 〈𝜎𝐴 〉〈𝜎𝐵 〉 = 0

Average:
〈𝜎𝐴 〉 = ∑ 𝑎𝑛 𝑃(𝑎𝑛 )
𝑛
〈𝜎𝐵 〉 = ∑ 𝑏𝑛 𝑃(𝑏𝑛 )
𝑛
〈𝜎𝐴 𝜎𝐵 〉 = ∑ ∑ 𝑎𝑛 𝑏𝑛 𝑃(𝑎𝑛 𝑏𝑛 )
𝑛 𝑛
**********
Assuming we have n objects in A and b:
〈𝜎𝐴 〉〈𝜎𝐵 〉 = ∑ 𝑎𝑛 𝑃(𝑎𝑛 ) ∑ 𝑏𝑛 𝑃(𝑏𝑛 ) =

𝑛 𝑛
𝑎1 𝑃(𝑎1 ) ∑ 𝑏𝑛 𝑃(𝑏𝑛 ) + 𝑎2 𝑃(𝑎2 ) ∑ 𝑏𝑛 𝑃(𝑏𝑛 ) + ⋯ + 𝑎𝑛 𝑃(𝑎𝑛 ) ∑ 𝑏𝑛 𝑃(𝑏𝑛 ) =

𝑛 𝑛 𝑛
𝑎1 𝑃(𝑎1 )𝑏1 𝑃(𝑏1 ) + 𝑎1 𝑃(𝑎1 )𝑏2 𝑃(𝑏2 ) + ⋯ + 𝑎1 𝑃(𝑎1 )𝑏𝑛 𝑃(𝑏𝑛 ) +

𝑎2 𝑃(𝑎2 )𝑏1 𝑃(𝑏1 ) + 𝑎2 𝑃(𝑎2 )𝑏2 𝑃(𝑏2 ) + ⋯ + 𝑎2 𝑃(𝑎2 )𝑏𝑛 𝑃(𝑏𝑛 ) +
⋮
𝑎𝑛 𝑃(𝑎𝑛 )𝑏1 𝑃(𝑏1 ) + 𝑎𝑛 𝑃(𝑎𝑛 )𝑏2 𝑃(𝑏2 ) + ⋯ + 𝑎𝑛 𝑃(𝑎𝑛 )𝑏𝑛 𝑃(𝑏𝑛 ) =
𝑎1 𝑏1 𝑃(𝑎1 )𝑃(𝑏1 ) + 𝑎1 𝑏2 𝑃(𝑎1 )𝑃(𝑏2 ) + ⋯ + 𝑎1 𝑃𝑏𝑛 𝑃(𝑎1 )𝑃(𝑏𝑛 ) +

𝑎2 𝑏1 𝑃(𝑎2 )𝑃(𝑏1 ) + 𝑎2 𝑏2 𝑃(𝑎2 )𝑃(𝑏2 ) + ⋯ + 𝑎2 𝑏𝑛 𝑃(𝑎2 )𝑃(𝑏𝑛 ) +
⋮
𝑎𝑛 𝑏1 𝑃(𝑎𝑛 )𝑃(𝑏1 ) + 𝑎𝑛 𝑏2𝑃(𝑎𝑛 ) 𝑃(𝑏2 ) + ⋯ + 𝑎𝑛 𝑏𝑛 𝑃(𝑎𝑛 )𝑃(𝑏𝑛 ) =
𝑎1 𝑏1 𝑃(𝑎1 𝑏1 ) + 𝑎1 𝑏2 𝑃(𝑎1 𝑏2 ) + ⋯ + 𝑎1 𝑏𝑛 𝑃(𝑎1 𝑏𝑛 ) +

𝑎2 𝑏1 𝑃(𝑎2 𝑏1 ) + 𝑎2 𝑏2 𝑃(𝑎2 𝑏2 ) + ⋯ + 𝑎2 𝑏𝑛 𝑃(𝑎2 𝑏𝑛 ) +
⋮
𝑎𝑛 𝑏1 𝑃(𝑎𝑛 𝑏1 ) + 𝑎𝑛 𝑏2 𝑃(𝑎𝑛 𝑏2 ) + ⋯ + 𝑎𝑛 𝑏𝑛 𝑃(𝑎𝑛 𝑏𝑛 ) =
∑ ∑ 𝑎𝑛 𝑏𝑛 𝑃(𝑎𝑛 𝑏𝑛 ) = 〈𝜎𝐴 𝜎𝐵 〉
𝑛 𝑛
page 50 of 105
Exercise 6.2
Show that if the two normalization conditions are satisfied:
then the state-vector is automatically normalized as well:

In other words, show that for this product state, normalizing the overall state-vector does not put
any additional constraints on the 𝛼′𝑠 and 𝛽′𝑠.
**********
Normalization means that
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 + 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 + 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 = 1
From (6.4) we know that
(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) = 1
We multiply the brackets:

(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 =
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 + 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 + 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 = 1
This is exactly the normalization condition.
page 51 of 105
Exercise 6.3
Prove that the state |𝑠𝑖𝑛𝑔⟩ cannot be written as a product state.

1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
The shape of a product state:

**********
If the singlet state can be composed out of a product state, the following equation should be valid:
1
𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
It follows that 𝛼𝑢 𝛽𝑢 = 0 and 𝛼𝑑 𝛽𝑑 = 0, as the vectors |𝑢𝑢⟩ and |𝑑𝑑⟩ do not appear in the singlet
state.
𝛼𝑢 𝛽𝑢 = 0 → 𝛼𝑢 = 0 𝑜𝑟 𝛽𝑢 = 0 or both.
𝛼𝑑 𝛽𝑑 = 0 → 𝛼𝑑 = 0 𝑜𝑟 𝛽𝑑 = 0 or both.
From 𝛼𝑢 𝛽𝑑 ≠ 0 it follows that 𝛼𝑢 ≠ 0 and 𝛽𝑑 ≠ 0, from 𝛼𝑑 𝛽𝑢 ≠ 0 it follows that 𝛼𝑑 ≠ 0 and 𝛽𝑢 ≠

0.
We get a contradiction and can conclude that it is not possible to combine the singlet state out of the
parameters of a product state.
page 52 of 105
Exercise 6.4
Use the matrix forms of 𝜎𝑥 , 𝜎𝑦 , 𝜎𝑧 and the column vectors for |𝑢⟩ and |𝑑⟩ to verify:
𝜎𝑧 |𝑢⟩ = |𝑢⟩ and 𝜎𝑧 |𝑑⟩ = −|𝑑⟩
𝜎𝑥 |𝑢⟩ = |𝑑⟩ and 𝜎𝑥 |𝑑⟩ = |𝑢⟩
𝜎𝑦 |𝑢⟩ = 𝑖|𝑑⟩ and 𝜎𝑦 |𝑑⟩ = −𝑖|𝑢⟩
Then, use:
𝜏𝑧 |𝑢⟩ = |𝑢⟩ and 𝜏𝑧 |𝑑⟩ = −|𝑑⟩
𝜏𝑥 |𝑢⟩ = |𝑑⟩ and 𝜏𝑥 |𝑑⟩ = |𝑢⟩
𝜏𝑦 |𝑢⟩ = 𝑖|𝑑⟩ and 𝜏𝑦 |𝑑⟩ = −𝑖|𝑢⟩
to write the equations for all possible combinations of the tensor product states 𝜎𝑧 |𝑢𝑢⟩ = |𝑢𝑢⟩ etc.
**********
Written as matrices and column vectors:

1 0 0 −𝑖 0 1
𝜎𝑧 = ( ) 𝜎𝑦 = ( ) 𝜎𝑥 = ( )
0 −1 𝑖 0 1 0
1 0
|𝑢⟩ = ( ) |𝑑⟩ = ( )
0 1
1 0 1 1∙1+0∙0 1
𝜎𝑧 |𝑢⟩ = ( )( ) = ( ) = ( ) = |𝑢⟩
0 −1 0 0 ∙ 1 + (−1) ∙ 0 0
1 0 0 1∙0+0∙1 0 0
𝜎𝑧 |𝑑⟩ = ( )( ) = ( ) = ( ) = − ( ) = −|𝑑⟩
0 −1 1 0 ∙ 0 + (−1) ∙ 1 −1 1
0 1 1 0∙1+1∙0 0
𝜎𝑥 |𝑢⟩ = ( )( ) = ( ) = ( ) = |𝑑⟩
1 0 0 1∙1+0∙0 1
0 1 0 0∙0+1∙1 1
𝜎𝑥 |𝑑⟩ = ( )( ) = ( ) = ( ) = |𝑢⟩
1 0 1 1∙0+0∙1 0
0 −𝑖 1 0 ∙ 1 + (−𝑖) ∙ 0 0 0
𝜎𝑦 |𝑢⟩ = ( )( ) = ( ) = ( ) = 𝑖 ( ) = 𝑖|𝑑⟩
𝑖 0 0 𝑖∙1+0∙0 𝑖 1
0 −𝑖 0 0 ∙ 0 + (−𝑖) ∙ 1 −𝑖 1
𝜎𝑦 |𝑑⟩ = ( )( ) = ( ) = ( ) = −𝑖 ( ) = −𝑖|𝑢⟩
𝑖 0 1 𝑖∙0+0∙1 0 0
The complete list of tensor product states:
𝜎𝑧 |𝑢𝑢⟩ = |𝑢𝑢⟩ 𝜎𝑧 |𝑢𝑑⟩ = |𝑢𝑑⟩ 𝜎𝑧 |𝑑𝑢⟩ = −|𝑑𝑢⟩ 𝜎𝑧 |𝑑𝑑⟩ = −|𝑑𝑑⟩

𝜎𝑥 |𝑢𝑢⟩ = |𝑑𝑢⟩ 𝜎𝑥 |𝑢𝑑⟩ = |𝑑𝑑⟩ 𝜎𝑥 |𝑑𝑢⟩ = |𝑢𝑢⟩ 𝜎𝑥 |𝑑𝑑⟩ = |𝑢𝑑⟩
𝜎𝑦 |𝑢𝑢⟩ = 𝑖|𝑑𝑢⟩ 𝜎𝑦 |𝑢𝑑⟩ = 𝑖|𝑑𝑑⟩ 𝜎𝑦 |𝑑𝑢⟩ = −𝑖|𝑢𝑢⟩ 𝜎𝑦 |𝑑𝑑⟩ = −𝑖|𝑢𝑑⟩
𝜏𝑧 |𝑢𝑢⟩ = |𝑢𝑢⟩ 𝜏𝑧 |𝑢𝑑⟩ = −|𝑢𝑑⟩ 𝜏𝑧 |𝑑𝑢⟩ = |𝑑𝑢⟩ 𝜏𝑧 |𝑑𝑑⟩ = −|𝑑𝑑⟩

𝜏𝑥 |𝑢𝑢⟩ = |𝑢𝑑⟩ 𝜏𝑥 |𝑢𝑑⟩ = |𝑢𝑢⟩ 𝜏𝑥 |𝑑𝑢⟩ = |𝑑𝑑⟩ 𝜏𝑥 |𝑑𝑑⟩ = |𝑑𝑢⟩
𝜏𝑦 |𝑢𝑢⟩ = 𝑖|𝑢𝑑⟩ 𝜏𝑦 |𝑢𝑑⟩ = −𝑖|𝑢𝑢⟩ 𝜏𝑦 |𝑑𝑢⟩ = 𝑖|𝑑𝑑⟩ 𝜏𝑦 |𝑑𝑑⟩ = −𝑖|𝑑𝑢⟩
page 53 of 105
Exercise 6.5
Prove the following theorem:
When any one of Alice’s and Bob’s spin operators acts on a product state, the result is still a product
state.
Show that in a product state, the expectation value of any component of 𝜎̅ or 𝜏̅ is the same as it
would be in the individual single-spin states.
**********
This proof only performed for 𝜎𝑧 .
Equation (6.5) tells us the shape of a product state:
|𝑃⟩ ≔ 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩

and
⟨𝑃| ≔ ⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗
A single state has the shape:
|𝐴⟩ ≔ 𝛼𝑢 |𝑢⟩ + 𝛼𝑑 |𝑑⟩ and ⟨𝐴| ≔ ⟨𝑢|𝛼𝑢∗ + ⟨𝑑|𝛼𝑑∗
The operator 𝜎𝑧 :
1 0
𝜎𝑧 = ( )
0 −1
We check the result of the expectation-value definition for both the generic product-state 𝑃 and the
generic single state 𝐴.
The single state:
〈𝜎𝑧 〉 = ⟨𝐴|𝜎𝑧 |𝐴⟩ =

⟨𝐴|𝜎𝑧 |(𝛼𝑢 |𝑢⟩ + 𝛼𝑑 |𝑑⟩)⟩ =
⟨𝐴|(𝛼𝑢 |𝑢⟩ − 𝛼𝑑 |𝑑⟩)⟩ =
⟨(⟨𝑢|𝛼𝑢∗ + ⟨𝑑|𝛼𝑑∗ )|(𝛼𝑢 |𝑢⟩ − 𝛼𝑑 |𝑑⟩)⟩ =
⟨𝑢|𝛼𝑢∗ 𝛼𝑢 |𝑢⟩ − ⟨𝑢|𝛼𝑢∗ 𝛼𝑑 |𝑑⟩ + ⟨𝑑|𝛼𝑑∗ 𝛼𝑢 |𝑢⟩ − ⟨𝑑|𝛼𝑑∗ 𝛼𝑑 |𝑑⟩ =
𝛼𝑢∗ 𝛼𝑢 − 0 + 0 − 𝛼𝑑∗ 𝛼𝑑
𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑
The product state with the behavior that Alice’s operator only acts on her half of the product state:
〈𝜎𝑧 〉 = ⟨𝑃|𝜎𝑧 |𝑃⟩ =

⟨𝑃|𝜎𝑧 |(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)⟩ =
⟨𝑃|(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)⟩ =
⟨⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ |𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩⟩ =
page 54 of 105
… boxes for better readability only …

⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + ⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − ⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − ⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩ +
⟨𝑢𝑑|𝛼𝑢∗ 𝛽∗𝑑 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽∗𝑑 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − ⟨𝑢𝑑|𝛼𝑢∗ 𝛽∗𝑑 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − ⟨𝑢𝑑|𝛼𝑢∗ 𝛽∗𝑑 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩ +
⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − ⟨𝑑𝑢 |𝛼∗𝑑 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩ +
⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩ =
⟨𝑢𝑢|𝑢𝑢⟩𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + ⟨𝑢𝑢|𝑢𝑑⟩𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑑 − ⟨𝑢𝑢|𝑑𝑢⟩𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 − ⟨𝑢𝑢|𝑑𝑑⟩𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑑 +

