Superfluidity and Superconductivity

Pair binding, condensation, off-diagonal long-range order, phase stiffness, an excitation gap, zero resistance, magnetic screening, flux quantization, and a Josephson current are objects and signatures often invoked in claims about coherent quantum matter. They are not synonyms. A defensible claim identifies which object was calculated or measured, which state and limit define it, and which alternatives the evidence actually excludes.

This gateway turns that declaration into the shortest useful route. It does not reproduce the derivations in the chapter leaves. It owns the distinction between neutral and charged coherent fluids, the regime choice among London, Ginzburg–Landau, BCS, vortex, Josephson, and proximity descriptions, and the evidence ledger needed before calling a material superfluid or superconducting.

The Quantum Matter Map owns the general system–state–observable problem tuple. Choosing a Model for Quantum Matter owns comparisons among non-nested model families. This page begins after the question has become one about coherent flow, pairing, electromagnetic rigidity, or a superconducting interface.

Helpful background. Neutral-fluid routes use Bose–Einstein Condensation, Off-Diagonal Long-Range Order, and collective-mode or response concepts. Fermionic routes use a declared Fermi surface, fermionic operators, and the BCS Mean-Field Theory owner. Electromagnetic routes additionally require gauge and response conventions. These are branch capabilities, not universal prerequisites.

Enter This Chapter

Choose the physical question before choosing a familiar theory.

• Is the system neutral? Use Superfluidity in Condensed Matter for the neutral-material claim ledger: condensation versus ODLRO and stiffness, static twist response, superfluid fraction, critical flow, two-fluid and sound response, and helium or film evidence. It applies the canonical BKT and finite-size owners rather than duplicating them.
• Is the claim equilibrium magnetic screening or flux response? Start with London Theory. It owns penetration depth, the Meissner distinction, fluxoid quantization, and the nonlocality limits of a stiffness-only electrodynamic description.
• Is the question about a slowly varying order parameter near a continuous transition? Use Ginzburg–Landau Theory for the material free-energy functional, coherence and penetration lengths, critical fields, and the static vortex solution in its controlled window.
• Is a microscopic weak-coupling fermionic account needed? Use BCS Theory for the pairing instability, reduced material model, self-consistent gap equation, quasiparticles, gap equation, thermodynamics, spectroscopy, and its limitations. The generic many-body derivation remains with BCS Mean-Field Theory and Bogoliubov Quasiparticles.
• Is the object a weak link or a phase difference? Use the Josephson Effect for current–phase dynamics, interference, driven phase locking, environmental dynamics, and the bridge to circuits.
• Is the object a moving or disordered vortex ensemble? First use Ginzburg–Landau Theory for the static core and type criterion. Then use Vortex Matter, Pinning, and Flux Flow for pinning, creep, collective phases, flux flow, and driven-vortex evidence.
• Is superconducting coherence leaking through a bulk interface region? Use Superconducting Proximity Effect for spatial anomalous amplitudes, inverse proximity, clean and diffusive coherence lengths, and controlled Eilenberger–Usadel reductions. Use Proximity and Andreev Physics for interface scattering, BTK conductance, Andreev levels, and hybrid devices.
• Is the pairing symmetry or mechanism nonconventional? BCS Theory gives the controlled baseline. Unconventional Superconductivity owns the stable crystal-, pseudospin-, and orbital-pairing taxonomy and the multi-probe evidence ladder. Material-specific high-temperature mechanism disputes remain with the frontier treatment.
• Is topology or device certification the real question? Route to Topological Superconductors or the appropriate circuit and quantum-information owner. A gap, a vortex, or a zero-bias feature does not by itself establish a topological phase or a qubit.

Coverage and Consolidations

All eight local leaves are substantive. Unconventional Superconductivity owns the settled pairing-symmetry and evidence taxonomy without adjudicating changing mechanism claims.

