The Inductor Hides the Charge
The Inductor Hides the Charge
A transmon buys its insensitivity to charge noise by flattening its energy landscape with a huge shunting capacitor — but flatness has a price: the plasma frequency and anharmonicity it can offer are both modest, boxed in by the same capacitor that protects it. Fluxonium refuses the trade. In place of, or alongside, that capacitor it adds a superinductor — a long chain of Josephson junctions wired to behave as a single, enormous linear inductance, orders of magnitude larger than any inductor a foundry can wind from wire.
That superinductor does something a capacitor cannot: it shunts the charge itself, routing the offending fluctuations to ground through an element with vanishing dissipation, rather than merely diluting their effect across a wide capacitance. The qubit's two lowest states are pulled apart into a deep double well, separated by a barrier the superinductor itself enforces, protected simultaneously from charge noise (via the inductive shunt) and, in its cousin the 0-π qubit, from flux noise as well (via a topological symmetry between two nearly degenerate wells). The coherence a transmon buys with size, fluxonium buys with an exotic circuit element that has no equivalent in ordinary electronics.
Where a transmon hides from noise by growing wide, fluxonium hides from it by growing an inductance that does not exist outside a superconducting circuit — a hiding place built not out of a bigger version of an ordinary part, but out of a part invented for the purpose.
Seed: Fluxonium / 0-π qubit.
Central fact: Fluxonium shunts a Josephson junction with a superinductor — a chain of junctions acting as an enormous linear inductance — protecting the qubit from charge noise directly, rather than diluting the effect with a large capacitor as the transmon does.
