Quantum Mechanics — Lecture notes for PHYS223

VI The harmonic oscillator

In quantum mechanics, the harmonic oscillator is an important paradigm because it provides a model for a variety of systems, such as the modes of the electrodynamic field (photons) and the vibrations of molecules and solids (phonons).

VI.1 Classical harmonic oscillator

The classical harmonic oscillator describes a particle subject to a restoring force F=-m⁢ω2⁢x proportional to the distance from an equilibrium position x=0. Newton’s equation m⁢x¨=F results in an oscillatory motion x⁢(t)=x0⁢cos⁡ω⁢t+(v0/ω)⁢sin⁡ω⁢t, where ω=2⁢π/T and T is the oscillation period. In this solution, x0=x⁢(0) is the initial position and v0=x˙⁢(0) is the initial velocity of the particle. According to F=-V′, the force F=-m⁢ω2⁢x corresponds to a parabolic potential energy

V⁢(x)=12⁢m⁢ω2⁢x2. (91)

VI.2 Schrödinger equation of the quantum harmonic oscillator

In order to solve this problem quantum mechanically, we follow our standard steps. The Schrödinger equation of the harmonic oscillator is given by

E⁢ψ⁢(x)=-ℏ22⁢m⁢ψ′′⁢(x)+12⁢m⁢ω2⁢x2⁢ψ⁢(x). (92)

This equation is again a linear differential equation of second order, but now one coefficient is position dependent. From our general considerations we already can anticipate the following:

  • •

    The solutions ψ⁢(x) are continuous.

  • •

    The derivatives ψ′⁢(x) of the solutions are also continuous.

  • •

    As V⁢(x)→∞ for |x|→∞, the particle cannot escape to infinity at finite energy E. This only permits bound states, which decay ψ⁢(x)→0 as |x|→∞.

  • •

    The energies of the bound states are discrete (i.e., only at certain energies we can find valid solutions of the Schrödinger equation).

  • •

    The ground state energy E0 will be larger than the classical minimal energy: E0>0.

VI.3 Mathematical solution

Let us first focus on finding the mathematical solutions of the Schrödinger equation. We start by bringing the Schrödinger equation into a simpler, rescaled form. Denote ψ⁢(x)=φ⁢(x⁢m⁢ω/ℏ), which corresponds to a rescaled position y=x⁢m⁢ω/ℏ. According to the chain rule, ψ′=m⁢ω/ℏ⁢φ′ and ψ′′=(m⁢ω/ℏ)⁢φ′′. We also write energy as E=ℏ⁢ω2⁢ϵ. In terms of φ, the Schrödinger equation then reads

ℏ⁢ω2⁢ϵ⁢φ⁢(y)=-ℏ⁢ω2⁢φ′′⁢(y)+ℏ⁢ω2⁢y2⁢φ⁢(y), (93)

or, after cancelling the common factor ℏ⁢ω2,

ϵ⁢φ⁢(y)=-φ′′⁢(y)+y2⁢φ⁢(y). (94)

Next, we introduce the new function φ⁢(y)=e-y2/2⁢f⁢(y). By applying the product rule, we then find the following standard differential equation:

ϵ⁢f=-f′′+2⁢y⁢f′+f. (95)

When ε≠2⁢n+1, where n=0,1,2,… is an integer, the solutions of Equation (95) grow rapidly for |y|→∞, even outgrowing the factor exp⁡(-y2/2) appearing in φ⁢(y)=e-y2/2⁢f⁢(y). These solutions do not fulfil our boundary conditions for |x|→∞ and hence have to be discarded.

For ε=2⁢n+1, n=0,1,2,…, however, Eq. (95) has simple solutions f⁢(y)=Hn⁢(y) which are just polynomials of degree n. These polynomials are known as the Hermite polynomials. The first two of these polynomials are simply H0=1 and H1=2⁢y. The other polynomials can be calculated recursively, Hn+1⁢(y)=2⁢y⁢Hn⁢(y)-2⁢n⁢Hn-1⁢(y).

VI.4 Ground state and first excited state

The first two solutions correspond to the ground state and the first excited state of the harmonic oscillator:

Ground state: n=0, f=1, which gives to f′=f′′=0. This is a solution of Eq. (95) if ϵ=1, which corresponds to an energy E0=12⁢ℏ⁢ω. The wave function is given by

ψ0⁢(x)=c0⁢exp⁡(-x2⁢m⁢ω2⁢ℏ),

where c0 can be determined from the normalisation condition ∫-∞∞|ψ⁢(x)|2⁢𝑑x=1 (see below). This gives c0=(m⁢ωπ⁢ℏ)1/4.

First excited state: n=1, f=2⁢y, hence f′=2 and f′′=0. This is a solution of Eq. (95) if ϵ=3, which corresponds to E1=32⁢ℏ⁢ω. The wave function is given by

ψ1⁢(x)=c1⁢x⁢exp⁡(-x2⁢m⁢ω2⁢ℏ),

with normalisation constant c1=(4⁢m3⁢ω3π⁢ℏ3)1/4.

VI.5 Normalisation of the ground state wave function

The ground state wave function of the harmonic oscillator provides us with a good occasion to practice once more the normalisation of the wave function. Since we want to interpret |ψ0⁢(x)|2=P⁢(x) as the probability density for position, we require

∫-∞∞|ψ0⁢(x)|2⁢𝑑x=1.

With ψ0⁢(x)=c0⁢exp⁡(-x2⁢m⁢ω2⁢ℏ) this integral reads

∫-∞∞c02⁢exp⁡(-x2⁢m⁢ωℏ)⁢𝑑x.

We use the standard integral ∫-∞∞exp⁡(-a⁢x2)⁢𝑑x=π/a, where we set a=m⁢ωℏ:

∫-∞∞c02⁢exp⁡(-x2⁢m⁢ωℏ)⁢𝑑x=c02⁢π⁢ℏm⁢ω.

The equates to 1 if c0=(m⁢ωπ⁢ℏ)1/4, in agreement with the value given in the previous section.

VI.6 General form of the bound states

In general, the normalised bound-state wave functions are given by

ψn=Cn⁢Hn⁢(x⁢α)⁢e-x2⁢α2/2, (96)

n=0,1,2,3,…, Cn=α1/2⁢π-1/4⁢(n!⁢2n)-1/2, where α=m⁢ωℏ.

The associated bound-state energies are

En=ℏ⁢ω⁢(n+12),n=0,1,2,…. (97)

Note that the difference between two consecutive energies is given by Planck’s relation E=ℏ⁢ω.

You could now draw diagrams of the potential, bound state energies, ground- and excited state wave functions, and the associated probability densities of the particle position.