Learning #41: Quantum Condensed Matter — Cooper Pairs to Topological Edge States

BCS Superconductivity: Cooper Pairs and the Energy Gap

The surprising discovery at the heart of BCS theory (Bardeen, Cooper, and Schrieffer, 1957) is that electrons in a metal can attract each other. Electrons are negatively charged and repel via the Coulomb force in vacuum, yet in a crystal lattice a subtle phonon-mediated interaction overcomes this repulsion at low temperature. One electron distorts the ionic lattice slightly, leaving a region of net positive charge that attracts a second electron. The two electrons, separated in real space by up to several hundred nanometres and carrying opposite momenta and opposite spins, form a bound state called a Cooper pair.

The binding energy of a Cooper pair is exponentially small in the electron-phonon coupling constant λ. The BCS gap equation at zero temperature gives:

Δ(0) ≈ 2ℏωD exp(−1 / N(0)V)

where ωD is the Debye frequency of the lattice, N(0) is the density of states at the Fermi level, and V is the effective electron-phonon coupling. This gap Δ is the energy cost of breaking a Cooper pair; it opens up symmetrically about the Fermi energy and explains why superconductors have zero DC resistance — scattering events that would degrade normal current cannot break pairs if the thermal energy kBT < 2Δ.

The macroscopic quantum state of all the Cooper pairs is described by a single condensate wave function Ψ(r) = |Ψ| eiθ. The phase θ is globally coherent across the superconductor — a remarkable fact with immediate experimental consequences. The supercurrent density is:

Js = (nse2/m*) (A − (ℏ/2e) ∇θ)

where ns is the superfluid density and A is the vector potential. This is the London equation; it predicts the Meissner effect (flux expulsion) and flux quantisation in units of the flux quantum Φ0 = h / 2e ≈ 2.07 × 10−15 Wb.

The Josephson Effect: Phase Tunnelling Across a Junction

In 1962, Brian Josephson predicted that if two superconductors are separated by a thin insulating barrier, Cooper pairs can tunnel through coherently without any applied voltage. The supercurrent depends only on the phase difference φ = θ1 − θ2 across the junction:

I = Ic sin φ

This is the DC Josephson effect: a static supercurrent with no voltage drop, tunable between +Ic and −Ic by adjusting φ. When a DC voltage V is applied across the junction, the phase evolves at the Josephson frequency:

dφ/dt = 2eV / ℏ

producing an oscillating current at frequency f = 2eV / h ≈ 483.6 MHz per microvolt. This AC Josephson effect is so precise that it forms the metrological standard for the volt: the BIPM has used Josephson junction arrays since 1990 to realise the volt to better than one part in 10⁹.

Two Josephson junctions arranged in a loop form a SQUID (superconducting quantum interference device). The critical current oscillates as a function of enclosed magnetic flux with period Φ0. SQUIDs detect magnetic fields as small as a few femtotesla — far below the noise floor of any semiconductor sensor — making them indispensable in brain imaging (MEG) and fundamental physics experiments searching for axions and gravitational waves.

Bose-Einstein Condensation: The Macroscopic Wave Function

Cooper pairs carry integer spin (two spin-1/2 electrons combine to give spin 0 or 1), making them bosons. Bosons are not subject to the Pauli exclusion principle and can, at sufficiently low temperature, all occupy the same single-particle ground state. This is Bose-Einstein condensation (BEC), predicted by Einstein in 1925 following Bose's work on photon statistics. The condensate fraction below the critical temperature Tc scales as:

N0/N = 1 − (T/Tc)3/2

for a three-dimensional, non-interacting gas. Because all condensate atoms occupy the same quantum state, their collective behaviour is governed by a single macroscopic wave function — sometimes called the order parameter:

Ψ(r, t) = √n0 eiθ(r,t)

where n0 is the condensate density. The dynamics of this wave function obey the Gross-Pitaevskii equation, a nonlinear Schrödinger equation that accounts for mean-field interactions between atoms. From this equation arise quantised vortices (circulation is quantised in units of h/m), sound-like Bogoliubov excitations, and interference fringes between two independent condensates — the latter first observed at MIT in 1997 by Andrews et al.

Alkali atom BECs (rubidium-87, sodium-23, lithium-7) were first realised in 1995 at temperatures of around 170 nanokelvin. Weakly interacting BECs are now used as pristine quantum simulators: by loading them into optical lattices one can emulate the Hubbard model, reproduce topological band structures, and even study the analogue of Hawking radiation near a sonic horizon in the condensate flow.

Topological Insulators: Bulk-Edge Correspondence

A topological insulator (TI) looks insulating in its interior but supports perfectly conducting states on its surfaces or edges. The surface states are not an accident of surface chemistry — they are guaranteed by the bulk-edge correspondence: a topological invariant computed from the bulk band structure dictates how many conducting edge modes must exist at any boundary with a trivially insulating region (including vacuum).

The simplest two-dimensional example is the quantum Hall state. Under a large perpendicular magnetic field, the bulk develops flat Landau levels with a gap between them, while the sample edges carry chiral edge modes that propagate in one direction only. The integer quantum Hall conductance is:

σxy = n e2 / h

where n is a Chern number — an integer topological invariant that counts how many times the occupied Bloch states wind around the Brillouin zone torus. Because n is an integer and cannot change without closing the bulk gap, the Hall conductance is robustly quantised to one part in 10⁹ regardless of disorder or sample geometry. This is why the quantum Hall effect defines the resistance standard (the von Klitzing constant RK = h/e2 ≈ 25,813 Ω).

In three-dimensional time-reversal-invariant topological insulators (such as Bi₂Se₃ and Bi₂Te₃), strong spin-orbit coupling inverts the band gap at certain time-reversal-invariant momenta, generating a Z₂ topological invariant. The resulting surface states form a single Dirac cone with spin-momentum locking: an electron moving in a given direction has its spin locked perpendicular to its momentum. This suppresses backscattering from non-magnetic impurities (time-reversal symmetry forbids U-turns), making the surface conduction robust.

Current research seeks to combine TI surfaces with superconductors. Theory predicts that the interface hosts Majorana fermions — quasiparticles that are their own antiparticle. Majorana bound states encode qubits non-locally, making them immune to local decoherence and offering a route to topologically protected quantum computation.

Try It Yourself

The following interactive simulations let you probe these phenomena directly — adjust temperatures, coupling strengths, and junction parameters and observe the quantum behaviour in real time.

Closing Thought

What unites Cooper pairs, Bose-Einstein condensates, and topological edge states is the idea that quantum mechanics is not just a theory of individual particles — it is a theory of collective order. A macroscopic wave function, a globally coherent phase, a topological invariant: these are all ways in which quantum correlations lock together the behaviour of 10²³ particles and produce effects that have no classical analogue whatsoever. The fact that such effects are now engineered into qubits, voltage standards, and medical imaging devices suggests that the 21st century belongs, in large part, to quantum condensed matter.

If you want to go deeper, the natural next step is to study the Bogoliubov-de Gennes equations that unify the BCS and BEC pictures, and then to look at how topological band invariants (Chern numbers, Z₂ invariants) are computed from Berry curvature in the Brillouin zone. Each of those threads opens onto an active frontier of condensed matter physics.