Nine empty prelaunch routes were consolidated into canonical owners. The generic Bose-condensation comparison belongs to the live neutral-superfluidity and many-body pages. The generic Cooper problem and pair algebra belong to BCS Mean-Field Theory; material BCS owns weak-coupling pair size and consequences; the Unconventional Superconductivity owner supplies crystal, pseudospin, and orbital pairing classification; and Pair-Density Waves and Exotic Orders owns finite-momentum and composite pairing. Gap spectroscopy and superconducting Bogoliubov quasiparticles are already developed in BCS Theory and the generic many-body owners. The Meissner effect belongs to London Theory; the type-I/type-II criterion to Ginzburg–Landau Theory; topological superconductivity to the live topology chapter; and superconducting circuits and qubits to Josephson, Circuit QED, and quantum-information hardware owners. The rapidly changing high-temperature mechanism record remains in the Frontiers chapter rather than being duplicated here.

Readiness Check

Before accepting a coherent-matter claim, be able to answer:

1. What material or platform, carrier charge, dimension, geometry, boundary, state, temperature, field, pressure, disorder, and preparation are meant?
2. Is the claimed object a pair, anomalous amplitude, condensate occupation, ODLRO eigenvalue, stiffness, current, gap, screening kernel, fluxoid, vortex, or phase difference?
3. Which symmetry is physical, which transformation is a gauge redundancy, and which phase convention has been chosen?
4. What observable and order of limits define the reported response?
5. Which probe, geometry, contacts, backgrounds, and resolution produce the recorded data?
6. What competing normal, ordered, inhomogeneous, finite-size, heating, or topological explanation must be excluded?

If any item is missing, repair it before selecting London, Ginzburg–Landau, BCS, Josephson, or proximity language. A fitted gap-shaped spectrum is not a complete superconducting diagnosis, and a model order parameter is not an instrument record.

Write the Coherence–Response Ledger

Record ten fields before comparing theories or experiments.

1. Platform and state. Give composition, carrier type, charge, lattice or continuum, dimension, geometry, boundary, ensemble, equilibrium or drive, preparation, temperature, field, pressure, disorder, and history.
2. Claimed coherent object. Name pair binding, condensate occupation, one- or two-body ODLRO, anomalous amplitude, stiffness, superfluid density, electromagnetic kernel, gap, current, fluxoid, vortex, or relative phase.
3. Symmetry and gauge convention. State the physical global and crystal symmetries, explicit sources, proposed spontaneous breaking, charge sign, vector-potential convention, and any auxiliary broken-symmetry saddle.
4. Microscopic or effective model. Declare the retained particles, bands, pairing channel, interaction or phenomenological functional, cutoff, and which parameters are derived, fitted, or measured.
5. Scale hierarchy. Compare Fermi or boson energies, pairing scale, temperature, stiffness, gap, coherence and penetration lengths, mean free path, sample dimensions, field, frequency, momentum, damping, and probe resolution.
6. Theory regime. Name London, Ginzburg–Landau, BCS, hydrodynamic, quasiclassical, Josephson, vortex-dynamics, or another description and state its small parameter, locality assumption, and failure test.
7. Requested observable. Specify a thermodynamic derivative, spectral function, conductivity kernel, penetration depth, current–phase relation, flux response, vortex velocity, or another normalized quantity and its order of limits.
8. Forward model. Connect the intrinsic object to contacts, fields, polarization, momentum selectivity, tunneling matrix elements, optical multilayers, domains, demagnetization, heating, backgrounds, and resolution.
9. Evidence and alternatives. Require complementary equilibrium, transport, thermodynamic, spectroscopic, phase-sensitive, or spatial evidence appropriate to the claim, with credible competing explanations.
10. Uncertainty and stopping rule. Separate model, parameter, solver, finite-size, sample, and measurement uncertainty; state the observation that would falsify, narrow, or escalate the interpretation.

The bounded conclusion should read: “for this state, geometry, limit order, model, and forward model, these observations support this coherent response over the stated alternatives.” It should not convert one transition, one fit, or one peak into a universal mechanism.

Keep the Coherent Objects Separate

A two-particle bound state can exist without macroscopic coherence. Occupation of a chosen orbital is basis dependent; BEC is a macroscopic eigenvalue of the one-body density matrix and is basis invariant. Neither condensate occupation nor ODLRO is identical to stiffness, the free-energy cost of a twist. A spectral gap is an excitation statement. Meissner screening is a transverse equilibrium response. Zero resistance is a transport statement. Josephson current is a relative-phase response of a coupled system.

