Fact-checked by Grok 2 weeks ago

Weak localization

Weak localization is a quantum interference effect in disordered electronic systems that enhances electron backscattering through constructive interference between time-reversed diffusion paths, leading to an increase in electrical resistance at low temperatures.^[1] This phenomenon is particularly prominent in two-dimensional conductors, where it manifests as a logarithmic divergence in the resistance as temperature approaches zero, distinct from classical diffusive transport.^[1] In such systems, the correction to conductivity arises from enhanced return probability to the origin, effectively reducing the mean free path and altering the Drude conductivity formula.^[1] The physical mechanism stems from the coherence of electron waves in the presence of time-reversal symmetry, where pairs of electrons propagating along closed loops interfere constructively for backscattering geometries.^[2] This interference is suppressed by factors that break time-reversal invariance, such as magnetic fields, which produce a characteristic positive magnetoresistance signature—a cusp in conductivity at low fields.^[3] In systems with strong spin-orbit coupling, weak anti-localization can dominate instead, enhancing conductivity through destructive interference and a π Berry phase, as seen in topological materials.^[3] Theoretically, weak localization was first understood through scaling arguments in the late 1970s, predicting logarithmic corrections in 2D systems, and formalized via diagrammatic perturbation theory using Cooperon propagators.^[4] Key developments include the 1980 work by Hikami, Larkin, and Nagaoka, which incorporated spin-orbit interactions and derived the magnetoconductivity formula widely used for experimental analysis.^[4] Experimentally, it was observed in thin metal films and inversion layers of MOSFETs in the 1970s, with resistance anomalies scaling logarithmically with temperature and sample size.^[5] Weak localization serves as a sensitive probe for microscopic parameters, including phase coherence length (typically 100 nm to 1 μm at millikelvin temperatures), dephasing rates due to electron-electron or electron-phonon interactions, and spin-orbit scattering times on picosecond scales.^[5] Applications extend to studying quantum transport in nanostructures, topological insulators—where surface states exhibit anti-localization—and magnetic impurities, revealing Kondo effects and Fermi liquid properties in dilute alloys.^[3]^[5] In modern contexts, it aids in characterizing 2D materials like graphene and transition metal dichalcogenides, distinguishing trivial from topological conduction channels.^[3]

Fundamentals

Definition and Overview

Weak localization (WL) is a quantum interference effect that manifests as a positive correction to the classical electrical resistivity in disordered metals or semiconductors at low temperatures. This correction arises from the coherent backscattering of electron waves, where pairs of time-reversed paths interfere constructively, enhancing the probability of electrons returning to their starting point and thereby impeding charge transport. In disordered systems, WL leads to an increase in resistance that scales logarithmically with decreasing temperature or increasing sample size, distinguishing it from the temperature-independent classical Drude conductivity. This logarithmic enhancement serves as a precursor to the stronger Anderson localization, where wavefunctions become fully localized due to disorder, potentially driving the system toward an insulating state. Unlike the Drude model, which treats electrons as classical particles scattered incoherently, WL highlights the wave-like nature of electrons and the role of phase coherence in transport. WL is prominent in the diffusive regime of electron transport, where the mean free path l (set by elastic scattering) is much shorter than the sample dimensions, ensuring multiple scattering events, while the phase coherence length L_\phi (limited by inelastic processes like electron-phonon or electron-electron interactions) remains longer than l, preserving quantum coherence over relevant scales. The characteristic scale of the conductivity correction is given by

\delta \sigma \sim \frac{e^2}{h} \ln \left( \frac{L_\phi}{l} \right),

where e is the electron charge and h is Planck's constant; this quantifies the interference contribution relative to the classical conductivity.