⟨𝑢𝑑|𝑢𝑢⟩𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑢 + ⟨𝑢𝑑|𝑢𝑑⟩𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 − ⟨𝑢𝑑|𝑑𝑢⟩𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑢 − ⟨𝑢𝑑|𝑑𝑑⟩𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 +
⟨𝑑𝑢|𝑢𝑢⟩𝛼∗𝑑 𝛽∗𝑢 𝛼𝑢 𝛽𝑢 + ⟨𝑑𝑢|𝑢𝑑⟩𝛼∗𝑑 𝛽∗𝑢 𝛼𝑢 𝛽𝑑 − ⟨𝑑𝑢|𝑑𝑢⟩𝛼∗𝑑 𝛽∗𝑢 𝛼𝑑 𝛽𝑢 − ⟨𝑑𝑢|𝑑𝑑⟩𝛼∗𝑑 𝛽∗𝑢 𝛼𝑑 𝛽𝑑 +
⟨𝑑𝑑|𝑢𝑢⟩𝛼∗𝑑 𝛽∗𝑑 𝛼𝑢 𝛽𝑢 + ⟨𝑑𝑑|𝑢𝑑⟩𝛼∗𝑑 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 − ⟨𝑑𝑑|𝑑𝑢⟩𝛼∗𝑑 𝛽∗𝑑 𝛼𝑑 𝛽𝑢 − ⟨𝑑𝑑|𝑑𝑑⟩𝛼∗𝑑 𝛽∗𝑑 𝛼𝑑 𝛽𝑑 =
With orthogonality in ⟨𝑢𝑢|𝑢𝑑⟩ etc. we get
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 − 𝛼∗𝑑 𝛽∗𝑢 𝛼𝑑 𝛽𝑢 − 𝛼∗𝑑 𝛽∗𝑑 𝛼𝑑 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 − 𝛼∗𝑑 𝛼𝑑 𝛽∗𝑢 𝛽𝑢 − 𝛼∗𝑑 𝛼𝑑 𝛽∗𝑑 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) − 𝛼∗𝑑 𝛼𝑑 (𝛽∗𝑢 𝛽𝑢 + 𝛽∗𝑑 𝛽𝑑 ) =
(𝛼𝑢∗ 𝛼𝑢 − 𝛼∗𝑑 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
(𝛼𝑢∗ 𝛼𝑢 − 𝛼∗𝑑 𝛼𝑑 )
This is the same result as in Alice’s system.
Reflections:
1 1
0 0
|𝑢𝑢⟩ is not simply the stacked vector ( ), but instead ( ). It’s built out of the tensor product:
1 0
0 0
1 1
1( )
1 1 0 0
( )⨂( ) = ( )=( )
0 0 1 0
0( )
0 0
The same holds for the other combinations.
1 0
The operator 𝜎𝑧 = ( ) can’t act on a state-vector with four components. Instead we need the
0 −1
1 0
tensor product of both Alice’s and Bob’s operator 𝜏𝑧 = ( ): 𝜎𝑧 ⨂𝜏𝑧 .
0 −1
1 0
We concentrate on Alice and let Bob alone, his operator will be the identity: 𝜏𝑧 = ( ).
0 1
1 0 1 0
𝜎𝑧𝑖 ≔ 𝜎𝑧 ⨂𝐼 = ( )⨂( )=
0 −1 0 1
1 0 1 0
1( ) 0( )
( 0 1 0 1 )=
1 0 1 0
0( ) −1 ( )
0 1 0 1
page 55 of 105
1 0 0 0
0 1 0 0
( )
0 0 −1 0
0 0 0 −1
We have the four state-vectors:
1 1
1( )
1 1 0 0
|𝑢𝑢⟩ = ( ) ⨂ ( ) = ( )=( )
0 0 1 0
0( )
0 0
0 0
1( )
1 0 1
|𝑢𝑑⟩ = ( ) ⨂ ( ) = ( 1 ) = ( )
0 1 0 0
0( )
1 0
1 0
0( )
0 1 0 0
|𝑑𝑢⟩ = ( ) ⨂ ( ) = ( )=( )
1 0 1 1
1( )
0 0
0 0
0( )
0 0 1 0
|𝑑𝑑⟩ = ( ) ⨂ ( ) = ( )=( )
1 1 0 0
1( )
1 1
Now we can check the effect of this new 𝜎𝑧𝑖 on the combined vectors |𝑢𝑢⟩ etc.
1 0 0 0 1 1
0 1 0 0 0 0
𝜎𝑧𝑖 |𝑢𝑢⟩ = ( )( ) = ( )
0 0 −1 0 0 0
0 0 0 −1 0 0
1 0 0 0 0 0
0 1 0 0 1 1
𝜎𝑧𝑖 |𝑢𝑑⟩ = ( )( ) = ( )
0 0 −1 0 0 0
0 0 0 −1 0 0
1 0 0 0 0 0 0
0 1 0 0 0 0 0
𝜎𝑧𝑖 |𝑑𝑢⟩ = ( )( ) = ( ) = −( )
0 0 −1 0 1 −1 1
0 0 0 −1 0 0 0
1 0 0 0 0 0 0
0 1 0 0 0 0 0
𝜎𝑧𝑖 |𝑑𝑑⟩ = ( )( ) = ( ) = −( )
0 0 −1 0 0 0 0
0 0 0 −1 1 −1 1
The results are as expected.
We will check one special case. As the operators 𝜎𝑧 and 𝜏𝑧 both change the sign of the “d”-
1 0 1 0
component, so 𝜎𝑧 𝜏𝑧 ≔ 𝜎𝑧 ⨂𝜏𝑧 = ( )⨂( ) should leave |𝑑𝑑⟩ intact.
0 −1 0 −1
1 0 1 0
1( ) 0( )
𝜎𝑧 𝜏𝑧 = ( 0 −1 0 1 )=
1 0 1 0
0( ) −1 ( )
0 1 0 −1
page 56 of 105
1 0 0 0
0 −1 0 0
( )
0 0 −1 0
0 0 0 1
We check:
1 0 0 0 0 0
0 −1 0 0 0 0
𝜎𝑧 𝜏𝑧 |𝑑𝑑⟩ = ( )( ) = ( )
0 0 −1 0 0 0
0 0 0 1 1 1
and get the correct result:
𝜎𝑧 𝜏𝑧 |𝑑𝑑⟩ = |𝑑𝑑⟩
page 57 of 105
Exercise 6.6
Assume Charlie has prepared the two spins in the singlet state. This time, Bob measures 𝜏𝑦 and Alice
measures 𝜎𝑥 . What is the expectation value of 𝜎𝑥 𝜏𝑦 ?
What does this say about the correlation between the two measurements?
1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
1
⟨𝑠𝑖𝑛𝑔| = (⟨𝑢𝑑| − ⟨𝑑𝑢|)
√2
**********
〈𝜎𝑥 𝜏𝑦 〉 = ⟨𝑠𝑖𝑛𝑔|𝜎𝑥 𝜏𝑦 |𝑠𝑖𝑛𝑔⟩ =
1
⟨𝑠𝑖𝑛𝑔|𝜎𝑥 𝜏𝑦 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔|𝜎𝑥 | (−𝑖|𝑢𝑢⟩ − 𝑖|𝑑𝑑⟩)⟩ =
√2
−𝑖
⟨𝑠𝑖𝑛𝑔|𝜎𝑥 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
−𝑖
⟨𝑠𝑖𝑛𝑔| (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2
|𝑠𝑖𝑛𝑔⟩ is not an eigenvector of 𝜎𝑥 𝜏𝑦 .
1 −𝑖
⟨ (⟨𝑢𝑑| − ⟨𝑑𝑢|)| (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2 √2
−𝑖
(⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑢𝑑|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩ − ⟨𝑑𝑢|𝑢𝑑⟩) =
2
−𝑖
(0 + 1 − 1 − 0)) = 0
2
The expectation value 〈𝜎𝑥 𝜏𝑦 〉 is 0.
What does this say about the correlation between the two measurements? The two measurements
are not correlated, the results of the measurements are independent.
page 58 of 105
Exercise 6.7
Next (after Charlie has had prepared the two spins in the singlet state), Charlie prepares the spins in a
different state, called |𝑇1 ⟩, where
1
|𝑇1 ⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩)
√2
T stands for triplet. The triplet states are completely different from the states in the coin and die
examples. What are the expectation values of the operators 𝜎𝑧 𝜏𝑧 , 𝜎𝑥 𝜏𝑥 , and 𝜎𝑦 𝜏𝑦 ?
Notice what a difference a sign can make.
**********
1
⟨𝑇1 | = (⟨𝑢𝑑| + ⟨𝑑𝑢|)
√2
The expectation value of 〈𝜎𝑧 𝜏𝑧 〉:
〈𝜎𝑧 𝜏𝑧 〉 = ⟨𝑇1 |𝜎𝑧 𝜏𝑧 |𝑇1 ⟩ =

1
⟨𝑇1 |𝜎𝑧 𝜏𝑧 | (|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑇1 |𝜎𝑧 | (−|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑇1 | (−|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
|𝑇1 ⟩ is eigenvector of 𝜎𝑧 𝜏𝑧 with eigenvalue of -1.
1 1
⟨ (⟨𝑢𝑑| + ⟨𝑑𝑢|)| (−|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2 √2
1
(−⟨𝑢𝑑|𝑢𝑑⟩ − ⟨𝑢𝑑|𝑑𝑢⟩ − ⟨𝑑𝑢|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩) =
2
1
(−1 − 0 − 0 − 1)) = −1
2
The expectation value 〈𝜎𝑧 𝜏𝑧 〉 is −1, correlation.
The expectation value of 〈𝜎𝑥 𝜏𝑥 〉:
〈𝜎𝑥 𝜏𝑥 〉 = ⟨𝑇1 |𝜎𝑥 𝜏𝑥 |𝑇1 ⟩ =

1
⟨𝑇1 |𝜎𝑥 𝜏𝑥 | (|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑇1 |𝜎𝑥 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇1 | (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2
|𝑇1 ⟩ is eigenvector of 𝜎𝑥 𝜏𝑥 with eigenvalue 1.
page 59 of 105
1 1
⟨ (⟨𝑢𝑑| + ⟨𝑑𝑢|)| (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2 √2
1
(⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑢𝑑|𝑢𝑑⟩ + ⟨𝑑𝑢|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩) =
2
1
(0 + 1 + 1 + 0) = 1
2
The expectation value 〈𝜎𝑥 𝜏𝑥 〉 is 1, correlation.
The expectation value of 〈𝜎𝑦 𝜏𝑦 〉:
〈𝜎𝑦 𝜏𝑦 〉 = ⟨𝑇1 |𝜎𝑦 𝜏𝑦 |𝑇1 ⟩ =
1
⟨𝑇1 |𝜎𝑦 𝜏𝑦 | (|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
𝑖
⟨𝑇1|𝜎𝑦 | (−|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
𝑖
⟨𝑇1 | (−𝑖|𝑑𝑢⟩ − 𝑖|𝑢𝑑⟩)⟩ =
√2
1
⟨𝑇1 | (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2
|𝑇1 ⟩ is eigenvector of 𝜎𝑦 𝜏𝑦 with eigenvalue 1.
1 1
⟨ (⟨𝑢𝑑| + ⟨𝑑𝑢|)| (|𝑑𝑢⟩ + |𝑢𝑑⟩)⟩ =
√2 √2
1
(⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑢𝑑|𝑢𝑑⟩ + ⟨𝑑𝑢|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩) =
2
1
(0 + 1 + 1 + 0) = 1
2
The expectation value 〈𝜎𝑦 𝜏𝑦 〉 is 1, correlation.
What a difference a sign can make: if we change the singlet state to the triplet state, the behavior
changes. With the singlet state, the correlation was always −1, now the expectation values changes:
〈𝜎𝑧 𝜏𝑧 〉 would become −1, 〈𝜎𝑥 𝜏𝑥 〉 = 1 and 〈𝜎𝑦 𝜏𝑦 〉 = 1 .
page 60 of 105
Exercise 6.8
Calculate the expectation values of the operators 𝜎𝑧 𝜏𝑧 , 𝜎𝑥 𝜏𝑥 , and 𝜎𝑦 𝜏𝑦 for the other two entangled
triplet states:
1
|𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)
√2
1
|𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩)
√2
(Addendum: at the end of this exercise we check the expectation values for the singlet state too.)
**********
1
First |𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)
√2
〈𝜎𝑧 𝜏𝑧 〉 = ⟨𝑇2 |𝜎𝑧 𝜏𝑧 |𝑇2 ⟩ =

1
⟨𝑇2 |𝜎𝑧 𝜏𝑧 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇2 |𝜎𝑧 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇2 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
|𝑇2 ⟩ is eigenvector of 𝜎𝑧 𝜏𝑧 with eigenvalue 1.
1 1
⟨ (⟨𝑢𝑢| + ⟨𝑑𝑑|)| (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2 √2
1
(⟨𝑢𝑢|𝑢𝑢⟩ + ⟨𝑢𝑢|𝑑𝑑⟩ + ⟨𝑑𝑑|𝑢𝑢⟩ + ⟨𝑑𝑑|𝑑𝑑⟩) =
2
1
(1 + 0 + 0 + 1)) = 1
2
The expectation value 〈𝜎𝑧 𝜏𝑧 〉 is 1, correlation.
〈𝜎𝑥 𝜏𝑥 〉 = ⟨𝑇2 |𝜎𝑥 𝜏𝑥 |𝑇2 ⟩ =

1
⟨𝑇2 |𝜎𝑥 𝜏𝑥 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇2 |𝜎𝑥 | (|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑇2 | (|𝑑𝑑⟩ + |𝑢𝑢⟩)⟩ =
√2
|𝑇2 ⟩ is eigenvector of 𝜎𝑥 𝜏𝑥 .
page 61 of 105
1 1
⟨ (⟨𝑢𝑢| + ⟨𝑑𝑑|)| (|𝑑𝑑⟩ + |𝑢𝑢⟩)⟩ =
√2 √2
1
(⟨𝑢𝑢|𝑑𝑑⟩ + ⟨𝑢𝑢|𝑢𝑢⟩ + ⟨𝑑𝑑|𝑑𝑑⟩ + ⟨𝑑𝑑|𝑢𝑢⟩) =
2
1
(0 + 1 + 1 + 0) = 1
2
The expectation value 〈𝜎𝑥 𝜏𝑥 〉 is 1, correlation.
〈𝜎𝑦 𝜏𝑦 〉 = ⟨𝑇2 |𝜎𝑦 𝜏𝑦 |𝑇2 ⟩ =
1
⟨𝑇2 |𝜎𝑦 𝜏𝑦 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
𝑖
⟨𝑇2 |𝜎𝑦 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
𝑖
⟨𝑇2 | (𝑖|𝑑𝑑⟩ + 𝑖|𝑢𝑢⟩)⟩ =
√2
1
⟨𝑇2 | (−|𝑑𝑑⟩ − |𝑢𝑢⟩)⟩ =
√2
|𝑇2 ⟩ is eigenvector of 𝜎𝑦 𝜏𝑦 with eigenvalue −1.
1 1
⟨ (⟨𝑢𝑢| + ⟨𝑑𝑑|)| (−|𝑑𝑑⟩ − |𝑢𝑢⟩)⟩ =
√2 √2
1
(−⟨𝑢𝑢|𝑑𝑑⟩ − ⟨𝑢𝑢|𝑢𝑢⟩ − ⟨𝑑𝑑|𝑑𝑑⟩ − ⟨𝑑𝑑|𝑢𝑢⟩) =
2
1
(−0 − 1 − 1 − 0) = −1
2
The expectation value 〈𝜎𝑦 𝜏𝑦 〉 is −1, correlation.
page 62 of 105
1
Second |𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩)
√2
〈𝜎𝑧 𝜏𝑧 〉 = ⟨𝑇3 |𝜎𝑧 𝜏𝑧 |𝑇3 ⟩ =

1
⟨𝑇3 |𝜎𝑧 𝜏𝑧 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇3 |𝜎𝑧 | (|𝑢𝑢⟩ + |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇3 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
|𝑇3 ⟩ is eigenvector of 𝜎𝑧 𝜏𝑧 with eigenvalue 1.
1 1
⟨ (⟨𝑢𝑢| − ⟨𝑑𝑑|)| (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2 √2
1
(⟨𝑢𝑢|𝑢𝑢⟩ − ⟨𝑢𝑢|𝑑𝑑⟩ − ⟨𝑑𝑑|𝑢𝑢⟩ + ⟨𝑑𝑑|𝑑𝑑⟩) =
2
1
(1 − 0 − 0 + 1)) = 1
2
The expectation value 〈𝜎𝑧 𝜏𝑧 〉 is 1, correlation.
〈𝜎𝑥 𝜏𝑥 〉 = ⟨𝑇3 |𝜎𝑥 𝜏𝑥 |𝑇3 ⟩ =

1
⟨𝑇3 |𝜎𝑥 𝜏𝑥 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑇3 |𝜎𝑥 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑇3 | (|𝑑𝑑⟩ − |𝑢𝑢⟩)⟩ =
√2
|𝑇3 ⟩ is eigenvector of 𝜎𝑥 𝜏𝑥 with eigenvalue −1.
1 1
⟨ (⟨𝑢𝑢| − ⟨𝑑𝑑|)| (|𝑑𝑑⟩ − |𝑢𝑢⟩)⟩ =
√2 √2
1
(⟨𝑢𝑢|𝑑𝑑⟩ − ⟨𝑢𝑢|𝑢𝑢⟩ − ⟨𝑑𝑑|𝑑𝑑⟩ + ⟨𝑑𝑑|𝑢𝑢⟩) =
2
1
(0 − 1 − 1 + 0) = −1
2
The expectation value 〈𝜎𝑥 𝜏𝑥 〉 is −1, correlation.
page 63 of 105
〈𝜎𝑦 𝜏𝑦 〉 = ⟨𝑇3 |𝜎𝑦 𝜏𝑦 |𝑇3 ⟩ =
1
⟨𝑇3 |𝜎𝑦 𝜏𝑦 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
𝑖
⟨𝑇3 |𝜎𝑦 | (|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
𝑖
⟨𝑇3 | (𝑖|𝑑𝑑⟩ − 𝑖|𝑢𝑢⟩)⟩ =
√2
1
⟨𝑇3 | (−|𝑑𝑑⟩ + |𝑢𝑢⟩)⟩ =
√2
|𝑇3 ⟩ is eigenvector of 𝜎𝑦 𝜏𝑦 with eigenvalue 1.
1 1
⟨ (⟨𝑢𝑢| − ⟨𝑑𝑑|)| (−|𝑑𝑑⟩ + |𝑢𝑢⟩)⟩ =
√2 √2
1
(−⟨𝑢𝑢|𝑑𝑑⟩ + ⟨𝑢𝑢|𝑢𝑢⟩ + ⟨𝑑𝑑|𝑑𝑑⟩ − ⟨𝑑𝑑|𝑢𝑢⟩) =
2
1
(−0 + 1 + 1 − 0) = 1
2
The expectation value 〈𝜎𝑦 𝜏𝑦 〉 is 1, correlation.
Addendum: for the singlet state we check these expectation values too.
Assume Charlie has prepared the two spins in the singlet state. What are the expectation values of
operators σz τz , σx τx , and σy τy?
1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
1
⟨𝑠𝑖𝑛𝑔| = (⟨𝑢𝑑| − ⟨𝑑𝑢|)
√2
〈𝜎𝑥 𝜏𝑥 〉 = ⟨𝑠𝑖𝑛𝑔|𝜎𝑥 𝜏𝑥 |𝑠𝑖𝑛𝑔⟩ =