For a slowly varying charged coherent field, a useful gauge-invariant combination is

$$\mathbf p_s=\hbar\boldsymbol\nabla\phi-q_\star\mathbf A,$$

where the signed coherent charge $q_\star$ and the transformations of $\phi$ and $\mathbf A$ must be declared consistently. A stiffness functional can depend on $\mathbf p_s$, but the phase alone is not an electromagnetic observable. Local electromagnetic gauge redundancy is not a physical symmetry that an exact state literally breaks. Broken-symmetry saddles remain useful; gauge-invariant currents, fluxes, correlations, and relative phases carry the physical content.

The response limit also matters. A conductivity probes a spatially uniform, time-dependent field, schematically $\mathbf q\to0$ before $\omega\to0$. Equilibrium magnetic screening is a static transverse spatial response and uses the opposite order. A delta function in conductivity and a Meissner kernel therefore answer different questions. A perfectly conducting normal state need not expel a field.

Likewise, distinguish a self-consistent pair field $\Delta(\mathbf k)$, the minimum positive Bogoliubov excitation energy, a pole or edge in a spectral function, a tunneling threshold, an optical onset, a pseudogap, and an induced minigap. In an ideal isotropic BCS model, the pair-field magnitude, minimum single-particle gap at the Fermi surface, and ideal tunneling edge can equal $\Delta$. A conventional optical pair-breaking scale is generally $2\Delta$ when regular absorption and its matrix elements are present; an ideal clean Galilean-invariant model at $\mathbf q=0$ can lack that regular onset. A pseudogap and proximity minigap are absent from the homogeneous baseline. The objects separate further in anisotropic, multiband, disordered, strongly coupled, inhomogeneous, or proximity systems.

Use a Branching Theory Graph

The chapter is not an accuracy ladder.

• Neutral many-body coherence branches from condensation and ODLRO to hydrodynamic response, stiffness, critical flow, dimensional crossover, and vortices.
• Charged long-wavelength equilibrium response branches to London Theory.
• A slowly varying material order parameter near its transition branches to Ginzburg–Landau Theory.
• A declared fermionic pairing model branches to BCS Theory and the generic Bogoliubov owner.
• Static vortices begin in Ginzburg–Landau Theory; collective vortex matter and driven flux motion continue in Vortex Matter, Pinning, and Flux Flow.
• Two coherent regions separated by a weak link branch to the Josephson Effect.
• A spatially inhomogeneous bulk interface branches to quasiclassical proximity theory; an interface-scattering or few-mode device question branches to Proximity and Andreev Physics.
• Crystal-symmetry classification and multi-probe inference branch to unconventional superconductivity; changing high-temperature mechanisms branch to Frontiers.
• BdG invariants, Majorana modes, and topological evidence branch to Topological Superconductors rather than remaining in this chapter gateway.

These branches overlap in real materials. A Josephson weak link may require proximity amplitudes, a vortex can carry subgap quasiparticles, and an unconventional order parameter can change London response. State the local owner for each step instead of forcing the complete problem into one theory.

Worked Audit: A Two-Dimensional Neutral Film

Suppose a thin neutral bosonic film develops a sharp change in flow response near a temperature $T_\star$. Momentum-distribution narrowing alone would be evidence for enhanced low-momentum occupation, not yet a superfluid diagnosis. The ledger must add the film size and thickness, interaction regime, boundary and disorder landscape, the imposed flow or twist, and the response observable.

The ten fields close as follows.