Historical Development

The concept of weak localization emerged as a perturbative quantum correction to classical transport in disordered metallic systems, building on the foundational ideas of Anderson localization introduced in 1958. Philip W. Anderson's seminal work demonstrated that strong disorder could lead to exponential localization of electron wavefunctions in three dimensions, preventing diffusion and metallic conduction, though the effect was non-perturbative and required strong scattering. This laid the groundwork for understanding quantum interference in disordered media, but early experimental anomalies in low-temperature resistivity, such as the logarithmic increase in resistance with decreasing temperature in thin metal films, remained unexplained by classical mechanisms like electron-phonon scattering.^[6] In the late 1970s, theoretical advances distinguished weak localization as a subtle, calculable effect arising from quantum interference in weakly disordered systems. The first explicit proposal came in 1979, when Lev P. Gor'kov, Anatoly I. Larkin, and David E. Khmelnitskii used diagrammatic perturbation theory to compute the quantum correction to conductivity in two-dimensional disordered metals, attributing it to enhanced backscattering due to constructive interference of time-reversed electron paths.^[7] Concurrently, Boris L. Altshuler, Arkady G. Aronov, and David E. Khmelnitskii extended this framework by incorporating electron-electron interactions with small energy transfers, showing how they contribute to the interference and resolve the logarithmic temperature dependence observed in experiments. This perturbative approach, rooted in the Cooperon propagator representing paired electron amplitudes, formalized weak localization as a precursor to stronger Anderson effects, applicable to diffusive regimes where the mean free path exceeds the de Broglie wavelength. Key milestones followed rapidly in 1980, solidifying the theory. Hidetoshi Fukuyama contributed to the understanding of weak localization in two-dimensional disordered systems. Simultaneously, Susumu Hikami, Anatoly I. Larkin, and Yosuke Nagaoka provided a comprehensive diagrammatic calculation for the two-dimensional case, including the effects of magnetic fields and spin-orbit scattering, which predicted a characteristic negative magnetoresistance that suppresses the effect.^[4] These works distinguished weak localization from full Anderson localization by emphasizing its weakness as a logarithmic, rather than exponential, correction, resolvable via perturbation theory in disorder strength. Experimental confirmation arrived soon after, validating the theoretical predictions. In 1980, David J. Bishop, Robert C. Dynes, and Daniel C. Tsui observed the predicted logarithmic temperature dependence and negative magnetoresistance in silicon inversion layers of metal-oxide-semiconductor field-effect transistors (MOSFETs) at millikelvin temperatures, attributing them to weak localization after ruling out classical explanations.^[8] This resolved ongoing debates about the low-temperature resistivity upturn in metals, confirming quantum interference as the underlying cause rather than unresolved scattering processes, and opened the door to using weak localization as a probe of phase coherence in disordered systems.

Theoretical Principles

Quantum Interference Mechanism

Weak localization arises from quantum interference effects in disordered conductors, where electron waves propagating along time-reversed paths interfere constructively, enhancing the probability of backscattering and thereby reducing the overall conductivity. In a disordered medium, electrons undergo multiple elastic scattering events, leading to diffusive motion; however, quantum coherence allows pairs of time-reversed trajectories—paths that retrace each other—to accumulate a phase difference of zero, resulting in constructive interference specifically for backscattered waves. This interference is captured by the maximally crossed diagram in perturbation theory, originally identified as contributing to an anomalous increase in resistivity beyond classical expectations.^[9] In the diagrammatic perturbation theory of electron transport, this interference is formalized through the Cooperon propagator, which describes the propagation of paired electron-hole excitations corresponding to time-reversed paths. The Cooperon, analogous to the particle-particle propagator in superconductivity but in the particle-hole channel, sums the infinite series of diffuson ladders with crossed links, representing coherent backscattering loops. Weak localization manifests as the contribution from the singular long-wavelength (q=0) mode of this Cooperon, which diverges in low dimensions due to recurrent scattering, yielding a perturbative correction to the classical Drude conductivity. This mode is suppressed by inelastic processes that break time-reversal symmetry or phase coherence.^[7] The mathematical basis for the conductivity correction in the weak-disorder limit is derived from the Kubo formula, incorporating the Cooperon contribution to first order in perturbation theory:

\delta \sigma = -\frac{e^2}{\pi h} \ln\left(\frac{\tau_\phi}{\tau}\right),

where e is the electron charge, h is Planck's constant, \tau is the elastic scattering time, and \tau_\phi is the phase relaxation time limiting the coherence of interfering paths. This logarithmic enhancement of localization reflects the diffusive nature of electron motion, with the correction growing as phase coherence lengthens.^[7] The role of phase coherence is central, as the interference effect requires maintenance of the electron wavefunction phase over timescales comparable to \tau_\phi. Dephasing arises from inelastic scattering mechanisms, such as electron-electron or electron-phonon interactions, which introduce random phase shifts and cut off the Cooperon divergence. The effect vanishes when the temperature exceeds the dephasing temperature T_\phi \sim \hbar / (k_B \tau_\phi), where \hbar is the reduced Planck's constant and k_B is Boltzmann's constant, restoring classical diffusive transport.^[10]

