Skip to main content
AmigoFish

Condensed Matter Ep 2: The Many-Body Problem

From Fermi Podcast · · 1 hr 1 min

This plays from the publisher's own server. We do not host a copy.

Open at the source

Not rated yet

Description

Write down the honest quantum state of a carbon atom's six electrons and you need seven hundred and twenty terms — every one saying the same physical thing, differing only in which electron you decided to call "electron one". Go to twenty electrons and it is two point four billion billion terms. The labels are a fiction: electrons are genuinely identical, and every one of those terms exists only to undo a naming decision you should never have made.

This conversation replaces the whole apparatus. Stop naming particles; start counting occupations. A row of boxes, each either lit or dark, and two operators whose entire job is to switch a box on or off. From that, three things fall out that no one has to impose. The Pauli exclusion principle becomes arithmetic — aim two creation operators at the same box and the swap rule makes the expression equal to minus itself, so it is zero, and the reason a chair holds your weight is one sign in one algebraic identity. Particle number leaves the notation, so the same operator describes one electron or ten to the twenty-three. And the ground state stops being empty: it is a filled sea, the useful objects are its ripples, and a missing electron below the surface behaves like a real positive particle — which is why aluminium's Hall voltage comes out with the wrong sign, and why the positron was predicted before it was found.

Then the model itself, built term by term rather than quoted. Electrons hop between neighbouring sites — the same tunnelling amplitude from the first conversation, in its third costume. Two electrons on the same site pay a fee. That is the entire Hubbard model, written in nineteen sixty-three, and it is the simplest hard problem in physics: in one dimension it was solved exactly by Lieb and Wu; in two it has resisted six decades of effort, and the reason is the same minus sign that gave us exclusion — it wrecks the Monte Carlo methods that work everywhere else, which is why people have started building the model out of cold atoms instead. And the two terms disagree so productively that magnetism appears in a Hamiltonian containing no magnetic term at all.

One thing is deliberately not earned. The quasiparticle — the claim that a soup of strongly interacting electrons behaves like a gas of weakly interacting ones — is stated precisely and then refused. Nothing here justifies it. It is Landau's hypothesis, and the next conversation has to pay for it.

——— Series: Physics - Tutorial · Grad level. Every equation spoken in full, no chalkboard required. Built by Fermi AI.

Back to Fermi Podcast · Report a problem