1
⟨𝑠𝑖𝑛𝑔|𝜎𝑥 𝜏𝑥 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔|𝜎𝑥 | (|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔| (|𝑑𝑢⟩ − |𝑢𝑑⟩)⟩ =
√2
|𝑠𝑖𝑛𝑔⟩ is eigenvector of 𝜎𝑥 𝜏𝑥 with eigenvalue -1, the expectation value of 〈𝜎𝑥 𝜏𝑥 〉 is -1.
page 64 of 105
〈𝜎𝑦 𝜏𝑦 〉 = ⟨𝑠𝑖𝑛𝑔|𝜎𝑦 𝜏𝑦 |𝑠𝑖𝑛𝑔⟩ =
1
⟨𝑠𝑖𝑛𝑔|𝜎𝑦 𝜏𝑦 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
𝑖
⟨𝑠𝑖𝑛𝑔|𝜎𝑦 | (−|𝑢𝑢⟩ − |𝑑𝑑⟩)⟩ =
√2
𝑖
⟨𝑠𝑖𝑛𝑔| (−𝑖|𝑑𝑢⟩ + 𝑖|𝑢𝑑⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔| (|𝑑𝑢⟩ − |𝑢𝑑⟩)⟩ =
√2
|𝑠𝑖𝑛𝑔⟩ is eigenvector of 𝜎𝑦 𝜏𝑦 with eigenvalue -1, the expectation value of 〈𝜎𝑥 𝜏𝑥 〉 is -1.
〈𝜎𝑧 𝜏𝑧 〉 = ⟨𝑠𝑖𝑛𝑔|𝜎𝑧 𝜏𝑧 |𝑠𝑖𝑛𝑔⟩ =

1
⟨𝑠𝑖𝑛𝑔|𝜎𝑧 𝜏𝑧 | (|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔|𝜎𝑧 | (−|𝑢𝑑⟩ − |𝑑𝑢⟩)⟩ =
√2
1
⟨𝑠𝑖𝑛𝑔| (−|𝑢𝑑⟩ + |𝑑𝑢⟩)⟩ =
√2
|𝑠𝑖𝑛𝑔⟩ is eigenvector of 𝜎𝑧 𝜏𝑧 with eigenvalue -1, the expectation value of 〈𝜎𝑥 𝜏𝑥 〉 is -1.
We get the following matrix:

〈 〉 𝑧𝑧 𝑥𝑥 𝑦𝑦
𝑇1 −1 1 1
𝑇2 1 1 −1
𝑇3 1 −1 1
𝑠𝑖𝑛𝑔 −1 −1 −1
page 65 of 105
Exercise 6.9
Prove that the four vectors |𝑠𝑖𝑛𝑔⟩, |𝑇1 ⟩, |𝑇2 ⟩, |𝑇3 ⟩ are eigenvectors of 𝜎 ∙ 𝜏:
𝜎 ∙ 𝜏 = 𝜎𝑥 𝜏𝑥 + 𝜎𝑦 𝜏𝑦 + 𝜎𝑧 𝜏𝑧
What are their eigenvalues?

1
|𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)
√2
1 1 1
|𝑇1 ⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩) |𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩) |𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩)
√2 √2 √2
**********
The complete list of tensor product states:


Applied to |𝑠𝑖𝑛𝑔⟩:
1 1 1
𝜎𝑥 𝜏𝑥 |𝑠𝑖𝑛𝑔⟩ = 𝜎𝑥 𝜏𝑥 (|𝑢𝑑⟩ − |𝑑𝑢⟩) = 𝜎𝑥 (|𝑢𝑢⟩ − |𝑑𝑑⟩) = (|𝑑𝑢⟩ − |𝑢𝑑⟩) = −|𝑠𝑖𝑛𝑔⟩
√2 √2 √2
1 1 1
𝜎𝑦 𝜏𝑦 |𝑠𝑖𝑛𝑔⟩ = 𝜎𝑦 𝜏𝑦 (|𝑢𝑑⟩ − |𝑑𝑢⟩) = 𝜎𝑦 (−𝑖|𝑢𝑢⟩ − 𝑖|𝑑𝑑⟩) = (|𝑑𝑢⟩ − |𝑢𝑑⟩) = −|𝑠𝑖𝑛𝑔⟩
√2 √2 √2
1 1 1
𝜎𝑧 𝜏𝑧 |𝑠𝑖𝑛𝑔⟩ = 𝜎𝑧 𝜏𝑧 (|𝑢𝑑⟩ − |𝑑𝑢⟩) = 𝜎𝑧 (−|𝑢𝑑⟩ − |𝑑𝑢⟩) = (−|𝑢𝑑⟩ + |𝑑𝑢⟩) = −|𝑠𝑖𝑛𝑔⟩
√2 √2 √2
Result: 𝜎 ∙ 𝜏|𝑠𝑖𝑛𝑔⟩ = −3|𝑠𝑖𝑛𝑔⟩
|𝑠𝑖𝑛𝑔⟩ is eigenvector of 𝜎 ∙ 𝜏 with eigenvalue -3.
𝜎 ∙ 𝜏 applied to |𝑇1 ⟩ gives:

1 1 1
𝜎𝑥 𝜏𝑥 |𝑇1 ⟩ = 𝜎𝑥 𝜏𝑥 (|𝑢𝑑⟩ + |𝑑𝑢⟩) = 𝜎𝑥 (|𝑢𝑢⟩ + |𝑑𝑑⟩) = (|𝑑𝑢⟩ + |𝑢𝑑⟩) = |𝑇1 ⟩
√2 √2 √2
1 1 1
𝜎𝑦 𝜏𝑦 |𝑇1 ⟩ = 𝜎𝑦 𝜏𝑦 (|𝑢𝑑⟩ + |𝑑𝑢⟩) = 𝜎𝑦 (−𝑖|𝑢𝑢⟩ + 𝑖|𝑑𝑑⟩) = (|𝑑𝑢⟩ + |𝑢𝑑⟩) = |𝑇1 ⟩
√2 √2 √2
1 1 1
𝜎𝑧 𝜏𝑧 |𝑇1 ⟩ = 𝜎𝑧 𝜏𝑧 (|𝑢𝑑⟩ + |𝑑𝑢⟩) = 𝜎𝑧 (−|𝑢𝑑⟩ + |𝑑𝑢⟩) = (−|𝑢𝑑⟩ − |𝑑𝑢⟩) = −|𝑇1 ⟩
√2 √2 √2
Result: 𝜎 ∙ 𝜏|𝑇1 ⟩ = |𝑇1 ⟩
|𝑇1 ⟩ is eigenvector of 𝜎 ∙ 𝜏 with eigenvalue 1.
page 66 of 105

1 1 1
𝜎𝑥 𝜏𝑥 |𝑇2 ⟩ = 𝜎𝑥 𝜏𝑥 (|𝑢𝑢⟩ + |𝑑𝑑⟩) = 𝜎𝑥 (|𝑢𝑑⟩ + |𝑑𝑢⟩) = (|𝑑𝑑⟩ + |𝑢𝑢⟩) = |𝑇2 ⟩
√2 √2 √2
1 1 1
𝜎𝑦 𝜏𝑦 |𝑇2 ⟩ = 𝜎𝑦 𝜏𝑦 (|𝑢𝑢⟩ + |𝑑𝑑⟩) = 𝜎𝑦 (𝑖|𝑢𝑑⟩ − 𝑖|𝑑𝑢⟩) = (−|𝑑𝑑⟩ − |𝑢𝑢⟩) = −|𝑇2 ⟩
√2 √2 √2
1 1 1
𝜎𝑧 𝜏𝑧 |𝑇2 ⟩ = 𝜎𝑧 𝜏𝑧 (|𝑢𝑢⟩ + |𝑑𝑑⟩) = 𝜎𝑧 (|𝑢𝑢⟩ − |𝑑𝑑⟩) = (|𝑢𝑢⟩ + |𝑑𝑑⟩) = |𝑇2 ⟩
√2 √2 √2
Result: 𝜎 ∙ 𝜏|𝑇2 ⟩ = |𝑇2 ⟩

1 1 1
𝜎𝑥 𝜏𝑥 |𝑇3 ⟩ = 𝜎𝑥 𝜏𝑥 (|𝑢𝑢⟩ − |𝑑𝑑⟩) = 𝜎𝑥 (|𝑢𝑑⟩ − |𝑑𝑢⟩) = (|𝑑𝑑⟩ − |𝑢𝑢⟩) = −|𝑇3 ⟩
√2 √2 √2
1 1 1
𝜎𝑦 𝜏𝑦 |𝑇3 ⟩ = 𝜎𝑦 𝜏𝑦 (|𝑢𝑢⟩ − |𝑑𝑑⟩) = 𝜎𝑦 (𝑖|𝑢𝑑⟩ + 𝑖|𝑑𝑢⟩) = (−|𝑑𝑑⟩ + |𝑢𝑢⟩) = |𝑇3 ⟩
√2 √2 √2
1 1 1
𝜎𝑧 𝜏𝑧 |𝑇3 ⟩ = 𝜎𝑧 𝜏𝑧 (|𝑢𝑢⟩ − |𝑑𝑑⟩) = 𝜎𝑧 (|𝑢𝑢⟩ + |𝑑𝑑⟩) = (|𝑢𝑢⟩ − |𝑑𝑑⟩) = |𝑇3 ⟩
√2 √2 √2
Result: 𝜎 ∙ 𝜏|𝑇3 ⟩ = |𝑇3 ⟩
|𝑠𝑖𝑛𝑔⟩ is eigenvector with eigenvalue 3,
|𝑇1 ⟩, |𝑇2 ⟩ und |𝑇3 ⟩ are eigenvectors with eigenvalue 1, they are degenerated.
page 67 of 105
Exercise 6.10
A system of two spins has the Hamiltonian:

𝜔ℏ
ℋ= 𝜎∙𝜏
2
Question 1)
What are the possible energies of the system, and what are the eigenvectors of the Hamiltonian?
Question 2)
Suppose the system starts in the state |𝑢𝑢⟩.
What is the state at any later time?
Answer the same question for initial states of |𝑢𝑑⟩, |𝑑𝑢⟩, |𝑑𝑑.
**********
Question 1)
The Hamilton-operator for the 2-spin-system:

𝜔ℏ
ℋ= 𝜎∙𝜏
2
From exercise 6.9 we know the eigenvectors and eigenvalues:
|𝑠𝑖𝑛𝑔⟩ with eigenvalue 3, |𝑇1 ⟩, |𝑇2 ⟩ und |𝑇3 ⟩ with eigenvalue 1.

3𝜔ℏ 𝜔ℏ
This gives the energies to 2
for |𝑠𝑖𝑛𝑔⟩ and each 2
for |𝑇1 ⟩, |𝑇2 ⟩ and |𝑇3 ⟩.
Question 2)
𝜔ℏ
ℋ|𝑢𝑢⟩ = 𝜎 ∙ 𝜏|𝑢𝑢⟩
2
𝜔ℏ
We use 𝜎 ∙ 𝜏 = 𝜎𝑥 𝜏𝑥 + 𝜎𝑦 𝜏𝑦 + 𝜎𝑧 𝜏𝑧 and apply this to |𝑢𝑢⟩ (we omit the factor 2
and add it in the
end):
𝜔ℏ 𝜔ℏ
𝜎𝑥 𝜏𝑥 |𝑢𝑢⟩ = 𝜎𝑥 |𝑢𝑑⟩ = |𝑑𝑑⟩, gives 𝜎 𝜏 |𝑢𝑢⟩
2 𝑥 𝑥
= 2
|𝑑𝑑⟩
𝜔ℏ 𝜔ℏ
𝜎𝑦 𝜏𝑦 |𝑢𝑢⟩ = 𝜎𝑦 𝑖|𝑢𝑑⟩ = −|𝑑𝑑⟩, gives 𝜎 𝜏 |𝑢𝑢⟩
2 𝑦 𝑦
=− 2
|𝑑𝑑⟩
𝜔ℏ 𝜔ℏ
𝜎𝑧 𝜏𝑧 |𝑢𝑢⟩ = 𝜎𝑧 |𝑢𝑢⟩ = |𝑢𝑢⟩, gives 𝜎 𝜏 |𝑢𝑢⟩ = |𝑢𝑢⟩
2 𝑧 𝑧 2
Building the sum, we get:

𝜔ℏ
ℋ|𝑢𝑢⟩ = |𝑢𝑢⟩
2
The other combinations analog:
𝜔ℏ
ℋ|𝑢𝑑⟩ = 𝜎 ∙ 𝜏|𝑢𝑑⟩
2
𝜔ℏ 𝜔ℏ
𝜎𝑥 𝜏𝑥 |𝑢𝑑⟩ = 𝜎𝑥 |𝑢𝑢⟩ = |𝑑𝑢⟩, gives 𝜎 𝜏 |𝑢𝑑⟩
2 𝑥 𝑥
= 2
|𝑑𝑢⟩
page 68 of 105
𝜔ℏ 𝜔ℏ
𝜎𝑦 𝜏𝑦 |𝑢𝑑⟩ = 𝜎𝑦 − 𝑖|𝑢𝑢⟩ = |𝑑𝑢⟩, gives 𝜎 𝜏 |𝑢𝑑⟩ = |𝑑𝑢⟩
2 𝑦 𝑦 2
𝜔ℏ 𝜔ℏ
𝜎𝑧 𝜏𝑧 |𝑢𝑑⟩ = 𝜎𝑧 − |𝑢𝑑⟩ = −|𝑢𝑑⟩, gives 𝜎 𝜏 |𝑢𝑑⟩
2 𝑧 𝑧
= 2
|𝑢𝑑⟩

𝜔ℏ
ℋ|𝑢𝑑⟩ = 𝜔ℏ|𝑑𝑢⟩ + |𝑢𝑑⟩
2
𝜔ℏ
ℋ|𝑑𝑢⟩ = 𝜎 ∙ 𝜏|𝑑𝑢⟩
2
𝜔ℏ 𝜔ℏ
𝜎𝑥 𝜏𝑥 |𝑑𝑢⟩ = 𝜎𝑥 |𝑑𝑑⟩ = |𝑢𝑑⟩, gives 𝜎 𝜏 |𝑑𝑢⟩
2 𝑥 𝑥
= 2
|𝑢𝑑⟩
𝜔ℏ 𝜔ℏ
𝜎𝑦 𝜏𝑦 |𝑑𝑢⟩ = 𝜎𝑦 𝑖|𝑑𝑑⟩ = |𝑢𝑑⟩, gives 𝜎 𝜏 |𝑑𝑢⟩
2 𝑦 𝑦
= 2
|𝑢𝑑⟩
𝜔ℏ 𝜔ℏ
𝜎𝑧 𝜏𝑧 |𝑑𝑢⟩ = 𝜎𝑧 |𝑑𝑢⟩ = −|𝑑𝑢⟩, gives 𝜎 𝜏 |𝑑𝑢⟩
2 𝑧 𝑧
=− 2
|𝑑𝑢⟩

𝜔ℏ
ℋ|𝑑𝑢⟩ = 𝜔ℏ|𝑢𝑑⟩ − |𝑑𝑢⟩
2
𝜔ℏ
ℋ|𝑑𝑑⟩ = 𝜎 ∙ 𝜏|𝑑𝑑⟩
2
𝜔ℏ 𝜔ℏ
𝜎𝑥 𝜏𝑥 |𝑑𝑑⟩ = 𝜎𝑥 |𝑑𝑢⟩ = |𝑢𝑢⟩, gives 𝜎 𝜏 |𝑑𝑑⟩
2 𝑥 𝑥
= 2
|𝑢𝑢⟩
𝜔ℏ 𝜔ℏ
𝜎𝑦 𝜏𝑦 |𝑑𝑑⟩ = 𝜎𝑦 − 𝑖|𝑑𝑢⟩ = |𝑢𝑢⟩, gives 𝜎 𝜏 |𝑑𝑑⟩
2 𝑦 𝑦
= 2
|𝑢𝑢
𝜔ℏ 𝜔ℏ
𝜎𝑧 𝜏𝑧 |𝑑𝑑⟩ = 𝜎𝑧 − |𝑑𝑑⟩ = |𝑑𝑑⟩, gives 𝜎 𝜏 |𝑑𝑑⟩ = |𝑑𝑑⟩
2 𝑧 𝑧 2