Role of Disorder and Electron Diffusion

In disordered metallic systems, non-magnetic impurities introduce elastic scattering that randomizes electron trajectories without flipping their spins, establishing the foundational diffusive motion essential for weak localization effects.^[4] These impurities create a random potential landscape, modeled classically by the Boltzmann transport equation, where electrons undergo repeated scattering events that lead to a mean free path l much shorter than the sample size but still allowing coherent propagation over relevant scales. The diffusive regime arises when electrons perform random walks due to this scattering, characterized by the diffusion constant D = \frac{v_F l}{d}, where v_F is the Fermi velocity, l is the mean free path, and d is the dimensionality of the system.^[4] This classical diffusion sets the stage for quantum corrections, as the multiple scattering paths enable the interference processes underlying weak localization, provided the disorder is weak enough to maintain a metallic state. The Anderson model further describes this disorder through random on-site potentials in a tight-binding lattice, predicting a transition from extended to localized states as disorder strength increases, with weak localization emerging in the weakly disordered limit before full localization occurs. A key parameter is the coherence length L_\phi = \sqrt{D \tau_\phi}, where \tau_\phi is the phase coherence time, which delineates the spatial extent over which quantum phase coherence persists despite inelastic scattering processes like electron-electron or electron-phonon interactions.^[4] The weak localization correction to conductivity scales logarithmically as \ln(L_\phi / l), enhancing backscattering and reducing conductance in low dimensions. This requires the prerequisite of weak disorder, quantified by k_F l \gg 1 (with k_F the Fermi wavevector), ensuring the system remains metallic with diffusive transport via numerous scattering events, yet coherent enough for interference to matter.^[4]

Weak Anti-Localization

Principles of WAL

Weak anti-localization (WAL) is a quantum correction to electrical conductivity in disordered systems, serving as the opposite of weak localization (WL) by producing a positive correction δσ > 0 due to destructive quantum interference in time-reversed electron paths induced by spin-orbit coupling (SOC).^[11] This effect arises when SOC causes a phase shift between paired electron waves propagating along closed loops, reducing the probability of backscattering and thereby enhancing overall conductivity.^[11] The mechanism of WAL stems from the modification of Cooperon contributions by SOC. Without SOC, the spin-singlet Cooperon mode leads to constructive interference, increasing backscattering and yielding the negative conductivity correction of WL. SOC introduces spin precession that randomizes the relative spin orientation along time-reversed paths, suppressing the singlet mode while promoting the three spin-triplet modes; these triplet modes experience destructive interference because the spin rotation accumulates a phase difference of π, effectively reducing coherent backscattering. In systems with strong SOC, where spin-orbit scattering dominates over phase decoherence, the triplet modes prevail, causing WAL to overpower WL.^[11]^[3] A central parameter governing WAL is the spin-orbit scattering time τ_so, which measures the timescale for spin randomization by SOC; WAL emerges prominently when τ_so < τ_φ, with τ_φ denoting the phase coherence time limited by inelastic scattering. In this regime, the WAL conductivity correction takes the form

\delta \sigma = +\frac{e^2}{2\pi h} \ln\left(\frac{\tau_\phi}{\tau_{so}}\right),

reflecting the logarithmic dependence on the ratio of coherence to spin-orbit timescales and establishing the magnitude of the enhancement.^[11] The transition from WL to WAL occurs progressively as SOC strength increases (decreasing τ_so relative to τ_φ and the elastic scattering time τ), shifting the balance from singlet-dominated constructive interference to triplet-dominated destructive interference. This crossover is captured in theoretical models like the Hikami-Larkin-Nagaoka framework through magnetoconductance analysis, where the fitting parameter α quantifies the relative WAL/WL contributions: positive α favors WL, while negative α (often approaching -1 for strong WAL) indicates WAL dominance.^[11]^[12]