1. Platform and state: a neutral interacting bosonic film of declared area, thickness, density, temperature, boundary, preparation, and equilibrium window.
2. Claimed object: the helicity modulus or mass-flow stiffness, not merely one orbital occupation.
3. Symmetry and convention: a physical global $U(1)$ charge symmetry, a declared twist/source convention, and no electromagnetic gauge field.
4. Model: a continuum or lattice boson model reduced to a phase-only description only after density fluctuations and defects are shown to be controlled.
5. Scales: healing length, vortex-core scale, film dimensions, interaction, thermal energy, probe frequency, drive, disorder, and resolution.
6. Regime: equilibrium two-dimensional hydrodynamics or BKT scaling, with finite-size and nonequilibrium failure tests.
7. Observable: a twist derivative, mass-current response, or independently calibrated stiffness with the thermodynamic limit stated.
8. Forward model: imposed rotation, flow, resonance, interference, or other probe geometry connected explicitly to the inferred response.
9. Evidence and alternatives: vortex signatures and consistent stiffness scaling compared with heating, inhomogeneity, finite-size crossover, and a thresholded dissipation onset.
10. Uncertainty and stop: propagate size, temperature, density, calibration, and scaling-form uncertainty; stop at a finite-size crossover claim if no controlled thermodynamic extrapolation exists.

For a two-dimensional finite system, the route is Bose–Einstein Condensation for occupation language, Off-Diagonal Long-Range Order for correlation, Low-Dimensional Quantum Gases for BKT and trapped-system theory, and Finite-Size Effects for the extrapolation contract. The live neutral-superfluid page applies those owners to stiffness, critical flow, and a declared probe. A claimed BKT transition requires scaling or stiffness evidence compatible with vortex-pair unbinding; a broadened density feature or dissipation onset alone admits inhomogeneity, heating, finite-size crossover, and probe-threshold alternatives. The analysis stops at “finite-size evidence for a stiffness crossover” unless the thermodynamic extrapolation is licensed.

Worked Audit: A Dirty Superconductor–Normal Bilayer

Now suppose tunneling into a diffusive normal film on a bulk superconductor shows a soft low-energy suppression and the bilayer transition temperature is lower than that of the isolated superconductor. The question is spatial, not merely whether Andreev reflection occurs at one ideal interface.

Route the bulk calculation to Superconducting Proximity Effect. It owns the distinction among anomalous amplitude, local $\Delta$, and spectral minigap; the clean-versus-diffusive regime test; inverse-proximity suppression; and the self-consistent thickness, temperature, and field audit. Proximity and Andreev Physics retains BTK conversion, channel transparency, and discrete Andreev levels, while the probe owners retain tunneling convolution and resolution. At this gateway, record only enough geometry, disorder, interface, and probe information to choose those owners and to reject a unique-transparency or topological claim from one soft spectrum.

Exit Checkpoint

You are ready to leave the gateway when you can:

• distinguish pairing, condensation, ODLRO, stiffness, spectral gaps, screening, zero resistance, vortices, and relative-phase response;
• state the physical symmetry and gauge convention without saying that local electromagnetic gauge redundancy literally breaks;
• select a neutral, London, Ginzburg–Landau, BCS, Josephson, vortex, bulk proximity, unconventional, or topological route and give its validity test;
• declare the response observable and the order of limits;
• connect the intrinsic object to a specified probe and geometry;
• name at least one credible alternative and one stopping or escalation test;
• identify the canonical page that owns the next derivation or evidence audit.

Canonical Boundaries

• Computational Quantum Matter owns material-facing solver routing, convergence, benchmarks, uncertainty, and probe-comparison records; this gateway retains phase, regime, response, defect, interface, and pairing-claim selection.
• The generic many-body pages own ideal-gas condensation, ODLRO, spontaneous symmetry breaking, the BCS saddle, and bosonic or fermionic Bogoliubov transformations.
• London, Ginzburg–Landau, Vortex Matter, BCS, and Josephson pages own their material derivations. This gateway owns only route selection and claim readiness.
• The live neutral-superfluidity page owns neutral material response; Superconducting Proximity Effect owns spatial quasiclassical correlations; the vortex page owns collective and driven vortex matter; and the unconventional page owns stable pairing-symmetry and evidence taxonomy.
• Proximity and Andreev Physics owns interface scattering, Andreev levels, and hybrid devices. Topological Superconductors owns BdG invariants, Majorana modes, and topological claim audits.
• Terahertz and Infrared Probes and Scanning Tunneling Microscopy and Spectroscopy own the instrument, inversion, resolution, and matrix-element layers for their measurements.
• Circuit QED and Quantum Information own circuit quantization, qubit architectures, control, readout, coherence budgets, and system-level claims. A material chapter should not duplicate them.