𝜔ℏ
ℋ|𝑑𝑑⟩ = 𝜔ℏ|𝑢𝑑⟩ + |𝑑𝑑⟩
2
page 69 of 105
Question 2 targets to chapter 4 with the aim of developing the time depend Schrödinger-equation.
We use “4.13 Recipe for a Schrödinger Ket” to perform this.
Recipe:
1. The Hamilton-operator:
𝜔ℏ
ℋ= 𝜎∙𝜏
2
2. We prepare an initial state |𝜓(0)⟩:
|𝜓(0)⟩ = |𝑢𝑢⟩
3. We calculate eigenvectors and eigenvalues of H:
1
𝐸1 ≔ |𝑠𝑖𝑛𝑔⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩) with eigenvalue 3
√2
1
𝐸2 ≔ |𝑇1 ⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩) with eigenvalue 1
√2
1
𝐸3 ≔ |𝑇2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩) with eigenvalue 1
√2
1
𝐸4 ≔ |𝑇3 ⟩ = (|𝑢𝑢⟩ − |𝑑𝑑⟩) with eigenvalue 1
√2
4. We calculate the initial coefficients 𝛼𝑗 (0) = ⟨𝐸𝑗 |𝜓(0)⟩
𝛼1 (0) = 0:
1
𝛼1 (0) = ⟨𝐸1 |𝜓(0)⟩ = ⟨𝑠𝑖𝑛𝑔|𝑢𝑢⟩ = (|𝑢𝑑⟩ − |𝑑𝑢⟩)|𝑢𝑢⟩ =
√2
1 1 1
= (⟨𝑢𝑑 − ⟨𝑑𝑢)|𝑢𝑢⟩ = (⟨𝑢𝑑|𝑢𝑢⟩ − ⟨𝑑𝑢|𝑢𝑢⟩) = (0 − 0) = 0
√2 √2 √2
𝛼2 (0) = 0:
1
𝛼2 (0) = ⟨𝐸2 |𝜓(0)⟩ = ⟨𝑇1 |𝑢𝑢⟩ = (|𝑢𝑑⟩ + |𝑑𝑢⟩)|𝑢𝑢⟩ =
√2
1 1 1
= (⟨𝑢𝑑 + ⟨𝑑𝑢)|𝑢𝑢⟩ = (⟨𝑢𝑑|𝑢𝑢⟩ + ⟨𝑑𝑢|𝑢𝑢⟩) = (0 + 0) = 0
√2 √2 √2
1
𝛼3 (0) = :
√2
1
𝛼3 (0) = ⟨𝐸3 |𝜓(0)⟩ = ⟨𝑇2 |𝑢𝑢⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)|𝑢𝑢⟩ =
√2
1 1 1 1
= (⟨𝑢𝑢 + ⟨𝑑𝑑)|𝑢𝑢⟩ = (⟨𝑢𝑢|𝑢𝑢⟩ + ⟨𝑑𝑑|𝑢𝑢⟩) = (1 + 0) =
√2 √2 √2 √2
page 70 of 105
1
𝛼4 (0) = :
√2
𝛼4 (0) = ⟨𝐸4 |𝜓(0)⟩ = ⟨𝑇3 |𝑢𝑢⟩ =
1 1 1 1
= (⟨𝑢𝑢| − ⟨𝑑𝑑)|𝑢𝑢⟩ = (⟨𝑢𝑢|𝑢𝑢⟩ − ⟨𝑑𝑑|𝑢𝑢⟩) = (1 − 0) =
√2 √2 √2 √2
5. We rewrite |𝜓(0)⟩ in terms of the eigenvectors |𝐸𝑗 ⟩ and the initial coefficients 𝛼𝑗 (0):
4
1 1
|𝜓(0)⟩ = ∑ 𝛼𝑗 (0)| 𝐸𝑗 ⟩ = 0|𝑠𝑖𝑛𝑔⟩ + 0|𝑇1 ⟩ + |𝑇2 ⟩ + |𝑇3 ⟩
𝑗=1
√2 √2
6. We replace each 𝛼𝑗 (0) with 𝛼𝑗 (𝑡), as a result |𝜓(0)⟩ becomes |𝜓(𝑡)⟩.

4
|𝜓(𝑡)⟩ = ∑ 𝛼𝑗 (𝑡)| 𝐸𝑗 ⟩
𝑗=1
𝑖
7. We use equation 4.30: 𝛼𝑗 (𝑡) = 𝛼𝑗 (0)𝑒 −ℏ𝐸𝑗(𝑡) and replace each 𝛼𝑗 (𝑡) with equation 4.30, using the
basic vectors (eigenvectors):
4
𝑖
|𝜓(𝑡)⟩ = ∑ 𝛼𝑗 (0)𝑒 −ℏ𝐸𝑗(𝑡) | 𝐸𝑗 ⟩
𝑗=1
1
We use: 𝛼1 (0) = 𝛼2 (0) = 0, 𝛼3 (0) = 𝛼4 (0) = , 𝐸1 = 3, 𝐸2 , 𝐸3 , 𝐸4 = 1
√2
1 𝑖 1 𝑖
|𝜓(𝑡)⟩ = 0|𝑠𝑖𝑛𝑔⟩ + 0|𝑇1 ⟩ + 𝑒 −ℏ𝑡 |𝑇2 ⟩ + 𝑒 −ℏ𝑡 |𝑇3 ⟩ =
√2 √2
1 𝑖
− 𝑡 1 𝑖
− 𝑡
= 𝑒 ℏ |𝑇2 ⟩ + 𝑒 ℏ |𝑇3 ⟩ =
√2 √2
1 −𝑖 𝑡
= 𝑒 ℏ (|𝑢𝑢⟩ + |𝑑𝑑⟩ + |𝑢𝑢⟩ − |𝑑𝑑⟩) =
2
2 𝑖
= 𝑒 −ℏ𝑡 (|𝑢𝑢⟩)
2
Summed up:
𝑖
|𝜓(𝑡)⟩ = 𝑒 −ℏ𝑡 (|𝑢𝑢⟩)
In the same way we calculate the other initial states |𝑢𝑑⟩, |𝑑𝑢⟩ 𝑢𝑛𝑑 |𝑑𝑑⟩.
Initial state |𝒖𝒅⟩:

1
𝛼1 (0) = = 𝛼2 (0) and 𝛼3 (0) = 0 = 𝛼4 (0), because of the products (⟨𝑢𝑢| − ⟨𝑑𝑑|)|𝑢𝑑⟩ etc.
√2
vanishing resp. resulting in 1.
page 71 of 105
1 3𝑖
− 𝑡 1 𝑖
− 𝑡
|𝜓(𝑡)⟩ = 𝑒 ℏ |𝑠𝑖𝑛𝑔⟩ + 𝑒 ℏ | 𝑇1 ⟩ + 0|𝑇2 ⟩ + 0|𝑇3 ⟩ =
√2 √2
1 3𝑖 1 𝑖
= 𝑒 − ℏ 𝑡 |𝑠𝑖𝑛𝑔⟩ + 𝑒 −ℏ𝑡 | 𝑇1 ⟩ =
2 2
1 3𝑖 𝑖
= (𝑒 − ℏ 𝑡 (|𝑢𝑑⟩ − |𝑑𝑢⟩) + 𝑒 −ℏ𝑡 (|𝑢𝑑⟩ + |𝑑𝑢⟩)) =
2
1 𝑖 2𝑖
= 𝑒 −ℏ𝑡 (𝑒 − ℏ 𝑡 (|𝑢𝑑⟩ − |𝑑𝑢⟩) + (|𝑢𝑑⟩ + |𝑑𝑢⟩))
2
Initial state |𝒅𝒖⟩:

1 1
𝛼1 (0) = − , 𝛼2 (0) = and 𝛼3 (0) = 0 = 𝛼4 (0), because of the products (⟨𝑢𝑢| − ⟨𝑑𝑑|)|𝑑𝑢⟩ etc.
√2 √2
1 3𝑖 1 𝑖
|𝜓(𝑡)⟩ = − 𝑒 − ℏ 𝑡 |𝑠𝑖𝑛𝑔⟩ + 𝑒 −ℏ𝑡 | 𝑇1 ⟩ + 0|𝑇2 ⟩ + 0|𝑇3 ⟩ =
√2 √2
1 3𝑖 1 𝑖
= − 𝑒 − ℏ 𝑡 |𝑠𝑖𝑛𝑔⟩ + 𝑒 −ℏ𝑡 | 𝑇1 ⟩ =
2 2
1 3𝑖 𝑖
= (𝑒 − ℏ 𝑡 (−|𝑢𝑑⟩ + |𝑑𝑢⟩) + 𝑒 −ℏ𝑡 (|𝑢𝑑⟩ + |𝑑𝑢⟩)) =
2
1 𝑖 2𝑖
= 𝑒 −ℏ𝑡 (𝑒 − ℏ 𝑡 (−|𝑢𝑑⟩ + |𝑑𝑢⟩) + (|𝑢𝑑⟩ + |𝑑𝑢⟩))
2
Initial state |𝒅𝒅⟩:

1 1
𝛼3 (0) = , 𝛼4 (0) = − und 𝛼1 (0) = 0 = 𝛼2 (0), because of the products (⟨𝑢𝑢| − ⟨𝑑𝑑|)|𝑑𝑑⟩ etc.
√2 √2
1 𝑖 1 𝑖
|𝜓(𝑡)⟩ = 0|𝑠𝑖𝑛𝑔⟩ + 0|𝑇1 ⟩ + 𝑒 −ℏ𝑡 |𝑇2 ⟩ − 𝑒 −ℏ𝑡 |𝑇3 ⟩ =
√2 √2
1 𝑖 1 𝑖
= 𝑒 −ℏ𝑡 |𝑇2 ⟩ − 𝑒 −ℏ𝑡 |𝑇3 ⟩ =
√2 √2
1 𝑖
= 𝑒 −ℏ𝑡 (|𝑢𝑢⟩ + |𝑑𝑑⟩ − |𝑢𝑢⟩ + |𝑑𝑑⟩) =
2
𝑖
= 𝑒 −ℏ𝑡 (|𝑑𝑑⟩)
𝑖
Note: for the states |uu⟩ and |dd⟩ 𝑒 −ℏ𝑡 will give them the shape of a
𝑖 𝑖
wave: 𝑒 −ℏ𝑡 (|𝑢𝑢⟩) → 𝑅𝑒 (𝑒 −ℏ𝑡 ) (|𝑢𝑢⟩)~cos (𝑡)|𝑢𝑢⟩.
For the states (|𝑢𝑑⟩) and (|𝑑𝑢⟩) we get another behavior.
1 − 𝑖 𝑡 −2𝑖𝑡
𝑒 ℏ (𝑒 ℏ (|𝑢𝑑⟩ − |𝑑𝑢⟩) + (|𝑢𝑑⟩ + |𝑑𝑢⟩))
2
starts at time 𝑡 = 0 in the state |𝑢𝑑⟩ but then begins oscillating between |𝑢𝑑⟩ and |𝑑𝑢⟩. The picture
might illustrate this.
page 72 of 105
Exercise 7.1
Write the tensor product 𝐼 ⨂ 𝜏𝑥 as a matrix, and apply that matrix to each of the |𝑢𝑢⟩, |𝑢𝑑⟩, |𝑑𝑢⟩
and |𝑑𝑑⟩ column vectors. Show that Alice’s half of the state-vector is unchanged in each case. Recall
that 𝐼 is the 2 × 2 unit matrix.
**********
0 1 0 1 0 1 0 0
1( ) 0( )
1 0 0 1 1 0 ) = (1 0 0 0
I ⊗ 𝜏𝑥 = ( )⊗( )=( 1 0 )
0 1 1 0 0 1 0 1 0 0 0 1
0( ) 1( )
1 0 1 0 0 0 1 0
1
0
applied to: |𝑢𝑢⟩ = ( )
0
0
0 1 0 0 1 0
(I ⊗ 𝜏𝑥 )|𝑢𝑢⟩ = (1 0 0 0 0 1
) ( ) = ( ) = |𝑢𝑑⟩
0 0 0 1 0 0
0 0 1 0 0 0
0
1
applied to: |𝑢𝑑⟩ = ( )
0
0
0 1 0 0 0 1
1 0 0 0 1 0
(I ⊗ 𝜏𝑥 )|𝑢𝑑⟩ = ( ) ( ) = ( ) = |𝑢𝑢⟩
0 0 0 1 0 0
0 0 1 0 0 0
0
0
applied to: |𝑑𝑢⟩ = ( )
1
0
0 1 0 0 0 0
(I ⊗ 𝜏𝑥 )|𝑑𝑢⟩ = (1 0 0 0 0 0
) ( ) = ( ) = |𝑑𝑑⟩
0 0 0 1 1 0
0 0 1 0 0 1
0
0
applied to: |𝑑𝑑⟩ = ( )
0
1
0 1 0 0 0 0
1 0 0 0 0 0
(I ⊗ 𝜏𝑥 )|𝑑𝑑⟩ = ( ) ( ) = ( ) = |𝑑𝑢⟩
0 0 0 1 0 1
0 0 1 0 1 0
page 73 of 105
Exercise 7.2
Calculate the matrix elements of 𝜎𝑧 ⨂ 𝜏𝑥 by forming the inner product analog to:
⟨𝑢𝑢|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝐼|𝑑𝑑⟩
⟨𝑢𝑑|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝐼|𝑑𝑑⟩
𝜎𝑧 ⨂ 𝐼 = ( )
⟨𝑑𝑢|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝐼|𝑑𝑑⟩
⟨𝑑𝑑|𝜎𝑧 𝐼|𝑢𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑢𝑑⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑑𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝐼|𝑑𝑑⟩
𝜎𝑧 operates to the left, 𝐼 to the right.
**********
𝜎𝑧 ⊗ 𝜏𝑥 according to the rules of the tensor product:
0 1 0 1 0 1 0 0
1( ) 0( )
1 0 0 1 0 ) = (1 0 0 0
𝜎𝑧 ⊗ 𝜏𝑥 = ( )⊗( )=( 1 0 1 )
0 −1 1 0 0 1 0 1 0 0 0 −1
0( ) −1 ( )
1 0 1 0 0 0 −1 0
𝜎𝑧 ⊗ 𝜏𝑥 :
⟨𝑢𝑢|𝜎𝑧 𝜏𝑥 |𝑢𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝜏𝑥 |𝑢𝑑⟩ ⟨𝑢𝑢|𝜎𝑧 𝜏𝑥 |𝑑𝑢⟩ ⟨𝑢𝑢|𝜎𝑧 𝜏𝑥 |𝑑𝑑⟩
⟨𝑢𝑑|𝜎𝑧 𝜏𝑥 |𝑢𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝜏𝑥 |𝑢𝑑⟩ ⟨𝑢𝑑|𝜎𝑧 𝜏𝑥 |𝑑𝑢⟩ ⟨𝑢𝑑|𝜎𝑧 𝜏𝑥 |𝑑𝑑⟩
(𝜎𝑧 ⊗ 𝜏𝑥 ) = =
⟨𝑑𝑢|𝜎𝑧 𝜏𝑥 |𝑢𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝜏𝑥 |𝑢𝑑⟩ ⟨𝑑𝑢|𝜎𝑧 𝜏𝑥 |𝑑𝑢⟩ ⟨𝑑𝑢|𝜎𝑧 𝜏𝑥 |𝑑𝑑⟩
(⟨𝑑𝑑|𝜎𝑧 𝜏𝑥 |𝑢𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝜏𝑥 |𝑢𝑑⟩ ⟨𝑑𝑑|𝜎𝑧 𝜏𝑥 |𝑑𝑢⟩ ⟨𝑑𝑑|𝜎𝑧 𝜏𝑥 |𝑑𝑑⟩)
𝜎𝑧 applied to the left, 𝜏𝑥 applied to the right:

⟨𝑢𝑢|𝑢𝑑⟩ ⟨𝑢𝑢|𝑢𝑢⟩ ⟨𝑢𝑢|𝑑𝑑⟩ ⟨𝑢𝑢|𝑑𝑢⟩
⟨𝑢𝑑|𝑢𝑑⟩ ⟨𝑢𝑑|𝑢𝑢⟩ ⟨𝑢𝑑|𝑑𝑑⟩ ⟨𝑢𝑑|𝑑𝑢⟩
=( )
⟨−𝑑𝑢|𝑢𝑑⟩ ⟨−𝑑𝑢|𝑢𝑢⟩ ⟨−𝑑𝑢|𝑑𝑑⟩ ⟨−𝑑𝑢|𝑑𝑢⟩
⟨−𝑑𝑑|𝑢𝑑⟩ ⟨−𝑑𝑑|𝑢𝑢⟩ ⟨−𝑑𝑑|𝑑𝑑⟩ ⟨−𝑑𝑑|𝑑𝑢⟩
Because of u and d being orthonormal vectors, this condenses to:
0 1 0 0
1 0 0 0
( )
0 0 0 −1
0 0 −1 0
This gives the same matrix as the tensor-product above.
page 74 of 105
Exercise 7.3
Consider the equation:
(𝐴 ⨂ 𝐵)(𝑎 ⨂ 𝑏) = (𝐴𝑎 ⨂ 𝐵𝑏)
𝐴 and 𝐵 represent 2 × 2 matrices (or operators), and 𝐴 and 𝑏 represent 2 × 1 column vectors.
a) Rewrite the equation
(𝐴 ⨂ 𝐵)(𝑎 ⨂ 𝑏) = (𝐴𝑎 ⨂ 𝐵𝑏)
in component form, replacing the symbols 𝐴, 𝐵, 𝑎, 𝑏 with the following matrices and column vectors:
𝐴11 𝐵11 𝐴11 𝐵12 𝐴12 𝐵11 𝐴12 𝐵12
𝐴11 𝐵21 𝐴11 𝐵22 𝐴12 𝐵21 𝐴12 𝐵22
𝐴⨂𝐵 =( )
𝐴21 𝐵11 𝐴21 𝐵12 𝐴22 𝐵11 𝐴22 𝐵12
𝐴21 𝐵21 𝐴21 𝐵22 𝐴22 𝐵21 𝐴22 𝐵22
𝑎11 𝑏11
𝑎11 𝑏11 𝑎11 𝑏21
(𝑎 ) ⨂ ( ) = ( )
21 𝑏21 𝑎21 𝑏11
𝑎21 𝑏21
b) Perform the matrix multiplication 𝐴𝑎 and 𝐵𝑏 on the right-hand side. Verify that each result is a
4 × 1 matrix.
c) Expand all three Kronecker products.
d) Verify the row and column sizes of each Kronecker product:
• 𝐴⨂𝐵:4 × 4
• 𝑎⨂𝑏:4 × 1
• 𝐴𝑎 ⨂ 𝐵𝑏 : 4 × 1
e) Perform the matrix multiplication on the left-hand side, resulting in a 4 × 1 column vector. Each
row should be the sum of four separate terms
f) Finally, verify that the resulting column vectors on the left and right sides are identical.
**********
a) Rewrite (A ⊗ B)(𝑎 ⊗ b) = (Aa ⊗ Bb) in component form:

𝐵 𝐵12 𝐵 𝐵12
𝐴11 ( 11 ) 𝐴12 ( 11 )
𝐴 𝐴12 𝐵 𝐵12 𝐵21 𝐵22 𝐵21 𝐵22
𝐴 ⊗ B = ( 11 ) ⊗ ( 11 )=( )=
𝐴21 𝐴22 𝐵21 𝐵22 𝐵 𝐵12 𝐵 𝐵12
𝐴21 ( 11 ) 𝐴22 ( 11 )
𝐵21 𝐵22 𝐵21 𝐵22
𝐴11 𝐵11 𝐴11 𝐵12 𝐴12 𝐵11 𝐴12 𝐵12
𝐴11 𝐵21 𝐴11 𝐵22 𝐴12 𝐵21 𝐴12 𝐵22
=( )
𝐴21 𝐵11 𝐴21 𝐵12 𝐴22 𝐵11 𝐴22 𝐵12
𝐴21 𝐵21 𝐴21 𝐵22 𝐴22 𝐵21 𝐴22 𝐵22
𝑎11 𝑏11
𝑎11 𝑏11 𝑎11 𝑏21
(𝑎 ⊗ b) = (𝑎 ) ⊗ ( ) = ( )
21 𝑏21 𝑎21 𝑏11
𝑎21 𝑏21
page 75 of 105
b) Perform the matrix multiplication:

𝐴11 𝐴12 𝑎11 𝐴 𝑎 + 𝐴12 𝑎21
Aa = ( ) ( ) = ( 11 11 )
𝐴21 𝐴22 𝑎21 𝐴21 𝑎11 + 𝐴22 𝑎21
𝐵 𝐵12 𝑏11 𝐵 𝑏 + 𝐵12 𝑏21
Bb = ( 11 ) ( ) = ( 11 11 )
𝐵21 𝐵22 𝑏21 𝐵21 𝑏11 + 𝐵22 𝑏21
(𝐴11 𝑎11 + 𝐴12 𝑎21 )(𝐵11 𝑏11 + 𝐵12 𝑏21 )
(𝐴11 𝑎11 + 𝐴12 𝑎21 )(𝐵21 𝑏11 + 𝐵22 𝑏21 )
(Aa ⊗ Bb) = =
(𝐴21 𝑎11 + 𝐴22 𝑎21 )(𝐵11 𝑏11 + 𝐵12 𝑏21 )
((𝐴21 𝑎11 + 𝐴22 𝑎21 )(𝐵21 𝑏11 + 𝐵22 𝑏21 ))
𝐴11 𝑎11 𝐵11 𝑏11 + 𝐴11 𝑎11 𝐵12 𝑏21 + 𝐴12 𝑎21 𝐵11 𝑏11 + 𝐴12 𝑎21 𝐵12 𝑏21
=( )
c) Expand all three Kronecker products: (A ⊗ B), (𝑎 ⊗ b) 𝑎𝑛𝑑 (Aa ⊗ Bb)
… already done …
d) Verify:
𝐴 ⊗ B = 4 x 4, (𝑎 ⊗ b) = 4 x 1, (Aa ⊗ Bb) = 4 x 1
e) Perform the matrix multiplication (A ⊗ B)(𝑎 ⊗ b):

𝐴11 𝐵11 𝐴11 𝐵12 𝐴12 𝐵11 𝐴12 𝐵12 𝑎11 𝑏11
𝐴11 𝐵21 𝐴11 𝐵22 𝐴12 𝐵21 𝐴12 𝐵22 𝑎11 𝑏21
=( )( )=
𝐴21 𝐵11 𝐴21 𝐵12 𝐴22 𝐵11 𝐴22 𝐵12 𝑎21 𝑏11
𝐴21 𝐵21 𝐴21 𝐵22 𝐴22 𝐵21 𝐴22 𝐵22 𝑎21 𝑏21
𝐴11 𝐵11 𝑎11 𝑏11 + 𝐴11 𝐵12 𝑎11 𝑏21 + 𝐴12 𝐵11 𝑎21 𝑏11 + 𝐴12 𝐵12 𝑎21 𝑏21
𝐴 𝐵 𝑎 𝑏 + 𝐴11 𝐵22 𝑎11 𝑏21 + 𝐴12 𝐵21 𝑎21 𝑏11 + 𝐴12 𝐵22 𝑎21 𝑏21
= ( 11 21 11 11 )
f) Finally, verify the identity:
Left side:
=( )
Right side:
=( )
page 76 of 105
Exercise 7.4
Calculate the density matrix for |Ψ⟩ = 𝛼|𝑢𝑢⟩ + 𝛽|𝑢𝑢⟩.
Answer:
𝜓(𝑢) = 𝛼; 𝜓 ∗ (𝑢) = 𝛼 ∗
𝜓(𝑑) = 𝛽; 𝜓 ∗ (𝑑) = 𝛽 ∗
𝛼 ∗𝛼 𝛼 ∗𝛽
𝜌𝑎′𝑎 = ( )
𝛽∗𝛼 𝛽∗𝛽
Now try plugging some numbers for 𝛼 and 𝛽.
1 1
Make sure they are normalized to 1. For example, 𝛼 = ,𝛽 =
√2 √2
**********
The density matrix is defined as
𝜌𝑎𝑎′ = ⟨𝑎|Ψ⟩⟨Ψ|𝑎′⟩
with the values of a and a’ being u and d (4 possible combinations).
We calculate the density matrix for:
|Ψ⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩
We need the bra too:
⟨Ψ|= ⟨𝑢|𝛼 ∗ + ⟨𝑢|𝛽 ∗
Then we calculate the elements of the density matrix.
𝜌𝑢𝑢
𝜌𝑢𝑢 = ⟨𝑢|Ψ⟩⟨Ψ|𝑢⟩ =
𝜌𝑢𝑢 = ⟨𝑢(𝛼|𝑢⟩ + 𝛽|𝑑⟩)⟩⟨(⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑢⟩ =
(𝛼⟨𝑢|𝑢⟩ + 𝛽⟨𝑢|𝑑⟩)(⟨𝑢|𝑢⟩𝛼 ∗ + ⟨𝑑|𝑢⟩𝛽 ∗ ) =
(𝛼 + 0)(𝛼 ∗ + 0) = 𝛼𝛼 ∗
𝜌𝑢𝑑
𝜌𝑢𝑑 = ⟨𝑢|Ψ⟩⟨Ψ|𝑑⟩ =
𝜌𝑢𝑑 = ⟨𝑢(𝛼|𝑢⟩ + 𝛽|𝑑⟩)⟩⟨(⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑑⟩ =
(𝛼⟨𝑢|𝑢⟩ + 𝛽⟨𝑢|𝑑⟩)(⟨𝑢|𝑑⟩𝛼 ∗ + ⟨𝑑|𝑑⟩𝛽 ∗ ) =
(𝛼 + 0)(0 + 𝛽 ∗ ) = 𝛼𝛽 ∗
page 77 of 105
𝜌𝑑𝑢
𝜌𝑑𝑢 = ⟨𝑑|Ψ⟩⟨Ψ|𝑢⟩ =
𝜌𝑑𝑢 = ⟨𝑑(𝛼|𝑢⟩ + 𝛽|𝑑⟩)⟩⟨(⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑢⟩ =
(𝛼⟨𝑑|𝑢⟩ + 𝛽⟨𝑑|𝑑⟩)(⟨𝑢|𝑢⟩𝛼 ∗ + ⟨𝑑|𝑢⟩𝛽 ∗ ) =
(0 + 𝛽)(𝛼 ∗ + 0) = 𝛽𝛼 ∗
𝜌𝑑𝑑
𝜌𝑑𝑑 = ⟨𝑑|Ψ⟩⟨Ψ|𝑑⟩ =
𝜌𝑑𝑑 = ⟨𝑑(𝛼|𝑢⟩ + 𝛽|𝑑⟩)⟩⟨(⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗ )|𝑑⟩ =
(𝛼⟨𝑑|𝑢⟩ + 𝛽⟨𝑑|𝑑⟩)(⟨𝑢|𝑑⟩𝛼 ∗ + ⟨𝑑|𝑑⟩𝛽 ∗ ) =
(0 + 𝛽)(0 + 𝛽 ∗ ) = 𝛽𝛽 ∗
The density matrix goes like this:

𝛼𝛼 ∗ 𝛼𝛽 ∗
𝜌𝑎𝑎′ = ( ∗ )
𝛽𝛼 𝛽𝛽 ∗
1 1
We plug in the example 𝛼 = and 𝛽 = and get:
√2 √2
1 1 1 1 1 1
𝜌𝑎𝑎′ = √2 √2 √2 √2 = (2 2)
1 1 1 1 1 1
2 2
(√2 √2 √2 √2)
1 √63
Choosing the values 𝛼 = 8 and 𝛽 = 8
(designed to fit 𝛼 2 + 𝛽 2 = 1):
11 1 √63 1 √63
𝜌𝑎𝑎′ = 88 8 8 = 64 64
√63 1 √63 √63 √63 63
( 8 8 8 8 ) ( 64 64 )
page 78 of 105
Exercise 7.5
a) Show:
𝑎 0 2 2
( ) = (𝑎 0)
0 𝑏 0 𝑏2
b) Now, suppose:
1⁄ 0
𝜌=( 3 )
0 2⁄3
Calculate:
𝜌2 , 𝑇𝑟(𝜌), 𝑇𝑟(𝜌2 )
c) If 𝜌 is a density matrix, does it represent a pure state or a mixed state?
**********
a) Show that:
𝑎 0 2 2
( ) = (𝑎 0)
0 𝑏 0 𝑏2
𝑎 0 2 𝑎 0 𝑎 0 𝑎∙𝑎+0∙0 𝑎∙0+0∙𝑏 2
( ) =( )( )=( ) = (𝑎 0)
0 𝑏 0 𝑏 0 𝑏 0∙𝑎+𝑏∙𝑎 0∙0+𝑏∙𝑏 0 𝑏2
b) Now, suppose
1
0
𝜌 = (3 )
2
0
3
Calculate 𝜌2 , 𝑇𝑟(𝜌), 𝑇𝑟(𝜌2 )
1 1 1
0 0 0
𝜌2 = 𝜌 ∙ 𝜌 = (3 ) ∙ (3 ) = (9 )
2 2 4
0 0 0
3 3 9
1
0 1 2 3
𝑇𝑟(𝜌) = 𝑇𝑟 (3 )= + = =1
2 3 3 3
0
3
1
0 1 4 5
𝑇𝑟(𝜌2 ) = 𝑇𝑟 (9 )= + = ≠1
4 9 9 9
0
9
c) If 𝜌 ist a density matrix, does it represent a pure stat or a mixed state?
A pure state is defined as a state in which 𝜌 is nonzero at only one point. This corresponds to a matrix
where only one entry on the diagonal is nonzero. As the sum of the diagonals need to be one for the
matrix to be a density matrix, this one entry must have the value 1 with the rest of all entries being
zero. So, the matrix 𝜌 presents a mixed state.
page 79 of 105
Exercise 7.6
By the standard rules of probability, the probability for a:
𝑃(𝑎) = ∑ 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎, 𝑏)

𝑏
This is just a diagonal entry in the density matrix:
𝑃(𝑎) = 𝜌𝑎𝑎
Use 𝑃(𝑎) = 𝜌𝑎𝑎 to show that if 𝜌 is a density matrix, then:
𝑇𝑟(𝜌) = 1
**********
Say that Alice’s system has more than one state but the states 𝑎1 , 𝑎2 , … , 𝑎𝑛 . The probability for each
one calculated according to 𝑃(𝑎) = 𝜌𝑎𝑎 is
𝑃(𝑎𝑖 ) = ∑ 𝜓 ∗ (𝑎𝑖 , 𝑏)𝜓(𝑎𝑖 , 𝑏)
𝑏
As Alice’s system must have a state (it couldn’t vanish), the sum over all possible states must be 1:
∑ 𝑃(𝑎𝑖 ) = 1
𝑖
Every probability 𝑃(𝑎𝑖 ) corresponds to a diagonal entry 𝜌𝑎𝑖𝑎𝑖 , so the sum over all diagonal elements
in the matrix must be 1. For a pure state with only one entry 𝜌𝑎𝑖𝑎𝑖 ≠ 0 this means that this entry
must be 1 too.
page 80 of 105
Exercise 7.7
We have the density matrix:

1⁄ 0
𝜌=( 2 )
0 1⁄2
Calculate 𝜌2 . How does this result confirm that 𝜌 represents an entangled state?
**********
1 2
1 1 1 2 1
0 0 0 ( ) 0 0
𝜌 2 = (2 ) = (2 ) ∙ (2 )= 2 = (4 )
1 1 1 1 2 1
0 0 0 0 ( ) 0
2 2 2 2 ) 4
(
In a mixed state more than one entry on the diagonal will be nonzero with the sum giving 1.
Only for pure states the equation 𝜌2 = 𝜌 holds, because the one and only “1” replicates itself.
page 81 of 105
Exercise 7.8
Consider the following states:

1
|𝜓1 ⟩ = (|𝑢𝑢⟩ + |𝑢𝑑⟩ + |𝑑𝑢⟩ + |𝑑𝑑⟩)
2
1
|𝜓2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩)
√2
1
|𝜓3 ⟩ = (3|𝑢𝑢⟩ + 4|𝑑𝑑⟩)
5
For each one, calculate Alice’s density matrix and Bob’s density matrix. Check their properties.
**********
Alice’s density matrix: calculate the sum 𝜌𝑎′𝑎 = ∑𝑏 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎′, 𝑏)
First, we expand the states |𝜓2 ⟩ and |𝜓3 ⟩ to the full set and get:
1 1 1 1
|𝜓1 ⟩ = |𝑢𝑢⟩ + |𝑢𝑑⟩ + |𝑑𝑢⟩ + |𝑑𝑑⟩
2 2 2 2
1 1
|𝜓2 ⟩ = |𝑢𝑢⟩ + |𝑑𝑑⟩ + 0|𝑢𝑑⟩ + 0|𝑑𝑢⟩
√2 √2
3 4
|𝜓3 ⟩ = |𝑢𝑢⟩ + |𝑢𝑑⟩ + 0|𝑑𝑢⟩ + 0|𝑑𝑑⟩
5 5
The values of 𝝍(𝒂, 𝒃) for the first case |𝝍𝟏 ⟩:
1
𝜓(𝑢, 𝑢) = = 𝜓(𝑢, 𝑑) = 𝜓(𝑑, 𝑢) = 𝜓(𝑑, 𝑑)
2
𝜌𝑎′𝑎 = ∑ 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎′, 𝑏)

𝑏
1 1 1 1 1 1 2 1
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑢𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
The density matrix 𝜌 of Alice goes like this:
1 1
𝜌 = (2 2)
1 1
2 2
Check: a density matrix must be Hermitian, and the trace of a density matrix must be 1, both
conditions are fulfilled for this result, Alice’s matrix represents a mixed state.
page 82 of 105
Bob’s density matrix: calculate the sum 𝜌𝑏′𝑏 = ∑𝑎 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎, 𝑏′)
The values of 𝜓(𝑎, 𝑏) for the first case |𝜓1 ⟩:

1
𝜓(𝑢, 𝑢) = = 𝜓(𝑢, 𝑑) = 𝜓(𝑑, 𝑢) = 𝜓(𝑑, 𝑑)
2
𝜌𝑏′𝑏 = ∑ 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎, 𝑏′)

𝑎
1 1 1 1 1 1 2 1
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑢) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
1 1 1 1 1 1 2 1
𝜌𝑑𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = ∙ + ∙ = + = =
2 2 2 2 4 4 4 2
The density matrix 𝜌 of Bob goes like this:
1 1
𝜌 = (2 2)
1 1
2 2
conditions are fulfilled for this result, Bob’s matrix represents a mixed state.
As all values are equal the swapping of the indices u, d has no effect to the density matrix, so Alice’s
and Bob’s matrices are equal.
The values of 𝝍(𝒂, 𝒃) for the second case |𝝍𝟐 ⟩:

1
|𝜓2 ⟩ = (|𝑢𝑢⟩ + |𝑑𝑑⟩) + 0(|𝑢𝑑⟩ + |𝑑𝑢⟩)
√2
We calculate Alice’s density matrix:
1 1 1 1
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = ∙ +0= =
√2 √2 √4 2
𝜌𝑢𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = 0 + 0 = 0
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = 0 + 0 = 0
1 1 1 1
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = 0 + ∙ = =
√2 √2 √4 2
1
0
𝜌 = (2 )
1
0
2
page 83 of 105
conditions are fulfilled for this result, Alice’s matrix represents a mixed state.
We calculate Bob’s density matrix:

1 1 1 1
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) = ∙ +0= =
√2 √2 √4 2
𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑢) = 0 + 0 = 0
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑑) = 0 + 0 = 0
1 1 1 1
𝜌𝑑𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = 0 + ∙ = =
√2 √2 √4 2
1
0
𝜌 = (2 )
1
0
2
conditions are fulfilled for this result, Bob’s matrix represents a mixed state.
As the values of the mixed kets |𝑢𝑑⟩ and |𝑑𝑢⟩ are 0, the matrices of Bob and Alice are the same.
The values of 𝝍(𝒂, 𝒃) for the third case |𝝍𝟑 ⟩:

1
|𝜓3 ⟩ = (3|𝑢𝑢⟩ + 4|𝑢𝑑⟩) + 0(|𝑑𝑢⟩ + |𝑑𝑑⟩)
5
We calculate Alice’s density matrix:
3 3 4 4 9 16 25
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = ∙ + ∙ = + = =1
5 5 5 5 25 25 25
3 4
𝜌𝑢𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = 0 ∙ + 0 ∙ = 0
5 5
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = 0
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = 0
1 0
𝜌=( )
0 0
conditions are fulfilled for this result, Alice’s matrix represents a pure state.
We calculate Bob’s density matrix:

3 3 9
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) = ∙ +0=
5 5 25
4 3 12
𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑢) = ∙ +0∙0=
5 5 25
3 4 12
𝜌𝑑𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑑) = ∙ +0∙0=
5 5 25
page 84 of 105
4 4 16
𝜌𝑑𝑑 = 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = ∙ +0=
5 5 25
9 12
𝜌=(25 25)
12 16
25 25
conditions are fulfilled for this result, Bob’s matrix doesn’t represent a pure state.
In this case both kets |𝑢𝑢⟩ ; |𝑑𝑑⟩ and |𝑢𝑑⟩ ; |𝑑𝑢⟩ change their values. This asymmetry changes the
density matrix from a pure state to a mixed state.
page 85 of 105
Exercise 7.9
Given any Alice observable 𝐴 and Bo observable 𝐵, show that for a product state, the correlation
𝐶(𝐴, 𝐵) is zero.
**********
Alice prepares her spin in state 〈𝐴〉 = 𝛼𝑢 |𝑢⟩ + 𝛼𝑑 |𝑑⟩.
Bob prepares his spin in state 〈𝐵〉 = 𝛽𝑢 |𝑢⟩ + 𝛽𝑑 |𝑑⟩.
The product of these two states 〈𝐴〉 ∙ 〈𝐵〉 gives:
(𝛼𝑢 |𝑢⟩ + 𝛼𝑑 |𝑑⟩) ∙ (𝛽𝑢 |𝑢⟩ + 𝛽𝑑 |𝑑⟩) =

(𝛼𝑢 |𝑢⟩ ∙ 𝛽𝑢 |𝑢⟩) + (𝛼𝑢 |𝑢⟩ ∙ 𝛽𝑑 |𝑑⟩) + (𝛼𝑑 |𝑑⟩ ∙ 𝛽𝑢 |𝑢⟩) + (𝛼𝑑 |𝑑⟩ ∙ 𝛽𝑑 |𝑑⟩) =
(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩) + (𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩) + (𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩) + (𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
The product state 〈𝐴𝐵〉 is the tensor product of both states, written as
𝐴⊗𝐵 =
{𝛼𝑢 |𝑢⟩ + 𝛼𝑑 |𝑑⟩} ⊗ {𝛽𝑢 |𝑢⟩ + 𝛽𝑑 |𝑑⟩} =
(𝛼𝑢 |𝑢⟩)(𝛽𝑢 |𝑢⟩ + 𝛽𝑑 |𝑑⟩) + (𝛼𝑑 |𝑑⟩)(𝛽𝑢 |𝑢⟩ + 𝛽𝑑 |𝑑⟩) =

(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩) + (𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩) + (𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩) + (𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
As the result for the “normal” multiplication and the tensor product is the same, the difference
equals zero.
page 86 of 105
Exercise 7.10
Given a measuring apparatus with the states |𝑏⟩ for initial blank state, |+1⟩ for “result of spin
measurement is up” and |−1⟩ for “result of spin measurement is down”.
Verify that the state-vector
αu |u, b⟩ + αd |d, b⟩
represents a completely unentangled state.
**********
We have two states, the state of the spin and the state of the apparatus:
(αu |u⟩ + αd |d⟩) and (1|b⟩ + 0|(+1)⟩ + 0|(−1⟩)), because in the beginning the apparatus is
certainly in the |𝑏⟩-state.
We build the product state according to p. 164:
|product state⟩ = {αu |u⟩ + αd |d⟩}⨂{1|b⟩ + 0|(+1)⟩ + 0|(−1⟩)} =

{αu |u⟩ + αd |d⟩}⨂{1|b⟩} =
αu |u⟩1|b⟩ + αd |d⟩1|b⟩ =
αu |u, b⟩ + αd |d, b⟩
Note: product states are not entangled.
page 87 of 105
Exercise 7.11
Calculate Alice’s density matrix for 𝜎𝑧 for the “near-singlet” state.
**********
Spin Operator Multiplication Table for the Up-Down-Basis:


The state-vector leads to the following wave-function:
𝜓𝑢𝑢 = 0|𝑢𝑢⟩ 𝜓𝑢𝑑 = √0.6|𝑢𝑑⟩ 𝜓𝑑𝑢 = −√0.4|𝑑𝑢⟩ 𝜓𝑑𝑑 = 0|𝑑𝑑⟩

As the values are all real, the complex conjugated are identical: 𝜓𝑢𝑢 = 𝜓 ∗ 𝑢𝑢 etc.
2
Obviously, the wave function is normalized: 02 + √0.6 + (−√0.4)2 + 02 = 1
The density matrix:
𝜓(𝑎, 𝑏) takes the form 𝜓(𝑎, 𝑏) = 𝜓𝑢𝑑 + 𝜓𝑑𝑢 = √0.6|𝑢𝑑⟩ − √0.4|𝑑𝑢⟩
and results in: 𝜓𝑢𝑢 = 0, 𝜓𝑢𝑑 = √0.6, 𝜓𝑑𝑢 = −√0.4, 𝜓𝑑𝑑 = 0
The density matrix of Alice: 𝜌𝑎′𝑎 = ∑𝑏 𝜓 ∗ (𝑎, 𝑏)𝜓(𝑎′ , 𝑏) (7.23)
expanded a, a’ (with 𝜓 ∗ = 𝜓 due to all coefficients being real):
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = 0.6

𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = 0
𝜌𝑑𝑢 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = 0
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = 0.4
gives Alice density matrix:
0.6 0 0.36 0
𝜌≔( ) ; 𝜌2 ≔ ( )
0 0.4 0 0.16
For Alice’s subsystem holds:
𝜌2 ≠ 𝜌 and Trace (𝜌2 ) < 1
page 88 of 105
Exercise 7.12
Verify the numerical values in each rap sheet.
This is a very specific exercise that refers to so called “rap sheets”. The solution here is for reasons of
completeness.
**********
State-Vector Rap Sheet 1


𝜓𝑢𝑢 = 𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ 𝜓𝑢𝑑 = 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ 𝜓𝑑𝑢 = 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ 𝜓𝑑𝑑 = 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩

with the normalization
The expectation values:
〈𝝈𝒛 〉 = ⟨𝝍|𝝈𝒛 |𝝍⟩ =

First the part 𝜎𝑧 |𝜓⟩:
𝜎𝑧 |𝜓⟩ = 𝜎𝑧 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =

(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
The left part ⟨𝜓| added: ⟨𝜓| = ⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
As only products of the form ⟨𝑢𝑢|𝑢𝑢⟩ contribute, the scalar product results in:
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 − 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 − 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 − 𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 − 𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 =

𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) − 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
(𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
(𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )(1) =
(𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )
page 89 of 105
〈𝝈𝒚 〉 = ⟨𝝍|𝝈𝒚 |𝝍⟩ =
First the part 𝜎𝑦 |𝜓⟩:
𝜎𝑦 |𝜓⟩ = 𝜎𝑦 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =
(𝑖𝛼𝑢 𝛽𝑢 |𝑑𝑢⟩ + 𝑖𝛼𝑢 𝛽𝑑 |𝑑𝑑⟩ − 𝑖𝛼𝑑 𝛽𝑢 |𝑢𝑢⟩ − 𝑖𝛼𝑑 𝛽𝑑 |𝑢𝑑)
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝑖𝛼𝑢 𝛽𝑢 |𝑑𝑢⟩ + 𝑖𝛼𝑢 𝛽𝑑 |𝑑𝑑⟩ − 𝑖𝛼𝑑 𝛽𝑢 |𝑢𝑢⟩ − 𝑖𝛼𝑑 𝛽𝑑 |𝑢𝑑)
−𝑖𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 − 𝑖𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 + 𝑖𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝑖𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 =
−𝑖𝛼𝑢∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 − 𝑖𝛼𝑢∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 + 𝑖𝛼𝑑∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 + 𝑖𝛼𝑑∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 =
𝑖𝛼𝑑∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 − 𝑖𝛼𝑢∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 + 𝑖𝛼𝑑∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 − 𝑖𝛼𝑢∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 =
𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )𝛽𝑢∗ 𝛽𝑢 + 𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )𝛽𝑑∗ 𝛽𝑑 =

𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )(1) =
𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )
〈𝝈𝒙 〉 = ⟨𝝍|𝝈𝒙 |𝝍⟩ =
First the part 𝜎𝑥 |𝜓⟩:
𝜎𝑥 |𝜓⟩ = 𝜎𝑥 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =

(𝛼𝑢 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑑𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑢𝑑⟩)
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝛼𝑢 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑑𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑢𝑑⟩)
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 + 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 =

𝛼𝑢∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑢∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 + 𝛼𝑑∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑑∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) + 𝛼𝑑∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =

(𝛼𝑢∗ 𝛼𝑑 + 𝛼𝑑∗ 𝛼𝑢 )(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 ) =
(𝛼𝑢∗ 𝛼𝑑 + 𝛼𝑑∗ 𝛼𝑢 )(1) =
(𝛼𝑢∗ 𝛼𝑑 + 𝛼𝑑∗ 𝛼𝑢 )
Now we must calculate
〈𝜎𝑥 〉2 + 〈𝜎𝑦 〉2 + 〈𝜎𝑧 〉2 =

2
(𝛼𝑢∗ 𝛼𝑑 + 𝛼𝑑∗ 𝛼𝑢 )2 + (𝑖(𝛼𝑑∗ 𝛼𝑢 − 𝛼𝑢∗ 𝛼𝑑 )) + (𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )2 =
page 90 of 105
(𝛼𝑢∗ 𝛼𝑑 )² + 2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 + (𝛼𝑑∗ 𝛼𝑢 )2 − (𝛼𝑑∗ 𝛼𝑢 )² + 2𝛼𝑑∗ 𝛼𝑢 𝛼𝑢∗ 𝛼𝑑 − (𝛼𝑢∗ 𝛼𝑑 )2

+ (𝛼𝑢∗ 𝛼𝑢 )² − 2𝛼𝑢∗ 𝛼𝑢 𝛼𝑑∗ 𝛼𝑑 + (𝛼𝑑∗ 𝛼𝑑 )2 =
2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 + 2𝛼𝑑∗ 𝛼𝑢 𝛼𝑢∗ 𝛼𝑑 − 2𝛼𝑢∗ 𝛼𝑢 𝛼𝑑∗ 𝛼𝑑 + (𝛼𝑢∗ 𝛼𝑢 )2 + (𝛼𝑑∗ 𝛼𝑑 )² =
2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 + 2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 − 2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 + (𝛼𝑢∗ 𝛼𝑢 )2 + (𝛼𝑑∗ 𝛼𝑑 )² =

(𝛼𝑢∗ 𝛼𝑢 )2 + 2𝛼𝑢∗ 𝛼𝑑 𝛼𝑑∗ 𝛼𝑢 + (𝛼𝑑∗ 𝛼𝑑 )² =
(𝛼𝑢∗ 𝛼𝑢 )2 + 2𝛼𝑢∗ 𝛼𝑢 𝛼𝑑∗ 𝛼𝑑 + (𝛼𝑑∗ 𝛼𝑑 )² =
(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )2 = 1
The expectation values: 〈𝝉𝒛 〉 = ⟨𝝍|𝝉𝒛 |𝝍⟩ =
First the part 𝜏𝑧 |𝜓⟩:
𝜏𝑧 |𝜓⟩ = 𝜏𝑧 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =

(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ − 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ − 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 − 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 + 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 − 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 =

𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 − 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 − 𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) + 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) =

(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) =
(1)(𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) =
(𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )
The expectation values: 〈𝝉𝒚 〉 = ⟨𝝍|𝝉𝒚 |𝝍⟩ =
First the part 𝜏𝑦 |𝜓⟩:
𝜏𝑦 |𝜓⟩ = 𝜏𝑦 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =
(𝑖𝛼𝑢 𝛽𝑢 |𝑢𝑑⟩ − 𝑖𝛼𝑢 𝛽𝑑 |𝑢𝑢⟩ + 𝑖𝛼𝑑 𝛽𝑢 |𝑑𝑑⟩ − 𝑖𝛼𝑑 𝛽𝑑 |𝑑𝑢)

(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝑖𝛼𝑢 𝛽𝑢 |𝑢𝑑⟩ − 𝑖𝛼𝑢 𝛽𝑑 |𝑢𝑢⟩ + 𝑖𝛼𝑑 𝛽𝑢 |𝑑𝑑⟩ − 𝑖𝛼𝑑 𝛽𝑑 |𝑑𝑢)
page 91 of 105
−𝑖𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑑 + 𝑖𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑢 − 𝑖𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑑 + 𝑖𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑢 =
−𝑖𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑑 + 𝑖𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑢 − 𝑖𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑑 + 𝑖𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑢 =
−𝑖𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑑 − 𝑖𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑑 + 𝑖𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑢 + 𝑖𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑢 =

−𝑖𝛽𝑢∗ 𝛽𝑑 (𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 ) + 𝑖𝛽𝑑∗ 𝛽𝑢 (𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 ) =
(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )(−𝑖𝛽𝑢∗ 𝛽𝑑 + 𝑖𝛽𝑑∗ 𝛽𝑢 ) =
𝑖(𝛽𝑑∗ 𝛽𝑢 − 𝛽𝑢∗ 𝛽𝑑 )
The expectation values: 〈𝝉𝒙 〉 = ⟨𝝍|𝝉𝒙 |𝝍⟩ =
First the part 𝜎𝑥 |𝜓⟩:
𝜏𝑥 |𝜓⟩ = 𝜏𝑥 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =

(𝛼𝑢 𝛽𝑢 |𝑢𝑑⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑢⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑑⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑢⟩)
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝛼𝑢 𝛽𝑢 |𝑢𝑑⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑢⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑑⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑢⟩)
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑑 + 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑢 + 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑑 + 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑢 =

𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑑 + 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑢 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑑 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑢 =
𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 ) + 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 ) =
(𝛼𝑢∗ 𝛼𝑢 + 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 ) =
(1)(𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 ) =
(𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 )
Now we must calculate
〈𝜏𝑥 〉2 + 〈𝜏𝑦 〉2 + 〈𝜏𝑧 〉2 =
(𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 )2 + (𝑖(𝛽𝑑∗ 𝛽𝑢 − 𝛽𝑢∗ 𝛽𝑑 ))2 + (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )2 =
(𝛽𝑢∗ 𝛽𝑑 + 𝛽𝑑∗ 𝛽𝑢 )2 − (𝛽𝑑∗ 𝛽𝑢 − 𝛽𝑢∗ 𝛽𝑑 )2 + (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )2 =
(𝛽𝑢∗ 𝛽𝑑 )² + 2𝛽𝑢∗ 𝛽𝑑 𝛽𝑑∗ 𝛽𝑢 + (𝛽𝑑∗ 𝛽𝑢 )2 − ((𝛽𝑑∗ 𝛽𝑢 )² − 2𝛽𝑑∗ 𝛽𝑢 𝛽𝑢∗ 𝛽𝑑 + (𝛽𝑢∗ 𝛽𝑑 )²)

+ (𝛽𝑢∗ 𝛽𝑢 )² − 2𝛽𝑢∗ 𝛽𝑢 𝛽𝑑∗ 𝛽𝑑 + (𝛽𝑑∗ 𝛽𝑑 )2 =
(𝛽𝑢∗ 𝛽𝑑 )² + 2𝛽𝑢∗ 𝛽𝑑 𝛽𝑑∗ 𝛽𝑢 + (𝛽𝑑∗ 𝛽𝑢 )2 − (𝛽𝑑∗ 𝛽𝑢 )2 + 2𝛽𝑑∗ 𝛽𝑢 𝛽𝑢∗ 𝛽𝑑 − (𝛽𝑢∗ 𝛽𝑑 )²)
+ (𝛽𝑢∗ 𝛽𝑢 )² − 2𝛽𝑢∗ 𝛽𝑢 𝛽𝑑∗ 𝛽𝑑 + (𝛽𝑑∗ 𝛽𝑑 )2 =
(𝛽𝑢∗ 𝛽𝑢 )² + 2𝛽𝑢∗ 𝛽𝑢 𝛽𝑑∗ 𝛽𝑑 + (𝛽𝑑∗ 𝛽𝑑 )2 =
(𝛽𝑢∗ 𝛽𝑢 + 𝛽𝑑∗ 𝛽𝑑 )² = 1
page 92 of 105
The composite expectation value: 〈𝜏𝑧 𝜎𝑧 〉 = ⟨𝜓|𝜏𝑧 𝜎𝑧 |𝜓⟩
𝜎𝑧 |𝜓⟩ = (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)

𝜏𝑧 𝜎𝑧 |𝜓⟩ =
𝜏𝑧 (𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ − 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩) =
(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ − 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
(⟨𝑢𝑢|𝛼𝑢∗ 𝛽𝑢∗ + ⟨𝑢𝑑|𝛼𝑢∗ 𝛽𝑑∗ + ⟨𝑑𝑢|𝛼𝑑∗ 𝛽𝑢∗ + ⟨𝑑𝑑|𝛼𝑑∗ 𝛽𝑑∗ )(𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ − 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ − 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩)
𝛼𝑢∗ 𝛽𝑢∗ 𝛼𝑢 𝛽𝑢 − 𝛼𝑢∗ 𝛽𝑑∗ 𝛼𝑢 𝛽𝑑 − 𝛼𝑑∗ 𝛽𝑢∗ 𝛼𝑑 𝛽𝑢 + 𝛼𝑑∗ 𝛽𝑑∗ 𝛼𝑑 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 𝛽𝑢∗ 𝛽𝑢 − 𝛼𝑢∗ 𝛼𝑢 𝛽𝑑∗ 𝛽𝑑 − 𝛼𝑑∗ 𝛼𝑑 𝛽𝑢∗ 𝛽𝑢 + 𝛼𝑑∗ 𝛼𝑑 𝛽𝑑∗ 𝛽𝑑 =
𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) − 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )
The product 〈𝜎𝑧 〉〈𝜏𝑧 〉:
〈𝜎𝑧 〉 = (𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )
〈𝜏𝑧 〉 = (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )
〈𝜎𝑧 〉〈𝜏𝑧 〉 = (𝛼𝑢∗ 𝛼𝑢 − 𝛼𝑑∗ 𝛼𝑑 )(𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) =
𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) − 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )

〈𝜏𝑧 𝜎𝑧 〉:
〈𝜏𝑧 𝜎𝑧 〉 = 𝛼𝑢∗ 𝛼𝑢 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 ) − 𝛼𝑑∗ 𝛼𝑑 (𝛽𝑢∗ 𝛽𝑢 − 𝛽𝑑∗ 𝛽𝑑 )
The correlation:
〈𝜏𝑧 𝜎𝑧 〉 = 〈𝜏𝑧 〉〈𝜎𝑧 〉 → 〈𝜏𝑧 𝜎𝑧 〉 − 〈𝜏𝑧 〉〈𝜎𝑧 〉 = 0
page 93 of 105
State-Vector Rap Sheet 2



1 1
𝜓𝑢𝑢 = 0|𝑢𝑢⟩ 𝜓𝑢𝑑 = |𝑢𝑑⟩ 𝜓𝑑𝑢 = − |𝑑𝑢⟩ 𝜓𝑑𝑑 = 0|𝑑𝑑⟩
√2 √2
1 2 1 2
Obviously, the wave function is normalized: 02 + ( 2) + (− ) + 02 = 1
√ √2
The density matrix:

1 1
𝜓(𝑎, 𝑏) takes the form 𝜓(𝑎, 𝑏) = 𝜓𝑢𝑑 + 𝜓𝑑𝑢 = |𝑢𝑑⟩ − |𝑑𝑢⟩
√2 √2
1 1
and results in: 𝜓𝑢𝑢 = 0, 𝜓𝑢𝑑 = , 𝜓𝑑𝑢 = − , 𝜓𝑑𝑑 = 0
√2 √2
expanding a, a’ to uu, ud, du and dd (with 𝜓 ∗ = 𝜓 due to all coefficients being real):
1 1 1
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = 0 ∙ 0 + ∙ =
√2 √2 2
−1 1
𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = 0 ∙ + ∙0=0
√2 √2
−1 1
𝜌𝑑𝑢 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = ∙0+0∙ =0
√2 √2
−1 −1 1
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = ∙ +0∙0=
√2 √2 2
1 1
0 0
𝜌 ≔ (2 ) ; 𝜌2 ≔ (4 )
1 1
0 0
2 4
page 94 of 105
〈𝜎𝑧 〉 = ⟨𝜓|𝜎𝑧 |𝜓⟩ =

1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜎𝑧 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑢𝑑⟩ + | 𝑑𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
⟨𝑢𝑑|𝑢𝑑⟩ + ⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩ =
2 2 2 2
1 1
+0+0− =0
2 2
〈𝜎𝑥 〉 = ⟨𝜓|𝜎𝑥 |𝜓⟩ =
1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜎𝑥 | (( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑑𝑑⟩ − | 𝑢𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
⟨𝑢𝑑|𝑑𝑑⟩ − ⟨𝑢𝑑|𝑢𝑢⟩ − ⟨𝑑𝑢|𝑑𝑑⟩ + ⟨𝑑𝑢|𝑢𝑢⟩ =
2 2 2 2
0−0−0+0=0
〈𝜎𝑦 〉 = ⟨𝜓|𝜎𝑦 |𝜓⟩ =
1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜎𝑦 | (( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) (𝑖 |𝑑𝑑⟩ + 𝑖 | 𝑢𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
𝑖 ⟨𝑢𝑑|𝑑𝑑⟩ + 𝑖 ⟨𝑢𝑑|𝑢𝑢⟩ − 𝑖 ⟨𝑑𝑢|𝑑𝑑⟩ − 𝑖 ⟨𝑑𝑢|𝑢𝑢⟩ =
2 2 2 2
0+0−0−0=0
〈𝜏𝑧 〉 = ⟨𝜓|𝜏𝑧 |𝜓⟩ =

1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑧 | (( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) (− |𝑢𝑑⟩ − | 𝑑𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
− ⟨𝑢𝑑|𝑢𝑑⟩ − ⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩ + ⟨𝑑𝑢|𝑑𝑢⟩ =
2 2 2 2
1 1
− +0+0+ = 0
2 2
page 95 of 105
〈𝜏𝑥 〉 = ⟨𝜓|𝜏𝑥 |𝜓⟩ =

1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑥 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑢𝑢⟩ − | 𝑑𝑑⟩) =
√2 √2 √2 √2
1 1 1 1
⟨𝑢𝑑|𝑢𝑢⟩ − ⟨𝑢𝑑|𝑑𝑑⟩ − ⟨𝑑𝑢|𝑢𝑢⟩ + ⟨𝑑𝑢|𝑑𝑑⟩ =
2 2 2 2
0−0−0+0=0
〈𝜏𝑦 〉 = ⟨𝜓|𝜏𝑦 |𝜓⟩ =
1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑦 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) (−𝑖 |𝑢𝑢⟩ − 𝑖 | 𝑑𝑑⟩) =
√2 √2 √2 √2
1 1 1 1
−𝑖 ⟨𝑢𝑑|𝑢𝑢⟩ − 𝑖 ⟨𝑢𝑑|𝑑𝑑⟩ + 𝑖 ⟨𝑑𝑢|𝑢𝑢⟩ + 𝑖 ⟨𝑑𝑢|𝑑𝑑⟩ =
2 2 2 2
−0 − 0 + 0 + 0 = 0
The composite expectation values:
〈𝜏𝑧 𝜎𝑧 〉 = ⟨𝜓|𝜏𝑧 𝜎𝑧 |𝜓⟩ =

1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑧 𝜎𝑧 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑧 | ( |𝑢𝑑⟩ + | 𝑑𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) (− |𝑢𝑑⟩ + | 𝑑𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
− ⟨𝑢𝑑|𝑢𝑑⟩ + ⟨𝑢𝑑|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩ =
2 2 2 2
1 1
− + 0 + 0 − = −1
2 2
page 96 of 105
〈𝜏𝑦 𝜎𝑦 〉 = ⟨𝜓|𝜏𝑦 𝜎𝑦 |𝜓⟩ =
1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑦 𝜎𝑦 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 𝑖 𝑖
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑦 | ( |𝑑𝑑⟩ + | 𝑢𝑢⟩) =
√2 √2 √2 √2
1 1 −𝑖² 𝑖²
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑑𝑢⟩ − | 𝑢𝑑⟩) =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑑𝑢⟩ + | 𝑢𝑑⟩) =
√2 √2 √2 √2
1 1 1 1
⟨𝑢𝑑|𝑑𝑢⟩ − ⟨𝑢𝑑|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩ − ⟨𝑑𝑢|𝑢𝑑⟩ =
2 2 2 2
1 1
0 − − + 0 = −1
2 2
〈𝜏𝑥 𝜎𝑥 〉 = ⟨𝜓|𝜏𝑥 𝜎𝑥 |𝜓⟩ =

1 1 1 1
⟨(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑥 𝜎𝑥 | ( |𝑢𝑑⟩ − | 𝑑𝑢⟩)⟩ =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| ) |𝜏𝑥 | ( |𝑑𝑑⟩ − | 𝑢𝑢⟩) =
√2 √2 √2 √2
1 1 1 1
(⟨𝑢𝑑| − ⟨𝑑𝑢| )( |𝑑𝑢⟩ − | 𝑢𝑑⟩) =
√2 √2 √2 √2
1 1 1 1
⟨𝑢𝑑|𝑑𝑢⟩ − ⟨𝑢𝑑|𝑢𝑑⟩ − ⟨𝑑𝑢|𝑑𝑢⟩ + ⟨𝑑𝑢|𝑢𝑑⟩ =
2 2 2 2
1 1
0 − − + 0 = −1
2 2
The correlation:
〈𝜏𝑧 𝜎𝑧 〉 − 〈𝜏𝑧 〉〈𝜎𝑧 〉 = −1 − 0 = −1
page 97 of 105
Vector Rap Sheet 3


𝜓𝑢𝑢 = 0|𝑢𝑢⟩ 𝜓𝑢𝑑 = √0.6|𝑢𝑑⟩ 𝜓𝑑𝑢 = −√0.4|𝑑𝑢⟩ 𝜓𝑑𝑑 = 0|𝑑𝑑⟩

2
Obviously, the wave function is normalized: 02 + √0.6 + (−√0.4)2 + 02 = 1
The density matrix:
𝜓(𝑎, 𝑏) takes the form 𝜓(𝑎, 𝑏) = 𝜓𝑢𝑑 + 𝜓𝑑𝑢 = √0.6|𝑢𝑑⟩ − √0.4|𝑑𝑢⟩
and results in: 𝜓𝑢𝑢 = 0, 𝜓𝑢𝑑 = √0.6, 𝜓𝑑𝑢 = −√0.4, 𝜓𝑑𝑑 = 0
expanded a, a’ (with 𝜓 ∗ = 𝜓 due to all coefficients being real):
𝜌𝑢𝑢 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑢, 𝑑) = 0.6

𝜌𝑢𝑑 = 𝜓 ∗ (𝑢, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑢, 𝑑)𝜓(𝑑, 𝑑) = 0
𝜌𝑑𝑢 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑢, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑢, 𝑑) = 0
𝜌𝑑𝑑 = 𝜓 ∗ (𝑑, 𝑢)𝜓(𝑑, 𝑢) + 𝜓 ∗ (𝑑, 𝑑)𝜓(𝑑, 𝑑) = 0.4
0.6 0 0.36 0
𝜌≔( ) ; 𝜌2 ≔ ( )
0 0.4 0 0.16
〈𝜎𝑧 〉 = ⟨𝜓|𝜎𝑧 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜎𝑧 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(√0,6|𝑢𝑑⟩ + √0,4|𝑑𝑢⟩) =
0.6⟨𝑢𝑑|𝑢𝑑⟩ + √0.24⟨𝑢𝑑|𝑑𝑢⟩ + √0.24⟨𝑑𝑢|𝑢𝑑⟩ − 0.4⟨𝑑𝑢|𝑑𝑢⟩ =

0.6 + 0 + 0 − 0.4 = 0.2
page 98 of 105
〈𝜎𝑥 〉 = ⟨𝜓|𝜎𝑥 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜎𝑥 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(√0,6|𝑑𝑑⟩ − √0,4|𝑢𝑢⟩) =
0.6⟨𝑢𝑑|𝑑𝑑⟩ − √0.24⟨𝑢𝑑|𝑢𝑢⟩ − √0.24⟨𝑑𝑢|𝑑𝑑⟩ + 0.4⟨𝑑𝑢|𝑢𝑢⟩ =

0−0−0+0=0
〈𝜎𝑦 〉 = ⟨𝜓|𝜎𝑦 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜎𝑦 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(𝑖√0,6|𝑑𝑑⟩ + 𝑖√0,4|𝑢𝑢⟩) =
𝑖0.6⟨𝑢𝑑|𝑑𝑑⟩ + 𝑖√0.24⟨𝑢𝑑|𝑢𝑢⟩ − 𝑖√0.24⟨𝑑𝑢|𝑑𝑑⟩ − 𝑖0.4⟨𝑑𝑢|𝑢𝑢⟩ =

0+0−0−0=0
〈𝜏𝑧 〉 = ⟨𝜓|𝜏𝑧 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑧 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(−√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩) =
−0.6⟨𝑢𝑑|𝑢𝑑⟩ − √0.24⟨𝑢𝑑|𝑑𝑢⟩ + √0.24⟨𝑑𝑢|𝑢𝑑⟩ + 0.4⟨𝑑𝑢|𝑑𝑢⟩ =

−0.6 + 0 + 0 + 0.4 = −0.2
〈𝜏𝑥 〉 = ⟨𝜓|𝜏𝑥 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑥 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(√0,6|𝑢𝑢⟩ − √0,4|𝑑𝑑⟩) =
0.6⟨𝑢𝑑|𝑢𝑢⟩ − √0.24⟨𝑢𝑑|𝑑𝑑⟩ − √0.24⟨𝑑𝑢|𝑢𝑢⟩ + 0.4⟨𝑑𝑢|𝑑𝑑⟩ =

0−0−0+0=0
〈𝜏𝑦 〉 = ⟨𝜓|𝜏𝑦 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑦 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(−𝑖√0,6|𝑢𝑢⟩ − 𝑖√0,4|𝑑𝑑⟩) =
−𝑖0.6⟨𝑢𝑑|𝑢𝑢⟩ − 𝑖√0.24⟨𝑢𝑑|𝑑𝑑⟩ + 𝑖√0.24⟨𝑑𝑢|𝑢𝑢⟩ + 𝑖0.4⟨𝑑𝑢|𝑑𝑑⟩ =