Systems Exhibiting WAL

Weak anti-localization (WAL) is prominently observed in heavy metals exhibiting strong spin-orbit coupling (SOC), such as gold (Au) and bismuth (Bi), where the relativistic effects enhance spin precession during electron diffusion, suppressing backscattering and leading to increased conductivity. In thin Au films at the atomic scale, WAL manifests as a distinct positive correction to magnetoconductivity, driven by the material's intrinsic SOC strength, with phase coherence lengths exceeding 100 nm at cryogenic temperatures. Similarly, in epitaxial Bi quantum films, WAL signatures appear in surface states alongside weak localization in bulk quantum wells, highlighting the role of topological protection in low-dimensional Bi structures.^[13] Two-dimensional systems, particularly the surface states of topological insulators like Bi₂Se₃, provide ideal platforms for WAL due to their helical spin texture, which amplifies SOC effects and protects against localization.^[14] In Bi₂Se₃ thin films, WAL is evident in the magnetotransport of decoupled top and bottom surfaces, with the effect persisting up to thicknesses of several quintuple layers under low magnetic fields.^[14] These systems demonstrate WAL when bulk contributions are minimized, allowing surface-state dominance. Observation of WAL requires specific conditions to ensure SOC outpaces dephasing mechanisms: low disorder levels, high electron mobility typically exceeding 1000 cm²/V·s, cryogenic temperatures below 10 K, and the absence of magnetic impurities that could break time-reversal symmetry.^[15] Under these conditions, the phase coherence length surpasses the mean free path, enabling quantum interference to dominate transport. Key observational signatures include negative magnetoconductivity at low magnetic fields (B < 0.1 T), reflecting the suppression of destructive interference by a perpendicular field, and a logarithmic decrease in resistance with temperature, R ∝ -ln(T), arising from the temperature-dependent dephasing rate.^[16] Historically, WAL was first observed in the 1980s in narrow-gap semiconductors like InSb thin films, where strong SOC in the conduction band led to positive magnetoresistance peaks indicative of 2D WAL at interfaces.^[17] In modern contexts, proximity-induced SOC has enabled WAL in graphene heterostructures, such as graphene/WSe₂ interfaces, where a transition from weak localization to WAL signals enhanced spin splitting, with coherence lengths up to microns at millikelvin temperatures.^[18]

Dimensional Dependence

Effects in Two Dimensions

In two dimensions, weak localization exhibits a marginal dimensionality where the quantum interference correction to the conductivity diverges logarithmically with decreasing temperature or increasing system size, leading to an eventual transition to strong localization for any nonzero disorder strength. The conductivity correction is given by \delta \sigma^{2D} = -\frac{e^2}{2\pi^2 \hbar} \ln\left(\frac{\tau_\phi}{\tau}\right), where e is the electron charge, \hbar is the reduced Planck's constant, \tau is the elastic scattering time, and \tau_\phi is the phase coherence time. This logarithmic divergence arises from the diffusive nature of electron motion in disordered 2D systems, where the return probability to the starting point is enhanced by constructive interference of time-reversed paths, reducing the overall conductivity. Unlike in three dimensions, where corrections are finite, the 2D case implies that all states are localized at zero temperature according to scaling theory, though weak localization effects are observable perturbatively at finite but low temperatures. The magnetoconductivity in 2D systems, which suppresses weak localization by breaking time-reversal symmetry, is described by the Hikami-Larkin-Nagaoka (HLN) equation: \Delta \sigma(B) = \frac{\alpha e^2}{2\pi^2 \hbar} \left[ \psi\left(\frac{1}{2} + \frac{B_\phi}{B}\right) - \psi\left(\frac{1}{2} + \frac{B_{so}}{B}\right) - \ln\left(\frac{B_{so}}{B_\phi}\right) \right], where \psi is the digamma function, B is the applied magnetic field, B_\phi = \hbar / (4 e L_\phi^2) with L_\phi = \sqrt{D \tau_\phi} the phase coherence length and D the diffusion constant, and B_{so} = \hbar / (4 e l_{so}^2) with l_{so} the spin-orbit scattering length. The prefactor \alpha distinguishes between weak localization and weak anti-localization: \alpha = 1 for the orthogonal ensemble (weak localization dominant in systems without significant spin-orbit coupling) and \alpha = -1/2 for the symplectic ensemble (weak anti-localization in strong spin-orbit coupling systems). This equation captures the crossover from positive magnetoconductivity at low fields (suppression of localization) to linear-in-B behavior at high fields.^[4] Experimentally, weak localization effects in two dimensions are prominently observed in thin metallic films, where the logarithmic temperature dependence of resistance and negative magnetoresistance align with theoretical predictions. Similarly, in high-mobility two-dimensional electron gases (2DEGs) confined in GaAs/AlGaAs heterostructures, low-temperature magnetotransport measurements reveal characteristic cusps in conductivity versus magnetic field, allowing extraction of \tau_\phi and confirmation of 2D diffusive transport. These systems provide clean platforms to study the competition between localization and dephasing mechanisms.

