Superconducting Order Parameter and Gauge-Invariant Phase
Node role: Concept.
Reader task: understand which superconducting phase differences are observable and how they connect to circuit flux variables.
Why it matters
Josephson junctions, flux-biased loops, and superconducting qubits all use a phase variable. The useful variable is not an absolute phase assigned to one piece of metal. It is an oriented, gauge-invariant phase difference whose definition includes the electromagnetic vector potential.
That distinction prevents a common but serious shortcut: circuit node flux is not simply the absolute phase of a macroscopic wavefunction multiplied by a constant.
Read this as three linked questions: what carries order, which phase difference is observable, and how that difference becomes a circuit branch coordinate. The microscopic distinctions and limits remain separate from the circuit map.
The collective object
Reorganized explanation. The collective order parameter, its relation to BCS pairing, and the distinction from the microscopic gap are summarized from superconductivity theory (Introduction to Superconductivity, n.d.).
A coarse-grained superconducting order parameter can be written
\[ \Psi(\mathbf r)=|\Psi(\mathbf r)|e^{i\theta(\mathbf r)}. \]
\(\Psi\) is a collective field, not a one-electron wavefunction. Its local phase \(\theta\) is used below to construct the observable branch phase, but an isolated value of \(\theta\) is not observable.
The magnitude \(|\Psi|\) measures the strength of superconducting order under a stated normalization; \(|\Psi|^2\) is not automatically an exact Cooper-pair count in every convention. The microscopic pair potential \(\Delta\) is a different object: its magnitude sets the quasiparticle gap in conventional BCS theory, but it is not generally the binding energy of one independent bosonic molecule.
The BCS ground state is a coherent many-electron state built from overlapping pair correlations. Conventional BCS pairs are therefore not non-overlapping exact bosons occupying independent single-pair orbitals.
From local phase to an observable
Reorganized explanation. Gauge invariance combines endpoint phases and the vector-potential integral; neither endpoint phase alone is an observable (Introduction to Superconductivity, n.d.). The signs below follow the declared path orientation.
Choose a path from superconducting electrode 1 to electrode 2 and orient every electrical quantity in that same direction. Define
\[ \boxed{ \delta_{12} =\theta_2-\theta_1 -\frac{2\pi}{\Phi_0}\int_1^2\mathbf A\cdot d\mathbf l } \]
with
\[ \Phi_0=\frac{h}{2e}, \qquad \varphi_0=\frac{\Phi_0}{2\pi}=\frac{\hbar}{2e}. \]
\(\Phi_0\) is the superconducting flux quantum and \(\varphi_0\) is the reduced flux quantum; both have units of webers. Under a gauge transformation, \(\mathbf A\) and the local phases change together while \(\delta_{12}\) remains unchanged. Reversing the branch orientation changes the sign of \(\delta_{12}\).
This is the phase that enters a Josephson current-phase relation. It is a difference across a declared branch, not a phase owned by either endpoint in isolation.
How circuit flux enters
This-page derivation. Start from node flux as integrated voltage (Vool and Devoret 2017, sec. 2.1.7) and express the branch phase with the selected static loop offset. This connects two descriptions; it does not identify absolute condensate phase with absolute node flux.
Circuit theory introduces a dimensional node-flux coordinate
\[ \Phi_n(t)=\int_{t_0}^{t}V_n(t')\,dt', \qquad [\Phi_n]=\mathrm{V\,s}=\mathrm{Wb}. \]
An oriented branch flux is assembled from node-flux differences and any declared loop-flux offset. For a Josephson branch, one may then write
\[ \delta_b=\frac{\Phi_b-\Phi_{b,\mathrm{offset}}}{\varphi_0} \]
under the selected tree, branch orientation, and static-flux convention.
Superconducting description
Start from local phases \(\theta_1\), \(\theta_2\) and the vector potential along an oriented path. Their observable combination is the gauge-invariant branch phase \(\delta_b\).
Circuit description
Integrate oriented branch voltage in time and include the declared static flux offset. This produces the branch flux used by the equation above.
The two descriptions meet at the oriented branch phase. They do not imply \(\Phi_n=\varphi_0\theta_n\) as a universal identity for absolute node variables.
The Ginzburg–Landau and BCS applicability distinctions below are reorganized from their superconductivity context (Introduction to Superconductivity, n.d.), not circuit-size rules.
- A junction model must declare branch orientation before current, voltage, and phase signs can be audited.
- A loop model must include the vector-potential or flux-offset contribution; endpoint phases alone do not close the loop constraint.
- A circuit artifact should use \(\Phi\) for dimensional generalized flux and a different symbol such as \(\delta\) or \(\varphi\) for dimensionless phase.
- Quasiparticle and gap physics can affect loss, but that does not make \(\Delta\) interchangeable with the circuit-scale Josephson energy \(E_J\).
The boundary of this description is equally important:
- Ginzburg–Landau theory is a controlled expansion near \(T_c\); using it far below \(T_c\) is an effective model, not a stronger derivation.
- Near boundaries, vortices, weak links, or strong nonequilibrium drive, amplitude variations may be as important as phase variations.
- The coherence length describes spatial recovery of superconducting order. By itself it is not a universal lithographic lower bound for a tunnel junction.
- Different sign conventions for electron charge and branch orientation are valid only when phase, voltage, current, and vector potential change consistently.
- Treating \(|\Psi|^2\) as an exact pair density without naming the adopted normalization can turn a qualitative model into a false measurement claim.
Connections
- Fluxoid and Flux Quantization closes the gauge-invariant phase around a superconducting loop.
- Josephson Current, Phase, Energy, and Inductance turns the branch phase into an electrical element model.
- Circuit Models to Bare Coordinates, Open EOM, and Normal Modes defines node flux independently as the time integral of node voltage and builds branch variables without identifying node flux with condensate phase.
- A quantum engineer’s guide supplies the broader superconducting-circuit design context.
- source-owner parameter tables records the project-facing phase vocabulary that consumes this distinction.
References
- L. N. Cooper, “Bound Electron Pairs in a Degenerate Fermi Gas,” Physical Review 104, 1189 (1956), doi:10.1103/PhysRev.104.1189.
- J. Bardeen, L. N. Cooper, and J. R. Schrieffer, “Theory of Superconductivity,” Physical Review 108, 1175 (1957), doi:10.1103/PhysRev.108.1175.
- L. P. Gor’kov, “Microscopic Derivation of the Ginzburg–Landau Equations in the Theory of Superconductivity,” Soviet Physics JETP 9, 1364 (1959), author manuscript record.
- Y. Nambu, “Quasi-Particles and Gauge Invariance in the Theory of Superconductivity,” Physical Review 117, 648 (1960), doi:10.1103/PhysRev.117.648.
| Field | Value |
|---|---|
| Status | Source-backed |
| Used by | Fluxoid and Flux Quantization, Josephson Current, Phase, Energy, and Inductance, Circuit Models to Bare Coordinates, Open EOM, and Normal Modes, and Circuit Lagrangian, Hamiltonian, and Canonical Quantization. |
| Implementation links | Workbench physics handoff and Workbench symbol glossary. |
| Open review question | Can a new designer explain why the observable is a branch phase difference and why it is not an absolute node phase? |