Common Routing Errors

“BEC and superfluidity are the same transition.” Condensate occupation and stiffness are distinct observables. Interactions, dimension, disorder, and finite size can separate their useful diagnostics.

“Zero resistance proves superconductivity.” Momentum conservation, ballistic transport, contact effects, filamentary paths, and thresholds can produce very small resistance. Add equilibrium screening or flux evidence and the required complementary tests.

“The electromagnetic gauge symmetry breaks.” Gauge redundancy labels descriptions of the same physical state. Phrase the claim through gauge-invariant correlations, stiffness, currents, fluxes, and relative phases.

“London, Ginzburg–Landau, and BCS are successive approximations to every sample.” They answer different long-wavelength, near-transition, and microscopic questions. Their domains overlap but are not totally ordered.

“Every measured gap equals the order parameter.” State the measured threshold or spectral object and its forward model. An induced, pseudogap, optical, tunneling, quasiparticle, and self-consistent pairing scale can differ.

“One zero-bias peak or vortex mode is topological.” Ordinary Andreev, disorder, confinement, Kondo, heating, and Caroli–de Gennes–Matricon alternatives must be excluded, and the bulk phase must be established.

Exercises

1. Gauge-invariant stiffness

Under a gauge transformation

$$\mathbf A\longmapsto\mathbf A+\boldsymbol\nabla\chi, \qquad \phi\longmapsto\phi+\frac{q_\star}{\hbar}\chi,$$

show that $\mathbf p_s=\hbar\boldsymbol\nabla\phi-q_\star\mathbf A$ is unchanged. Explain why a phase expectation value alone is not an observable.

Solution

Substitution gives

$$\mathbf p_s' =\hbar\boldsymbol\nabla\phi+q_\star\boldsymbol\nabla\chi -q_\star\mathbf A-q_\star\boldsymbol\nabla\chi =\mathbf p_s.$$

The phase changes with the representative gauge, whereas currents, fluxes, correlation functions with the required gauge structure, and relative phases across a junction can be operationally defined. A broken-symmetry phase is a useful coordinate on the effective saddle, not by itself a gauge-invariant measurement.

2. Perfect conductivity versus Meissner response

Suppose a clean normal model has nonzero Drude weight—for example, a Galilean-invariant continuum at nonzero density. Does its zero-frequency conductivity delta function imply equilibrium magnetic flux expulsion? Identify the two limiting procedures.

Solution

No. Uniform conductivity takes the long-wavelength limit first and then probes low frequency. Static transverse screening takes the zero-frequency limit and then examines long wavelength. The limits need not commute. Momentum conservation can protect a transport delta function without producing a transverse Meissner kernel.

3. Route four claims

Choose the first canonical route for each question: the penetration depth of a bulk conventional sample; the temperature-dependent self-consistent gap; the current–phase relation of a weak link; and the spatial pair amplitude through a dirty normal layer.

Solution

Use London Theory for penetration depth, BCS Theory for the microscopic gap, and Josephson Effect for the weak-link current–phase relation. The dirty-layer pair amplitude belongs to Superconducting Proximity Effect, which tests the diffusive reduction and solves the spatial anomalous correlation and inverse suppression. BCS Mean-Field Theory still owns the reusable parent pairing saddle, Proximity and Andreev Physics owns BTK and few-mode device scattering, and Ginzburg–Landau Theory connects penetration and coherence lengths only near the transition.

4. Audit a gap claim

Tunneling shows a suppressed density of states near zero bias, while optical absorption has a different onset and the diamagnetic response is weak. List four objects that must not be identified automatically and two next tests.