Behavior in Three Dimensions

In three dimensions, weak localization gives rise to a finite, perturbative correction to the classical Drude conductivity, without the divergent logarithmic enhancement seen in two dimensions. This correction arises from quantum interference of electron waves along time-reversed paths, leading to enhanced backscattering and a reduction in conductivity on the order of \delta \sigma^{3D} \approx -\frac{e^2}{2\pi^2 \hbar} \frac{1}{l}, where l is the elastic mean free path; a weaker temperature-dependent term \propto 1/L_\phi (with L_\phi the phase coherence length) provides the primary observable signature at low temperatures, scaling as a power law \delta \sigma^{3D} \sim -T^{1/2} assuming \tau_\phi \propto T^{-1}. The relative magnitude \delta \sigma / \sigma \sim (k_F l)^{-2} remains small in the weakly disordered metallic regime (k_F l \gg 1), ensuring no transition to an insulating state. Theoretically, three dimensions lie above the lower critical dimension d=2 for Anderson localization in the orthogonal ensemble, where weak localization acts as a minor perturbation rather than driving instability. Renormalization group analysis reveals a \beta function \beta(g) = (d-2) - \frac{c}{g} (with g the dimensionless conductance and c > 0 a model-dependent constant), yielding \beta(g) \approx 1 - c/g > 0 for large g in d=3; this implies a stable metallic fixed point with only weak logarithmic flow toward lower conductance, and localization confined to a mobility edge in strongly disordered systems. Experimental observations of three-dimensional weak localization occur in bulk disordered metals such as gold (Au) and palladium (Pd), where samples maintain a thickness exceeding L_\phi (typically 10–100 nm at millikelvin temperatures) to preserve bulk behavior, as opposed to thinner films that cross over to two-dimensional effects. In such systems, magnetoresistance measurements reveal a positive, sublinear field dependence consistent with suppression of the interference by orbital dephasing, with the effect most prominent in high-purity samples at low temperatures (T \lesssim 1 K). The observability of weak localization in three dimensions requires L_\phi to exceed the sample size in propagation directions, ensuring coherent diffusion over relevant scales; however, inelastic scattering (e.g., from electron-phonon interactions) limits L_\phi, rendering the effect inherently weaker and less divergent than in lower dimensions, often necessitating ultralow temperatures and minimal disorder for detection.

Experimental Probes

Magnetic Field Dependence

The application of a magnetic field suppresses weak localization through its orbital effect, which breaks time-reversal symmetry by introducing an Aharonov-Bohm phase shift in the electron wave functions propagating along time-reversed paths. This phase shift arises from the magnetic flux threading the closed diffusive loops of area on the order of L_\phi^2, where the flux quantum is \Phi_0 = h/(2e); when the flux through such loops reaches approximately \Phi_0, the constructive interference responsible for enhanced backscattering is disrupted, reducing the weak localization correction to conductivity. The characteristic magnetic field scale for this suppression is B_c \sim \hbar / (4 e L_\phi^2), below which the weak localization effect is prominent and above which it saturates, unmasking the underlying classical positive magnetoresistance due to Lorentz deflection of carriers. For B \gg B_c, the quantum interference is fully quenched, and the magnetoresistance transitions from the quantum regime to the classical Drude-like behavior.^[19] Experimentally, this manifests as negative magnetoresistance \Delta \rho(B) < 0 at low fields (B \lesssim B_c), reflecting the suppression of the weak localization-induced increase in resistivity, followed by a crossover to positive magnetoresistance at higher fields from orbital deflection effects. In contrast, for weak anti-localization, the low-field regime shows negative magnetoconductivity due to the suppression of the Cooperon channels contributing to the conductivity enhancement, reflecting the suppression of the weak anti-localization-induced decrease in resistivity.^[19]