−0 − 0 + 0 + 0 = 0
page 99 of 105
The composite expectation values:
〈𝜏𝑧 𝜎𝑧 〉 = ⟨𝜓|𝜏𝑧 𝜎𝑧 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑧 𝜎𝑧 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑧 |(√0,6|𝑢𝑑⟩ + √0,4|𝑑𝑢⟩) =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(−√0,6|𝑢𝑑⟩ + √0,4|𝑑𝑢⟩) =
−0.6⟨𝑢𝑑|𝑢𝑑⟩ + √0.24⟨𝑢𝑑|𝑑𝑢⟩ + √0.24⟨𝑑𝑢|𝑢𝑑⟩ − 0.4⟨𝑑𝑢|𝑑𝑢⟩ =
−0.6 + 0 + 0 − 0.4 = −1
〈𝜏𝑥 𝜎𝑥 〉 = ⟨𝜓|𝜏𝑥 𝜎𝑥 |𝜓⟩ =
⟨(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑥 𝜎𝑥 |(√0,6|𝑢𝑑⟩ − √0,4|𝑑𝑢⟩)⟩ =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)|𝜏𝑥 |(√0,6|𝑑𝑑⟩ − √0,4|𝑢𝑢⟩) =
(⟨𝑢𝑑|√0,6 − ⟨𝑑𝑢|√0,4)(√0,6|𝑑𝑢⟩ − √0,4|𝑢𝑑⟩) =
0.6⟨𝑢𝑑|𝑑𝑢⟩ − √0.24⟨𝑢𝑑|𝑢𝑑⟩ − √0.24⟨𝑑𝑢|𝑑𝑢⟩ + 0.4⟨𝑑𝑢|𝑢𝑑⟩ =
0 − √0.24 − √0.24 + 0 = −2 ∙ √0.24 = −√0.96

The correlation:
〈𝜏𝑧 𝜎𝑧 〉 − 〈𝜏𝑧 〉〈𝜎𝑧 〉 = −1 − 0.2 ∙ (−0.2) = −1 + 0.04 = −0.96
page 100 of 105

Exercise 8.1
Prove that the position operator 𝑋 and the momentum operator 𝐷 are linear operators.
𝑿: 𝑿𝜑(𝑥) = 𝑥𝜑(𝑥)
𝑑𝜑(𝑥)
𝑫: 𝑫𝜑(𝑥) =
𝑑𝑥
**********
Properties of a linear operator 𝑳:
1) 𝑳(𝑓 + 𝑔) = 𝑳𝑓 + 𝑳𝑔
2) 𝑳(𝑡𝑓) = 𝑡𝑳𝑓
For 𝑿
1)
𝑿(𝜑(𝑥) + 𝜃(𝑥)) = 𝑥(𝜑(𝑥) + 𝜃(𝑥)) =

𝑥𝜑(𝑥) + 𝑥𝜃(𝑥) = 𝑿𝜑(𝑥) + 𝑿𝜃(𝑥)
2)
𝑿(𝑧𝜑(𝑥)) = 𝑥𝑧𝜑(𝑥) = 𝑧𝑥𝜑(𝑥) = 𝑧𝑿(𝜑(𝑥))
For 𝑫
1)
𝑑(𝜑(𝑥) + 𝜃(𝑥))
𝑫(𝜑(𝑥) + 𝜃(𝑥)) = =
𝑑𝑥
𝑑𝜑(𝑥) 𝑑𝜃(𝑥)
+ = 𝑫𝜑(𝑥) + 𝜃𝑫(𝑥)
𝑑𝑥 𝑑𝑥
2)
𝑑(𝑧𝜑(𝑥)) 𝑧(𝑑𝜑(𝑥))
𝑫(𝑧𝜑(𝑥)) = = = 𝑧𝑫(𝜑(𝑥))
𝑑𝑥 𝑑𝑥
page 101 of 105

Exercise 9.1
Applicate the Hamiltonian
ℏ2 𝜕 2 𝜓(𝑥)
− = 𝐸𝜓(𝑥)
2𝑚 𝜕𝑥 2
to the wave function:
𝑖𝑝
𝑥
𝜓(𝑥) = 𝑒 ℏ
Show that this wave function is a solution, if we set:
𝑝2
𝐸=
2𝑚
**********
We start with the Hamiltonian:
ℏ2 𝜕 2 𝜓(𝑥)
− = 𝐸𝜓(𝑥)
2𝑚 𝜕𝑥 2
We insert the wave function:
𝑖𝑝𝑥
ℏ2 𝜕 2 𝑒 ℏ
− =;
2𝑚 𝜕𝑥 2
We derive partial twice:
Once:
𝑖𝑝𝑥
𝜕𝑒 ℏ 𝑖𝑝 𝑖𝑝𝑥
= 𝑒 ℏ
𝜕𝑥 ℏ
Twice:
𝑖𝑝 𝑖𝑝𝑥
𝜕( 𝑒 ℏ ) 𝑖𝑝 𝑖𝑝 𝑖𝑝𝑥 𝑝2 𝑖𝑝𝑥
ℏ
= 𝑒 ℏ = − 2𝑒 ℏ
𝜕𝑥 ℏ ℏ ℏ
… fill in the result:
𝑖𝑝𝑥
ℏ2 𝜕 2 𝑒 ℏ ℏ2 𝑝2 𝑖𝑝𝑥
− = (− ) (− 𝑒 ℏ )=
2𝑚 𝜕𝑥 2 2𝑚 ℏ2
𝑝2 𝑖𝑝𝑥 𝑝2
( 𝑒 )=
ℏ 𝜓(𝑥)
2𝑚 2𝑚
We get:
ℏ2 𝜕 2 𝜓(𝑥) 𝑝2
− = 𝜓(𝑥): = 𝐸𝜓(𝑥)
2𝑚 𝜕𝑥 2 2𝑚
We can conclude:
𝑝2
=𝐸
2𝑚
page 102 of 105

Exercise 9.2
Prove the following equation by expanding each side and comparing the results:
[𝑃2 , 𝑋] = 𝑃[𝑃, 𝑋] + [𝑃, 𝑋]𝑃
𝑃 is the momentum operator, 𝑋 is the position operator – both are matrices.

[𝑃, 𝑋] is the commutator relation: [𝑃, 𝑋] = 𝑃𝑋 − 𝑋𝑃
**********
[𝑃2 , 𝑋] = 𝑃[𝑃, 𝑋] + [𝑃, 𝑋]𝑃
Left side:
[𝑃2 , 𝑋] = 𝑃𝑃𝑋 − 𝑋𝑃𝑃
Right side:
𝑃[𝑃, 𝑋] + [𝑃, 𝑋]𝑃 =

𝑃(𝑃𝑋 − 𝑋𝑃) + (𝑃𝑋 − 𝑋𝑃)𝑃 =
𝑃𝑃𝑋 − 𝑃𝑋𝑃 + 𝑃𝑋𝑃 − 𝑋𝑃𝑃 =
𝑃𝑃𝑋 − 𝑋𝑃𝑃
Obviously both sides of the equation are the same.
page 103 of 105

Exercise 9.3
Show that the right-hand side of

𝑑 𝑑
[𝑉(𝑥), 𝑃]𝜓(𝑥) = 𝑉(𝑥) (−𝑖ℏ ) 𝜓(𝑥) − (−𝑖ℏ ) 𝑉(𝑥)𝜓(𝑥)
𝑑𝑥 𝑑𝑥
simplifies to the right-hand side of:
𝑑𝑉(𝑥)
[𝑉(𝑥), 𝑃]𝜓(𝑥) = 𝑖ℏ
𝑑𝑥
Hint: First expand the second term by taking the derivative of the product.
Then look for cancellations.
**********
𝑑 𝑑
[𝑉(𝑥), 𝑃]𝜑(𝑥) = 𝑉(𝑥) (−𝑖ℏ ) 𝜑(𝑥) − (−𝑖ℏ ) 𝑉(𝑥)𝜑(𝑥)
𝑑𝑥 𝑑𝑥
We calculate the derivations:
𝑑 𝑑
𝑉(𝑥) (−𝑖ℏ ) 𝜑(𝑥) − (−𝑖ℏ ) 𝑉(𝑥)𝜑(𝑥) =
𝑑𝑥 𝑑𝑥
𝑑𝜑(𝑥) 𝑑𝑉(𝑥) 𝑑𝜑(𝑥)
𝑉(𝑥) (−𝑖ℏ ) + 𝑖ℏ ( ∙ 𝜑(𝑥) + ∙ 𝑉(𝑥)) =
𝑑𝑥 𝑑𝑥 𝑑𝑥
𝑑𝜑(𝑥) 𝑑𝑉(𝑥) 𝑑𝜑(𝑥)

−𝑖ℏ ∙ 𝑉(𝑥) + 𝑖ℏ ∙ 𝜑(𝑥) + 𝑖ℏ ∙ 𝑉(𝑥) =
𝑑𝑥 𝑑𝑥 𝑑𝑥
𝑑𝑉(𝑥)
𝑖ℏ ∙ 𝜑(𝑥)
𝑑𝑥
We get
𝑑𝑉(𝑥)
[𝑉(𝑥), 𝑃]𝜑(𝑥) = 𝑖ℏ ∙ 𝜑(𝑥)
𝑑𝑥
or
𝑑𝑉(𝑥)
[𝑉(𝑥), 𝑃] = 𝑖ℏ
𝑑𝑥
page 104 of 105

Exercise 10.1
Find the second derivative of 𝑥:
Show thereby that it solves:
−𝜔2 𝑥 = 𝑥̈
**********
We calculate:
𝑥̇ = −𝐴 ∙ 𝑠𝑖𝑛(𝜔𝑡) ∙ 𝜔 + 𝐵 ∙ 𝑐𝑜𝑠(𝜔𝑡) ∙ 𝜔 =
𝜔(𝐵 ∙ 𝑐𝑜𝑠(𝜔𝑡) − 𝐴 ∙ 𝑠𝑖𝑛(𝜔𝑡))
𝑥̈̇ = 𝜔(−𝐴 ∙ 𝑐𝑜𝑠(𝜔𝑡) ∙ 𝜔 − 𝐵 ∙ 𝑠𝑖𝑛(𝜔𝑡) ∙ 𝜔) =
−𝜔2 (𝐴 ∙ 𝑐𝑜𝑠(𝜔𝑡) + 𝐵 ∙ 𝑠𝑖𝑛(𝜔𝑡)) =
−𝜔2 𝑥
The same calculation with complex numbers:
𝑥 = 𝑒 𝑖𝜔𝑡
𝑥̇ = 𝑖𝜔𝑒 𝑖𝜔𝑡
𝑥̈ = 𝑖 2 𝜔2 𝑖𝑒 𝑖𝜔𝑡 = −𝜔2 𝑒 𝑖𝜔𝑡 = −𝜔2 𝑥
page 105 of 105

Quantum Mechanics The Theoretical Minimum Exercises Complete1

Uploaded by

Copyright:

Available Formats

Quantum Mechanics The Theoretical Minimum Exercises Complete1

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Quantum Mechanics The Theoretical Minimum Exercises Complete1

Uploaded by

Copyright:

Available Formats

What are inner products and how are they used?

What are inner products and how are they used?

What are orthonormal bases and how can they be constructed?

What are orthonormal bases and how can they be constructed?

The exercises of „Quantum Mechanics, The Theoretical Minimum“

Hope I can help you with learning quantum mechanics.

c) Show that 𝛼 ∗ 𝛽 must be pure imaginary.

Hermitian: the diagonal is pure real and: 𝑀 = 𝑀†

The square of uncertainty (or standard deviation) of A, (△ 𝐴)2 :

then the state-vector is automatically normalized as well:

𝛼𝑢 𝛽𝑢 |𝑢𝑢⟩ + 𝛼𝑢 𝛽𝑑 |𝑢𝑑⟩ + 𝛼𝑑 𝛽𝑢 |𝑑𝑢⟩ + 𝛼𝑑 𝛽𝑑 |𝑑𝑑⟩

𝜎𝑧 |𝑢⟩ = |𝑢⟩ and 𝜎𝑧 |𝑑⟩ = −|𝑑⟩

a) Rewrite the equation:

𝐴11 𝐵11 𝐴11 𝐵12 𝐴12 𝐵11 𝐴12 𝐵12

c) Expand all three Kronecker products.

d) Verify the row and column sizes of each Kronecker product:

c) If 𝜌 is a density matrix, does it represent a pure state or a mixed state?

Use 𝑃(𝑎) = 𝜌𝑎𝑎 to show that if 𝜌 is a density matrix, then:

𝑎1∗ 𝑎1 + 𝑎2∗ 𝑎2 + ⋯ + 𝑎𝑛∗ 𝑎𝑛 =

Interchange: ⟨𝐵|𝐴⟩ = ⟨𝐴|𝐵⟩∗

𝑎1∗∗ 𝑏1∗ + 𝑎2∗∗ 𝑏2∗ + ⋯ + 𝑎𝑛∗∗ 𝑏𝑛∗ =

Prove that the vector |𝑟⟩ is orthogonal to vector |𝑙⟩:

If |𝑟⟩ is orthogonal to |𝑙⟩, then the scalar product ⟨𝑟|𝑙⟩ must be 0.

Eq. 2.7: ⟨𝑖|𝑜⟩ = 0

Eqs. 2.8 ⟨𝑜|𝑢⟩⟨𝑢|𝑜⟩:

Eqs. 2.8 ⟨𝑜|𝑑⟩⟨𝑑|𝑜⟩:

Eqs. 2.8 ⟨𝑖|𝑢⟩⟨𝑢|𝑖⟩:

Eqs. 2.8 ⟨𝑖|𝑑⟩⟨𝑑|𝑖⟩:

Eqs. 2.9 ⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩:

Eqs. 2.9 ⟨𝑜|𝑟⟩⟨𝑟|𝑜⟩:

Eqs. 2.9 ⟨𝑜|𝑙⟩⟨𝑙|𝑜⟩:

Eqs. 2.9 ⟨𝑖|𝑟⟩⟨𝑟|𝑖⟩:

Eqs. 2.9 ⟨𝑖|𝑙⟩⟨𝑙|𝑖⟩:

Are |𝑖⟩ and |𝑜⟩ unique in that respect?

Assume that |𝑖⟩ and |𝑜⟩ are given as:

|𝑖⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩ and ⟨𝑖| = ⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗

|𝑜⟩ = 𝛾|𝑢⟩ − 𝛿|𝑑⟩ and ⟨𝑜| = ⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗

a) Use Eqs. 2.8 to show that

⟨𝑢|𝑜⟩ = ⟨𝑢|(𝛾|𝑢⟩ − 𝛿|𝑑⟩) =

⟨𝑜|𝑑⟩ = (⟨𝑢|𝛾 ∗ − ⟨𝑑|𝛿 ∗ )|𝑑⟩ =

⟨𝑑|𝑜⟩ = ⟨𝑑|(𝛾|𝑢⟩ − 𝛿|𝑑⟩) =

⟨𝑢|𝑖⟩ = ⟨𝑢|(𝛼|𝑢⟩ + 𝛽|𝑑⟩) =

⟨𝑑|𝑖⟩ = ⟨𝑑|(𝛼|𝑢⟩ + 𝛽|𝑑⟩) =

b) Use the above result and Eqs. 2.9 to show that

|𝑖⟩ = 𝛼|𝑢⟩ + 𝛽|𝑑⟩ and ⟨𝑖| = ⟨𝑢|𝛼 ∗ + ⟨𝑑|𝛽 ∗

c) Show that 𝛼 ∗ 𝛽 must be pure imaginary.

Let 𝛼 = 𝑟 + 𝑖𝑠, 𝛼 ∗ = 𝑟 − 𝑖𝑠, 𝛽 = 𝑡 + 𝑖𝑢, 𝛽 ∗ = 𝑡 − 𝑖𝑢.

𝛼 ∗ 𝛽 + 𝛽 ∗ 𝛼 = 0 → (𝑟 − 𝑖𝑠)(𝑡 + 𝑖𝑢) + (𝑟 + 𝑖𝑠)(𝑡 − 𝑖𝑢) = 0

The linear independent vectors define a matrix P, a linear map:

Proof by contradiction eq. 3.14:

𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼 = 𝜆1 ∙ cos 𝛼

𝑐𝑜𝑠 𝜃 ∙ cos 𝛼 + sin 𝜃 ∙ sin 𝛼 = cos (𝜃 − 𝛼)

We try the first one: 𝑎 = − sin 𝛼 ; 𝑏 = cos 𝛼

Check the second solution 𝑎 = sin 𝛼 ; 𝑏 = − cos 𝛼:

Let 𝑛𝑧 = 𝑐𝑜𝑠𝜃, 𝑛𝑥 = 𝑠𝑖𝑛𝜃𝑐𝑜𝑠𝜙, and 𝑛𝑦 = 𝑠𝑖𝑛𝜃𝑠𝑖𝑛𝜙.

Compute the eigenvalues and eigenvectors for the matrix 𝜎𝑛 :

With the above definitions this transforms to:

−𝑐𝑜𝑠 2 𝜃 + 𝜆2 − (𝑠𝑖𝑛2 𝜃𝑐𝑜𝑠 2 𝜙 + 𝑠𝑖𝑛2 𝜃𝑠𝑖𝑛2 𝜙) =