Solution

Do not equate the tunneling threshold, optical onset, self-consistent pair field, minimum quasiparticle gap, pseudogap, or induced minigap. At least four of these suffice. Next tests should include an equilibrium stiffness or screening measurement and a second probe with a declared forward model—for example, calibrated temperature- and field-dependent tunneling, heat capacity, penetration depth, terahertz conductivity, or a phase-sensitive Josephson measurement. The weak diamagnetic signal requires checks for volume fraction, geometry, demagnetization, inhomogeneity, and resolution.

5. Diagnose a vortex-resistance signal

A type-II sample has nonzero resistance below its pairing onset only in a magnetic field. Give a route that does not call this the destruction of pairing by default.

Solution

Use Ginzburg–Landau Theory to establish the mixed-state and static vortex scales, then the vortex-matter route for pinning, creep, flux flow, collective phases, heating, and driven nonequilibrium. Compare current, field, frequency, sample geometry, disorder, and imaging or magnetic evidence. Mobile vortices can dissipate while the pair amplitude remains nonzero; the result need not mark the microscopic pairing transition.

6. Test a Ginzburg–Landau classification

Near a continuous transition, a single-component local fit gives $\xi=20\,\mathrm{nm}$ and $\lambda=100\,\mathrm{nm}$. Classify the minimal Ginzburg–Landau model and state why the classification is bounded.

Solution

The fitted ratio is $\kappa=\lambda/\xi=5$, which exceeds $1/\sqrt{2}$. The minimal local single-component Ginzburg–Landau model is therefore in its type-II regime. This conclusion is bounded by the assumed near-transition, local, single-component description. Nonlocality, multicomponent order, anisotropy, disorder, interfaces, demagnetization, and a temperature range outside the fit can invalidate a universal material label.

7. Audit a Josephson voltage–frequency claim

A junction under a constant voltage magnitude $|V|$ shows radiation at $f=2e|V|/h$. What does this establish, and what does it not establish?

Solution

With a declared voltage and phase convention, the frequency is compatible with the ordinary AC Josephson relation for charge-$2e$ coherence. It tests a relative-phase response and provides a metrological relation in its controlled regime. It does not by itself determine the microscopic pairing mechanism, the full current–phase relation, the electromagnetic environment, junction transparency, or topology. Those require the Josephson owner, device model, and complementary evidence.

8. Route an unconventional-pairing claim

Penetration depth and thermal transport are both consistent with nodes, while tunneling is sample dependent. What can be concluded before the pairing representation and mechanism are known?

Solution

The combined evidence can support low-energy nodal excitations over the stated temperature, field, disorder, and resolution windows. It does not yet fix the crystal or pseudospin representation, the sign structure, or the pairing mechanism. Check surface orientation, impurity and multiband alternatives, phonon and electronic backgrounds, and whether the probes weight the same regions of momentum space. Use BCS Theory for the baseline and Unconventional Superconductivity for the stable symmetry and evidence taxonomy; route changing mechanism claims to Frontiers.

9. Separate BEC, ODLRO, and superfluid response

Give one reason that a macroscopic condensate occupation does not by itself establish a nonzero critical velocity, and one reason a two-dimensional fluid can show superfluid response without true finite-temperature BEC.

Solution

An ideal homogeneous Bose gas can have macroscopic occupation while its quadratic excitation spectrum makes the Landau ratio $\inf_{\mathbf p\ne0}\epsilon(\mathbf p)/|\mathbf p|$ vanish, so the ideal model does not obtain a positive Landau bound from interactions. Conversely, a two-dimensional interacting fluid can have a BKT phase with algebraic order, finite stiffness, and vortex binding while the one-body density matrix lacks a thermodynamic macroscopic eigenvalue at nonzero temperature. Occupation, one-body ODLRO, stiffness, and measured critical flow must therefore be tested separately.

10. Audit the ideal BCS density of states

For a clean isotropic model,

$$\frac{N_s(E)}{N_n(0)} =\operatorname{Re} \frac{|E|}{\sqrt{E^2-\Delta^2}}.$$

State the subgap and high-energy limits, then explain why a rounded tunneling edge is not automatically a microscopic lifetime.