Temperature and Disorder Effects

The temperature dependence of weak localization arises primarily from dephasing processes that destroy the phase coherence of electron wavefunctions, limiting the size of interfering loops to the coherence length L_\phi = \sqrt{D \tau_\phi}, where D is the diffusion constant and \tau_\phi is the dephasing time. In two-dimensional disordered metals at low temperatures, electron-electron interactions dominate dephasing, yielding a dephasing rate $1/\tau_\phi \propto T. This linear temperature scaling leads to L_\phi \propto T^{-1/2}, and thus the weak localization correction to the conductivity takes the form \delta \sigma \propto -\ln T, manifesting as a logarithmic increase in resistance at low temperatures. At higher temperatures, electron-phonon scattering contributes significantly to dephasing, with $1/\tau_\phi \propto T^p where p typically ranges from 1 to 3 depending on the phonon spectrum and material. These thermal effects suppress weak localization by reducing L_\phi, eventually rendering the quantum interference negligible when L_\phi becomes comparable to the mean free path l. In diffusive regimes, this temperature modulation allows probing of the underlying electron dynamics without external fields. Disorder strength, quantified by the elastic mean free path l, tunes the extent of multiple scattering paths essential for weak localization. In the weakly disordered limit where k_F l \gg 1 (with k_F the Fermi wavevector), enhanced backscattering probabilities amplify the interference correction, increasing its magnitude logarithmically with system size. Stronger disorder, corresponding to smaller l, initially boosts the weak localization effect by promoting more closed loops, but simultaneously shortens \tau_\phi through increased inelastic scattering, thereby reducing overall coherence. When k_F l \approx 1, the Ioffe-Regel criterion is met, marking a crossover to strong Anderson localization where extended states give way to exponentially localized ones. Experimentally, temperature and disorder effects on weak localization are investigated through resistance measurements as a function of temperature in thin metallic films or two-dimensional electron gases, revealing characteristic low-temperature upturns attributed to \delta \sigma \propto -\ln T. Disorder is systematically varied by controlling film thickness, which increases surface scattering and reduces l as thickness decreases below 100 nm, or by doping to introduce impurity scattering.^[20] In systems exhibiting weak anti-localization due to strong spin-orbit coupling, dephasing similarly limits the interference range with a comparable temperature dependence, though the spin-orbit interaction itself remains largely unaffected by thermal fluctuations. This makes weak anti-localization more robust at intermediate temperatures compared to pure weak localization, but still subject to eventual suppression by enhanced $1/\tau_\phi.

Modern Developments

Applications in Low-Dimensional Materials

In graphene and other two-dimensional materials, weak localization or anti-localization effects arise from quantum interference influenced by valley-dependent spin-orbit coupling (SOC) and the chiral nature of Dirac fermions. Early experiments on graphene flakes in 2008 demonstrated weak localization, manifesting as negative magnetoconductance at low fields due to intervalley scattering suppressing the expected anti-localization.^[21] In subsequent high-mobility suspended graphene devices (post-2008), weak anti-localization (WAL) was observed with positive magnetoconductance, allowing probing of the π Berry phase inherent to graphene's band structure when intervalley scattering is minimized. For instance, in clean suspended graphene samples, WAL corrections were measured with phase coherence lengths exceeding 1 μm at millikelvin temperatures, confirming topological protection against backscattering and enabling studies of intervalley scattering rates.^[22] These measurements have since been extended to probe topological phase transitions in graphene heterostructures, where WAL suppression signals the crossover from trivial to nontrivial band topology. In topological insulators such as Bi₂Te₃, WAL on the surface states serves as a hallmark of protected helical edge modes, distinguishing two-dimensional surface conduction from bulk contributions.^[23] Studies on thin Bi₂Te₃ films in the early 2010s revealed a characteristic cusp in magnetoresistance at low fields (B < 0.1 T), fitted to the Hikami-Larkin-Nagaoka model with a phase coherence length of ~500 nm, indicating dominant surface-state transport even in samples with residual bulk doping.^[23] This WAL effect persists up to ~10 K, reflecting the robustness of spin-momentum locking against disorder, and has been used to quantify the separation of surface and bulk conductivities by analyzing field-angle dependence.^[24] In Bi₂Te₃ microflakes, WAL amplitude scales inversely with thickness below 100 nm, providing a direct probe of topological surface states and their suppression by magnetic impurities.^[24] In quasi-one-dimensional semiconductor nanowires, such as InAs, weak localization exhibits anisotropic suppression under parallel magnetic fields, where the flux through diffusive loops (confined to the wire cross-section) vanishes, leading to negligible interference correction.^[25] Magnetoconductance measurements on InAs nanowires with diameters ~50 nm show a sharp negative correction at zero field, evolving to positive under perpendicular fields (B ⊥ wire axis), with dephasing times τ_φ ~ 10 ns at 100 mK, highlighting the role of one-dimensional confinement in enhancing phase coherence.^[25] This harmonic suppression by parallel fields (B ∥ wire) has been exploited to isolate spin-orbit effects, as the WAL-to-WL crossover depends on Rashba coupling strength, tuned via gate voltage in hybrid InAs structures.^[26] Recent advancements in high-mobility two-dimensional electron gases (2DEGs), such as those in InSb quantum wells, have revealed interplay between weak localization and quantum Hall states, where disorder-induced corrections modulate plateau widths at filling factors ν ~ 1–5. In 2023 experiments on gate-defined 2DEGs with mobilities exceeding 10⁵ cm²/Vs, WAL signatures at low fields (B < 0.05 T) coexist with Shubnikov-de Haas oscillations, allowing extraction of localization lengths ~2 μm and demonstrating how elevated temperatures (up to 1 K) suppress interference without fully quenching Hall quantization. These studies in 2020s-era devices underscore the role of WL in broadening quantum Hall transitions, providing benchmarks for disorder control in topological 2DEG platforms.

