Anderson Localization: Why Disorder Can Freeze Wave Transport (Field Guide)

Date: 2026-03-16
Category: explore

The weird claim

Adding more randomness to a material can make transport not just worse, but effectively stop.

Not because particles are trapped by a classical barrier, but because waves repeatedly scatter and interfere with themselves until long-range transport dies.

That is Anderson localization.

Core intuition (without heavy formalism)

In a clean crystal, wave-like states can extend across the sample (Bloch-like transport).
In a disordered medium, waves scatter many times.

A key interference pattern is between a path and its time-reversed twin:

• they acquire the same phase
• they interfere constructively for return-to-origin probability
• backscattering is enhanced

Do this once: small correction (weak localization).
Do this strongly enough across many paths: diffusion can collapse (strong localization).

So the system goes from:

• diffusion-like spreading to
• exponentially localized modes with finite localization length (\xi).

A practical mental model

Think in terms of two lengths:

• mean free path (\ell): typical distance between scattering events
• localization length (\xi): size of an eigenstate envelope

If states stay extended at sample scale, you conduct/diffuse.
If (\xi) is small versus system size, transport becomes exponentially suppressed.

A common heuristic near onset is the Ioffe–Regel idea (k\ell \sim 1): when wavelength and scattering scale become comparable, wave transport gets fragile.

Dimensionality really matters

For non-interacting waves with standard uncorrelated disorder:

• 1D: all states localize (in thermodynamic limit)
• 2D: scaling theory predicts no true metallic phase in the simplest symmetry class at (T=0)
• 3D: there can be a genuine transition with a mobility edge (localized below/above a critical energy depending on convention)

Important caveat: symmetry/topology/interactions can change this story (spin–orbit, magnetic field, topological classes, many-body effects).

How to recognize localization experimentally

Typical fingerprints:

1. Exponential spatial tails of mode intensity/probability.
2. Suppressed diffusion: wavepacket spread slows dramatically (not (\propto t) as in normal diffusion).
3. Enhanced coherent backscattering and interference signatures.
4. Transport scaling changes with size/energy/disorder consistent with localization theory.

No single metric is enough in messy systems (absorption, decoherence, finite-size effects can fake or mask signals).

Not the same as every insulator story

• Anderson localization: disorder + interference dominated.
• Mott localization: interaction/correlation dominated (even in cleaner lattices).
• Many-body localization (MBL): interacting disordered quantum systems with constrained thermalization (still actively debated in realistic settings).

Real materials often mix mechanisms; clean attribution is part of the hard science.

Why this is bigger than electrons

Anderson’s original context was electronic transport, but the wave mechanism is broader.

Localization-like physics has been explored in:

• light in disordered photonic media/lattices,
• ultracold atom matter waves in controlled disorder,
• acoustic/elastic waves,
• other complex-wave platforms.

That makes Anderson localization a unifying idea: disorder can be a phase-control knob, not just a nuisance.

Quick “field checklist” for thinking clearly

When someone claims localization, ask:

1. Is transport suppression from interference, not just absorption/loss?
2. Are finite-size and boundary effects ruled out?
3. Is there scaling evidence (with size/energy/disorder), not one pretty plot?
4. What symmetry class and dimensionality are we actually in?
5. Could interactions/dephasing be the real driver?

If these are fuzzy, the claim is likely premature.

References

1. P. W. Anderson (1958), Absence of Diffusion in Certain Random Lattices, Physical Review 109, 1492–1505.
   https://doi.org/10.1103/PhysRev.109.1492

2. E. Abrahams, P. W. Anderson, D. C. Licciardello, T. V. Ramakrishnan (1979), Scaling Theory of Localization: Absence of Quantum Diffusion in Two Dimensions, Physical Review Letters 42, 673–676.
   https://doi.org/10.1103/PhysRevLett.42.673

3. F. Evers, A. D. Mirlin (2008), Anderson transitions, Reviews of Modern Physics 80, 1355–1417.
   https://doi.org/10.1103/RevModPhys.80.1355

4. J. Billy et al. (2008), Direct observation of Anderson localization of matter waves in a controlled disorder, Nature 453, 891–894.
   https://doi.org/10.1038/nature07000

5. G. Roati et al. (2008), Anderson localization of a non-interacting Bose–Einstein condensate, Nature 453, 895–898.
   https://doi.org/10.1038/nature07071

6. A. Lagendijk, B. van Tiggelen, D. S. Wiersma (2009), Fifty years of Anderson localization, Physics Today 62(8), 24–29.
   https://doi.org/10.1063/1.3206091

7. M. Segev, Y. Silberberg, D. N. Christodoulides (2013), Anderson localization of light, Nature Photonics 7, 197–204.
   https://doi.org/10.1038/nphoton.2013.30

8. T. van der Beek et al. (2023), Anderson localization of electromagnetic waves in three dimensions, Nature Physics 19, 1014–1020.
   https://doi.org/10.1038/s41567-023-02091-7

One-line takeaway

Anderson localization is the “interference wins over diffusion” regime: enough disorder can turn transport from a flow problem into a trapping problem.