36
Part One · Quantum mechanics

Ladder operators

In the last chapter you solved the oscillator with a power series, a page of algebra. Paul Dirac found a way to get the same answer in four lines, without solving any differential equation at all. His trick was to factor the energy into two operators that move you up or down one rung of the ladder. Those two operators turned out to be far more than a trick. Applied to the electromagnetic field, the step up one rung is the creation of a photon.

On the plate, the ladder of oscillator states: each rung one quantum higher than the last.

Factoring the energy

In natural units (ℏ=m=ω=1\hbar = m = \omega = 1) the oscillator energy is H=12(p2+x2)H = \tfrac12(p^2 + x^2). For numbers, x2+p2=(x−ip)(x+ip)x^2 + p^2 = (x - ip)(x + ip). For operators the order matters (Chapter 29), so define

a=x+ip2,a†=x−ip2,[a,a†]=1a = \frac{x + ip}{\sqrt2}, \qquad a^\dagger = \frac{x - ip}{\sqrt2}, \qquad [a, a^\dagger] = 1

Multiplying out gives a†a=12(x2+p2)−12a^\dagger a = \tfrac12(x^2 + p^2) - \tfrac12, so H=ℏω (a†a+12)H = \hbar\omega\,(a^\dagger a + \tfrac12). The operator N=a†aN = a^\dagger a is the number operator. Everything about the oscillator now follows from the single relation [a,a†]=1[a, a^\dagger] = 1.

The ladder, from algebra alone

Using the commutator, [N,a†]=a†[N, a^\dagger] = a^\dagger. So if N∣n⟩=n∣n⟩N|n\rangle = n|n\rangle, then N(a†∣n⟩)=(n+1) a†∣n⟩N(a^\dagger|n\rangle) = (n + 1)\,a^\dagger|n\rangle: a†a^\dagger raises the number by one, and aa lowers it by one. With normalization:

a†∣n⟩=n+1 ∣n+1⟩,a∣n⟩=n ∣n−1⟩a^\dagger|n\rangle = \sqrt{n+1}\,|n+1\rangle, \qquad a|n\rangle = \sqrt{n}\,|n-1\rangle

Going down cannot go on for ever, because ⟨n∣N∣n⟩=∥a∣n⟩∥2≥0\langle n|N|n\rangle = \|a|n\rangle\|^2 \ge 0. The ladder must stop at a state with a∣0⟩=0a|0\rangle = 0, whose energy is 12ℏω\tfrac12\hbar\omega. Writing a∣0⟩=0a|0\rangle = 0 in position space is a first-order equation, (x+∂x)ϕ0=0(x + \partial_x)\phi_0 = 0, solved by the Gaussian. Climb to 3 and the plate shows the factors you collect on the way. Established

Matrix elements for free

Because x=(a+a†)/2x = (a + a^\dagger)/\sqrt2, position connects each rung only to its neighbours:

⟨n∣x∣m⟩=12(m δn,m−1+m+1 δn,m+1)\langle n|x|m\rangle = \frac{1}{\sqrt2}\left(\sqrt{m}\,\delta_{n,m-1} + \sqrt{m+1}\,\delta_{n,m+1}\right)

The plate shows the matrix of xx: only the two diagonals next to the main one are filled. This is a selection rule. A harmonic molecule driven by light, which couples through xx, can only climb one rung at a time, which is why the infrared spectrum of Chapter 35 has one strong line. The weak overtones come from the anharmonic Morse shape.

A picture of a quantum state

A classical oscillator is a point in the plane of position and momentum, going round a circle. A quantum state cannot be a point, but Eugene Wigner found a function on that plane, W(x,p)W(x, p), that comes as close as possible: integrate it over pp and you get ∣ψ(x)∣2|\psi(x)|^2; integrate over xx and you get the momentum distribution.

But it is not a true probability, because it can go negative. Look at the number state with 1 quanta. Amber is positive and red is negative. A state of exactly one quantum is negative at the centre. No classical spread of positions and momenta could produce it, and that negativity is now used as a test that a state is truly quantum. Established

DeeperFor the number state ∣n⟩|n\rangle, Wn(x,p)=(−1)nπe−(x2+p2)Ln ⁣(2(x2+p2))W_n(x,p) = \frac{(-1)^n}{\pi}e^{-(x^2+p^2)}L_n\!\big(2(x^2+p^2)\big), with LnL_n the Laguerre polynomial. At the origin it is (−1)n/π(-1)^n/\pi.

The most classical quantum state

Number states have a definite count of quanta but no definite phase; nothing in them oscillates. The state that does oscillate like a classical pendulum is the coherent state, an eigenstate of the lowering operator, a∣α⟩=α∣α⟩a|\alpha\rangle = \alpha|\alpha\rangle. Its Wigner function is the ground-state Gaussian, displaced to the point α\alpha and circling the origin, exactly like the classical point.

Set its size, ∣α∣=|\alpha| = 2.00. Count its quanta and you get a Poisson distribution with mean ∣α∣2|\alpha|^2, shown in the bars. Roy Glauber showed in 1963 that a laser produces exactly this kind of state, and shared the 2005 Nobel Prize for it.

Squeezing the noise

The uncertainty principle limits the product σxσp\sigma_x\sigma_p, not each factor. A squeezed state has less noise in one direction and more in the other. Squeeze by 0.6 and the round blob becomes an ellipse, still with the minimum area.

This is engineering now, not theory. Since 2019 the LIGO gravitational-wave detectors have injected squeezed light into their interferometers, reducing the quantum noise in the quantity they measure and letting them hear more distant black-hole mergers. Established

What you now have

The ladder operators and [a,a†]=1[a, a^\dagger] = 1, the spectrum from algebra, the selection rule for xx, the Wigner picture and its negativity, coherent states with Poisson statistics, and squeezed states. When Part Five quantizes fields, each mode of light will be one of these oscillators, and a†a^\dagger will create a photon.

· · ·

Three questions to carry with you.

Why must the ladder have a bottom rung?tap
⟨N⟩ = ‖a|n⟩‖² cannot be negative, so lowering must end at a state with a|0⟩ = 0.
What does a negative Wigner function tell you?tap
That the state cannot be any classical mixture of positions and momenta: it is non-classical.
What kind of state does a laser emit?tap
Approximately a coherent state: definite phase, Poisson-distributed photon number.

And one to work out. A coherent state has ∣α∣2=4|\alpha|^2 = 4. What is the probability of finding zero quanta, e−∣α∣2e^{-|\alpha|^2}?

P(0) =

Sources: P. A. M. Dirac, The Principles of Quantum Mechanics. E. Wigner, Phys. Rev. 40, 749 (1932). R. J. Glauber, Phys. Rev. 131, 2766 (1963). C. C. Gerry and P. L. Knight, Introductory Quantum Optics. M. Tse et al., Phys. Rev. Lett. 123, 231107 (2019) (squeezed light in LIGO). Checked by code: W₁(0,0) = −1/π; Poisson P(0) for mean 4 = 0.0183.

Next · Chapter 37
Central potentials

Every orbital shape, s, p and d, comes from one fact: the force points to the center.

Turn the page