Relevance to Quantum Phenomena

Weak localization manifests as the perturbative precursor to Anderson localization in disordered quantum systems, where quantum interference of electron waves enhances backscattering probability, leading to a logarithmic correction to conductivity that signals the onset of insulating behavior. This effect provides critical insight into metal-insulator transitions, as described by the scaling theory of localization, by illustrating how weak disorder gradually localizes states and reduces conductance at low temperatures without fully insulating the system.^[27] Theoretical extensions beyond the non-interacting framework incorporate electron-electron interactions through Altshuler-Aronov corrections, which introduce additional contributions to the conductivity via exchange and Fock processes that modify the cooperon propagator responsible for interference. These corrections are vital for accurately modeling transport in real materials where interactions compete with disorder effects. Post-2010 developments have further linked weak localization to many-body localization, an interacting counterpart where disorder halts thermalization and preserves initial quantum states, extending single-particle interference ideas to ergodicity-breaking phenomena in isolated quantum systems. In topological insulators, weak antilocalization emerges as a hallmark of Z₂ topology, driven by spin-orbit coupling that locks spin to momentum and destructively interferes backscattered paths, thereby enhancing surface-state conductivity and distinguishing topological protection from trivial disorder effects.^[3] Extensions in the 2020s have applied this signature to higher-order topological insulators, where weak antilocalization probes protected states on lower-dimensional boundaries like hinges, revealing robust quantum transport amid strong disorder.^[28] Dephasing studies using weak localization in superconducting nanowires directly inform qubit coherence times, as the phase-breaking rate derived from magnetoconductance corrections quantifies decoherence sources like electron-phonon interactions or flux noise that limit quantum gate fidelity. For example, in systems designed for topological quantum computing, such as InSb nanowires hosting Majorana modes, weak localization analysis helps optimize disorder levels to extend dephasing times beyond microseconds, enhancing prospects for scalable qubit arrays.^[29]