Solution

For $|E|<\Delta$, the displayed ideal expression has zero real part. For $|E|\gg\Delta$, the ratio approaches one, while an integrable square-root coherence edge occurs at $|E|=\Delta$. The normalization must use the same per-spin or total density-of-states convention in numerator and denominator. Measured conductance additionally convolves the sample spectrum with temperature, tip spectrum, matrix elements, and resolution. Disorder, anisotropy, multiple gaps, spatial inhomogeneity, pair breaking, and true lifetime effects can all round the edge, so a phenomenological broadening parameter is not uniquely a scattering rate.

References

1. J. Bardeen, L. N. Cooper, and J. R. Schrieffer, “Theory of Superconductivity,” Physical Review 108, 1175–1204 (1957), doi:10.1103/PhysRev.108.1175.
2. M. R. Beasley, “Notes on the Ginzburg–Landau Theory,” in Superconductivity, edited by R. D. Parks, Marcel Dekker, 1969.
3. W. Belzig, F. K. Wilhelm, C. Bruder, G. Schön, and A. D. Zaikin, “Quasiclassical Green’s Function Approach to Mesoscopic Superconductivity,” Superlattices and Microstructures 25, 1251–1288 (1999), doi:10.1006/spmi.1999.0710.
4. G. Blatter, M. V. Feigel’man, V. B. Geshkenbein, A. I. Larkin, and V. M. Vinokur, “Vortices in High-Temperature Superconductors,” Reviews of Modern Physics 66, 1125–1388 (1994), doi:10.1103/RevModPhys.66.1125.
5. P. G. de Gennes, Superconductivity of Metals and Alloys, Addison-Wesley, 1989 reprint.
6. G. Eilenberger, “Transformation of Gorkov’s Equation for Type II Superconductors into Transport-Like Equations,” Zeitschrift für Physik A 214, 195–213 (1968), doi:10.1007/BF01379803.
7. A. A. Golubov, M. Yu. Kupriyanov, and E. Il’ichev, “The Current–Phase Relation in Josephson Junctions,” Reviews of Modern Physics 76, 411–469 (2004), doi:10.1103/RevModPhys.76.411.
8. B. D. Josephson, “Possible New Effects in Superconductive Tunnelling,” Physics Letters 1, 251–253 (1962), doi:10.1016/0031-9163(62)91369-0.
9. J. M. Kosterlitz and D. J. Thouless, “Ordering, Metastability and Phase Transitions in Two-Dimensional Systems,” Journal of Physics C 6, 1181–1203 (1973), doi:10.1088/0022-3719/6/7/010.
10. A. J. Leggett, Quantum Liquids: Bose Condensation and Cooper Pairing in Condensed-Matter Systems, Oxford University Press, 2006.
11. D. R. Nelson and J. M. Kosterlitz, “Universal Jump in the Superfluid Density of Two-Dimensional Superfluids,” Physical Review Letters 39, 1201–1205 (1977), doi:10.1103/PhysRevLett.39.1201.
12. O. Penrose and L. Onsager, “Bose–Einstein Condensation and Liquid Helium,” Physical Review 104, 576–584 (1956), doi:10.1103/PhysRev.104.576.
13. D. J. Scalapino, S. R. White, and S. Zhang, “Insulator, Metal, or Superconductor: The Criteria,” Physical Review B 47, 7995–8007 (1993), doi:10.1103/PhysRevB.47.7995.
14. J. R. Schrieffer, Theory of Superconductivity, revised edition, Westview Press, 1999.
15. M. Sigrist and K. Ueda, “Phenomenological Theory of Unconventional Superconductivity,” Reviews of Modern Physics 63, 239–311 (1991), doi:10.1103/RevModPhys.63.239.
16. M. Tinkham, Introduction to Superconductivity, second edition, Dover Publications, 2004.
17. K. D. Usadel, “Generalized Diffusion Equation for Superconducting Alloys,” Physical Review Letters 25, 507–509 (1970), doi:10.1103/PhysRevLett.25.507.
18. C. N. Yang, “Concept of Off-Diagonal Long-Range Order and the Quantum Phases of Liquid He and of Superconductors,” Reviews of Modern Physics 34, 694–704 (1962), doi:10.1103/RevModPhys.34.694.