3.7.

4 Hermitian Operator
• The wavefunction can be complex (no observable)
• Measured quantity is always real (observable)

à Operator can be complex

• In order to fulfill that the expectation value is real , the following applies
for the operator A

and

à This quality denotes the operator A as Hermitian

In general with and

In QM all operators must be Hermitian, due to the fact the eigenvalues are observables + real
 3.7.5 Properties of Operator
If two eigenfunctions of a Hermitian operator have two different eigenvalues, they
are orthogonal to each other. It should be true that (here in 3D dx à d3x)
1 2

Proof:
- Multiply equation 1 from left with Y*2 + integrate

- Multiply the complex conjugate of equation 2 from left with Y1 + integrate

- Form the difference

Def. of a Hermitian operator à zero, q.e.d

 3.8 Dirac Notation (Bra-ket Notation)
The Dirac notation is a mathematical formalism that is well suited for QM

Key ideas:
- A quantum state is presented by a ket,
column vector with
elements a1, a2,…aN

- The complex conjugate of a ket is a bra with

raw vector whose elements
are the complex conjugates
of the ket elements ai
- Inner product (scalar product)
projection of one vector onto another

i. e. amount of overlap of two states

example 1:
total overlap
of a state with itself
integral of the probability normalization
density function condition
example 2:
two orthogonal states no overlap
 3.8.1 Eigenvalue Equation
- An operator Q transforms any given ket into another ket

- The adjoint of an operator (i. e. transpose of its complex conjugate) transforms any given
bra into another bra

- All operators have a set of kets on which their action is to multiply the ket by a scalar q
eigenvalue equation for
operator Q

- Typically the set of kets that satisfy an eigenvalue equation are denoted by where
n = 1, 2, 3, … and their corresponding scalars by qn

- Eigenvalue equation for Q:

where n = 1, 2, 3, …

à are called eigenkets (eigenstates, eigenfunctions) with n = 1,2,3,...

à Corresponding scalar factors qn are called eigenvalues

à Complete set of eigenvalues are called eigenvalue spectrum
 3.8.2 Hermitian Operator
- An Hermitian operator equals its adjoint

- Hermitian operators have three special mathematical properties:

à They have real eigenvalues

à They have orthogonal eigenfunctions

with d Kronecker-Delta

à The eigenfunctions form a complete set

- Matrix element of an operator Q

an operator can be
thought of as a matrix

à The matrix element of an Hermitian operator equals its complex conjugate

formal definition of an
Hermitian operator
 3.8.3 Expectation Value
- The expectation value of a given operator is related to its matrix element by:

- The expectation value of a Hermitian operator Q, where is an eigenstate of Q,

corresponds to one of its eigenvalues, qn

- Since the eigenvalues of a Hermitian operator are real, then the expectation value of Q is real
+ equals its complex conjugate

- Expectation value of a Hermitian operator Q in a superposition state (see later)

 3.8.4 Superposition States - 1
- The eigenstates of Hermitian operators are orthogonal to one another (i. e. they are linearly
independent / they form a complete set)

à They constitute a basis set of eigenstates

à Any state can be written as a linear superposition of eigenstates (where n = 1, 2, 3, …)

coefficient cn represents the

amount of the nth eigenstate
in the superposition

- Analogous: The x, y, and z unit vector ( , , ) constitute the basis set of vectors for 3D
Euclidean space
à all other vectors in this space can be expressed as a linear superposition of , ,

ax, ay, az represents the amount of a

along each of the three axis
 3.8.4 Superposition States - 2
- The nth coefficient: inner product of the nth eigenstate
with the superposition state

- Expectation value of a Hermitian operator Q in a superposition state

sum of the expectation

values of pure states

à The expectation value comprises the eigenvalues of all the eigenstates in the superposition

But: a measurement must return a single eigenvalue

What do we measure?