References

[1]
[PDF] Weak Localization - UF Physics Department
Feb 27, 2018 · The difference is due to constructive interference between the two paths. This is the basic idea behind weak localization. I. DIFFUSION EQUATION.
[2]
[PDF] Weak localization and weak anti-localization in topological insulators
Sep 4, 2014 · In this article, we review recent progresses in both theory and experiment of weak (anti-)localization in topological insulators, where the ...
[3]
Weak localization and weak anti-localization in topological insulators
Sep 4, 2014 · Weak localization and weak anti-localization are quantum interference effects in quantum transport in a disordered electron system.
[4]
Spin-Orbit Interaction and Magnetoresistance in the Two ...
Spin-Orbit Interaction and Magnetoresistance in the Two Dimensional Random System Free ... Hikami and J. Zinn-Justin, Nucl. Phys. B (to be published.) 10 ...
[5]
International Journal of Modern Physics B
### Summary of Weak Localization Review
[6]
http://www.jetpletters.ru/ps/1364/article_20629.pdf
[7]
[PDF] LP Gor'kov, AI Larkin, and DE Khmelnitskii - JETP Letters
The connection between this result and the conclusions of Ref. 1 about the total localization in the two- dimensional case is discussed. PACS numbers: 72.10.Bg.Missing: weak | Show results with:weak
[8]
https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.44.1153
[9]
Breakdown of the Concentration Expansion for the Impurity ...
Phys. Rev. Lett. 16, 984 (1966) ... Langer, Phys. Rev. 120, 714 (1960) 124, 1003 (1961); M. L. Glasser, Phys. Rev. 129, 472 (1963). Outline Information.
[10]
https://arxiv.org/pdf/1409.1299.pdf
[11]
https://doi.org/10.1143/PTP.63.707
[12]
Conductivity Corrections for Topological Insulators with Spin-Orbit ...
Jul 24, 2015 · It predicts a crossover from weak localization to antilocalization as a function of the strength of scattering off spin-orbit impurities.
[13]
Magnetoconductance in epitaxial bismuth quantum films
Aug 17, 2021 · Furthermore, quantum corrections described by weak antilocalization (WAL) for the surface states, and by weak localization (WL) for quantum well ...Abstract · Article Text · INTRODUCTION · RESULTS
[14]
Coherent topological transport on the surface of Bi 2 Se 3 - Nature
Jun 26, 2013 · For thick samples, the magnitude of weak antilocalization indicates two decoupled (top and bottom) symplectic surfaces.Missing: heavy | Show results with:heavy
[15]
Weak antilocalization in Cd3As2 thin films | Scientific Reports - Nature
Mar 3, 2016 · We report on the low-temperature magnetoresistance measurements on a ~50 nm-thick Cd 3 As 2 film. The weak antilocalization under perpendicular magnetic field ...Missing: heavy | Show results with:heavy
[16]
Signature of weak-antilocalization in sputtered topological insulator ...
Jun 13, 2022 · We report a low-temperature magneto transport study of Bi 2 Se 3 thin films of different thicknesses (40, 80 and 160 nm), deposited on sapphire (0001) ...Missing: heavy | Show results with:heavy
[17]
Effect of hetero‐interface on weak localization in InSb thin film layers
May 4, 2005 · Effect of hetero-interface on weak localization in InSb thin film layers ... The positive MR arising from the weak anti-localization (WAL) found ...<|separator|>
[18]
Spin–orbit proximity effect in graphene | Nature Communications
Sep 26, 2014 · This transition from weak localization to weak anti-localization behaviour provides independent evidence for the proximity effect-induced SOC.
[19]
https://doi.org/10.1155/2014/737296
[20]
[PDF] WEAK LOCALIZATION IN THIN FILMS a time-of-flight experiment ...
In this review article the physics of weak localization is discussed. It represents an interference experiment with conduction electrons split into pairs of ...
[21]
Impurity Effect on Weak Antilocalization in the Topological Insulator
Apr 21, 2011 · We study the weak antilocalization (WAL) effect in topological insulator B i 2 ⁢ T e 3 thin films at low temperatures.
[22]
Weak antilocalization in topological insulator Bi 2 Te 3 microflakes
Jan 16, 2013 · In particular, positive magnetoresistances (MRs) in the two-dimensional weak-antilocalization (WAL) effect were measured in low magnetic fields, ...Abstract · Article Text · INTRODUCTION · RESULTS AND DISCUSSIONMissing: anti- | Show results with:anti-
[23]
Anisotropic magnetoconductance of a InAs nanowire: Angle ...
The magnetoconductance of an InAs nanowire is investigated with respect to the relative orientation between external magnetic field and the nanowire axis.
[24]
Phase-coherent transport and spin relaxation in InAs nanowires ...
Apr 28, 2015 · At low magnetic fields, the magnetoconductance characteristics exhibit a crossover between weak antilocalization and weak localization by ...
[25]
Universal mechanism for Anderson and weak localization - PNAS
In this paper, we demonstrate that both Anderson and weak localizations originate from the same universal mechanism, acting on any type of vibration, in any ...
[26]
Weak antilocalization in the transition metal telluride | Phys. Rev. B
Dec 19, 2023 · Low-temperature magnetotransport shows that T a 2 ⁢ P d 3 ⁢ T e 5 exhibits weak antilocatization (WAL) effect in both perpendicular orientation and parallel ...Missing: seminal | Show results with:seminal
[27]
Electronic Transport and Quantum Phenomena in Nanowires
This destructive interference is referred to as weak antilocalization. An applied magnetic field destroys this weak antilocalization effect, resulting in a ...