Topological Background#

Topology in physics: genus, winding numbers, topological invariants. Why topological properties are robust against local perturbations.

Topology is the branch of mathematics that studies properties preserved under continuous deformations. A coffee mug and a donut are topologically equivalent (both have one hole); a sphere is topologically distinct from both (no holes). This section introduces the topological concepts that underpin topological quantum computing.

Topology: The Mathematics of Robustness#

The central insight is that topological properties are discrete invariants. A surface either has 0 holes or 1 hole or 2 holes; there is no continuous way to go from 0 to 1. This discreteness makes topological quantities inherently robust against small perturbations. Stretching, squeezing, or slightly deforming a surface does not change the number of holes.

The genus \(g\) of a closed surface counts the number of handles:

  • Sphere: \(g = 0\)

  • Torus (donut): \(g = 1\)

  • Double torus: \(g = 2\)

The genus is related to the Euler characteristic \(\chi = 2 - 2g\) by the Gauss-Bonnet theorem, which connects the global topology to the local curvature integrated over the entire surface:

\[\int_S K \, dA = 2\pi \chi\]

where \(K\) is the Gaussian curvature. This theorem is remarkable: no matter how you deform the surface locally, the integral of curvature over the whole surface remains fixed.

Winding Numbers#

A simpler topological invariant is the winding number. Consider a closed path in the plane that avoids the origin. The winding number counts how many times the path wraps around the origin:

\[n = \frac{1}{2\pi} \oint d\theta\]

This integer cannot change under continuous deformation of the path (as long as the path never crosses the origin). The winding number appears throughout physics:

  • Vortices in superfluids: The phase of the order parameter winds by \(2\pi n\) around a vortex core

  • Magnetic flux quantization: The phase of the superconducting order parameter in a loop is quantized, with the winding number counting the number of flux quanta

  • Band topology: The Berry phase accumulated by a Bloch state around the Brillouin zone characterizes the topological class of the band

Topological Invariants in Quantum Systems#

In condensed matter physics, quantum states of matter can be classified by topological invariants, discrete numbers that remain unchanged under smooth deformations of the Hamiltonian (as long as the energy gap does not close).

The integer quantum Hall effect provides the prototypical example. When a 2D electron gas is placed in a strong magnetic field, the Hall conductance is quantized:

\[\sigma_{xy} = \nu \frac{e^2}{h}\]

The integer \(\nu\) is the Chern number, a topological invariant of the occupied energy bands. This quantization is extraordinarily precise (parts per billion) because it is protected by topology, not by fine-tuning of material parameters.

Topological insulators are materials that are insulating in the bulk but conduct on their surfaces due to topologically protected edge states. The bulk topological invariant (a \(\mathbb{Z}_2\) index) guarantees the existence of these edge states, which cannot be removed by disorder or impurities.

Particle Exchange in Two Dimensions#

The connection to quantum computing comes through particle statistics. In three dimensions, exchanging two identical particles twice returns the system to its original state. The exchange operator satisfies \(P^2 = 1\), so \(P = \pm 1\), giving bosons and fermions.

In two dimensions, the story is different. The paths of two particles in 2D form braids: one exchange followed by another is topologically distinct from no exchange. The fundamental group of the configuration space is the braid group \(B_n\), which is infinite and non-Abelian (for \(n \geq 3\)).

This mathematical structure permits anyons: particles whose exchange statistics are described by arbitrary phases (Abelian anyons) or matrices (non-Abelian anyons). The key point is that these exchange properties are topological. They depend on how many times one particle goes around another, not on the speed, trajectory, or details of the path.

From Topology to Fault Tolerance#

The robustness of topological invariants translates directly into fault tolerance for quantum computation:

Property

Mathematical Analog

Physical Consequence

Discrete invariant

Genus, winding number

Error must overcome a finite gap

Global nature

Integral over entire surface

Local perturbations have no effect

Braid group structure

Non-commutative exchanges

Gates encoded in particle trajectories

Topological protection

Continuous deformations preserve invariant

Exponential suppression of errors

Topology is not geometry

Geometry cares about distances and angles. Topology cares only about connectivity and structure. A circle of radius 1 and a circle of radius 100 are topologically identical. This distinction is exactly why topological quantum states are robust: the environment can distort the “geometry” of the quantum state without affecting its “topology.”

Further Reading#

  • Nakahara, M., Geometry, Topology and Physics, 2nd edition, CRC Press (2003).

  • Hasan, M. Z. and Kane, C. L., “Colloquium: Topological insulators,” Rev. Mod. Phys. 82, 3045 (2010). doi:10.1103/RevModPhys.82.3045

  • Wen, X.-G., “Colloquium: Zoo of quantum-topological phases of matter,” Rev. Mod. Phys. 89, 041004 (2017). doi:10.1103/RevModPhys.89.041004