à QM is probabilistic à thus we can only determine the probability of a particular outcome

à The probability of measuring q1 is |c1|2, the probability of measuring q2 is |c2|2 , …

à The normalization condition is that the sum of probabilities equals one

 3.8.5 Identity Operator
- The identity operator I leaves a state unchanged

- The identity operator can be expressed in terms of eigenstates by considering a

superposition state

à The operation that leaves a state unchanged is the sum of the outer product of each eigenstate
with itself
sum of the outer product

where n = 1, 2, 3, ..
 3.8.6 Example 1: Matrix Elements for the
Hamilton Operator
- Energy eigenstates of the infinite square well, orthonormal basis (see chapter 4)

- Calculate matrix representation in this basis

(i) Hamilton operator

- Expectation value of energy for a superposition state

 3.8.7 Example 2: Matrix Elements for the
Momentum Operator – 1
- Energy eigenstates of the infinite square well, orthonormal basis (see chapter 4)

- Calculate matrix representation in this basis

(ii) Momentum operator

- Calculation of matrix element (this is not so simple, see next slide)

à Calculation of expectation value in energy eigenstate yields zero

 3.8.7 Example 2: Matrix Elements for the
Momentum Operator – 2
- Calculation of the expectation value for a superposition state:

à The result is not zero + time depending

Chapter 4: Wave Mechanics
 Outlook: Time-evolution Postulate

Postulate 4 (time-evolution postulate): the time evolution of the wavefunction is

governed by the Schrödinger equation:

à Hamiltonian operator H is the operator that represents a particle’s total energy

 4.1 Mathematical Operators
An operator is an object that acts on and modifies another mathematical object

• Wavevector operator:

• Frequency operator:

Example:

- Consider a plane wave (frequency w, wavevector k) as wavefunction

- Acting on the wavefunction with operator k:

eigenvalue equation for operator

k with its eigenvalue k
4.1.1 Examples of Mathematical Operators
An operator is an object that acts on and modifies another mathematical object

• Wavevector operator:

• Frequency operator:

• Momentum operator:

• Kinetic energy operator:

 4.1.2 Hamiltonian Operator
Hamiltonian operator H represents a particle’s total energy
(kinetic energy + potential energy)

potential energy operator V: here unknown

function on space (constant in time)

• Hamiltonian:

• Momentum:

• Kinetic energy:

• Potential energy:
4.2 Time-dependent Schrödinger Equation
Deriving the Schrödinger equation:
- The total energy of a quantum mechanical particle is:

- Acting on a wavefunction with H:

Time-dependent Schrödinger equation (TDSE)

Hamiltonian, H

à Describes the time evolution of a particle of mass m, moving under the

influence of a potential energy function V(x)
à Describes the properties of particles on atomic + sub-atomic scales
à Quantum analog of Newton’s law
4.2.1 Separating the Schrödinger Equation

Hamiltonian, H
• Separate into two parts: a spatial part (only x) + a temporal part (only t)

• Wavefunction as the product of a function in space F(x) + a function in time T(t)

à Reduces to two ordinary differential equations:

1.) à equation only of x

2.) à equation only in t

variable E is a constant; equals the total energy of the system

Allows to distinguish a particle’s spatial behavior from its time evolution

 4.2.2 Time-independent Schrödinger Equation

• x-equation: corresponds to the Hamiltonian

• t-equation: corresponds to , where

Time-independent Schrödinger equation (TISE)

Hamiltonian, H
4.2.3 Example: Constant Potential V(x) = V0
• Starting point: TISE + set V(x) = V0 with V0< E

with kinetic energy T

• Rearranging + use term for kinetic energy T

à Standard second-order differential equation with complex exponential solution

A, B are constants set by

with boundary conditions

 4.2.4 Temporal Part of the TDSE

• First-order differential equation

• Independent of the potential energy function V(x)

à Solution T(t) is independent of the system under consideration

C is a constant set by
boundary conditions
with
 4.2.5 Total Solution
The total solution is given by the product of space + time solution:

- T(t) is universal for all potentials V(x)

à Just need to solve f(x) + multiply the solution by

- Normalization condition:

- For a constant potential V0 (see example), the overall wavefunctions are a

superposition of traveling plane waves:
must be confined to a finite
region since infinite plane
waves are not normalizable
+x-direction -x-direction