We now derive the central result of angular-momentum theory: the allowed eigenvalues of
L^2 and L^z. The argument uses only the commutation algebra and the fact that the
ladder must terminate — no differential equations required.
Setting up the eigenvalue problem
Because L^2 and L^z commute, they share simultaneous eigenstates. Label them
∣λ,μ⟩ by their eigenvalues,
L^2∣λ,μ⟩=ℏ2λ∣λ,μ⟩,L^z∣λ,μ⟩=ℏμ∣λ,μ⟩,
with λ,μ dimensionless. Since L^2−L^z2=L^x2+L^y2 is a
sum of squares of Hermitian operators, its expectation value is non-negative, giving the bound
λ≥μ2.
So for fixed λ the value of μ cannot grow without limit — there must be a largest and a
smallest μ.
The ladder must terminate
Let μmax be the top rung. Then L^+∣λ,μmax⟩=0, otherwise it would
produce a state with larger μ. Using L^−L^+=L^2−L^z2−ℏL^z and acting on the top state gives
0=ℏ2(λ−μmax2−μmax)⟹λ=μmax(μmax+1).
Similarly, the bottom rung μmin satisfies L^+L^−∣λ,μmin⟩=0,
which gives λ=μmin(μmin−1). Equating the two expressions for λ yields
μmin=−μmax.
Quantization
Starting at μmin=−μmax and applying L^+ repeatedly must land exactly on
μmax after an integer number N of steps:
μmax=μmin+N=−μmax+N⟹μmax=2N,N=0,1,2,…
Defining ℓ≡μmax, this means ℓ can take the values
0,21,1,23,2,…. The eigenvalues are therefore
L^2∣ℓ,m⟩=ℏ2ℓ(ℓ+1)∣ℓ,m⟩,L^z∣ℓ,m⟩=ℏm∣ℓ,m⟩
with m running in unit steps from −ℓ to +ℓ.
Half-integers and orbital motion
The algebra permits both integer and half-integer ℓ. Orbital angular momentum, tied to a
single-valued spatial wavefunction eimϕ, allows only integer ℓ. Half-integer values are
realized by intrinsic spin, which has no spatial wavefunction to constrain it — a distinction we
explore later in the module.
Try it
For the d-orbital multiplet ℓ=2, compute the eigenvalue of L^2 in units of
ℏ2 (that is, ℓ(ℓ+1)) and return it.
Run your code to see the quantum state.
Sign in on the full site to ask questions and join the discussion.