Field-Induced Interface Pinning

• Field-induced interface pinning is the phenomenon where an external field stabilizes or localizes interfacial regions by modulating inherent chemical, structural, or disorder-induced biases.
• Research shows that applied fields can trigger localization transitions, modify tunneling electroresistance, and dynamically arrest domain wall motion in varied materials.
• The concept spans disciplines—from ferroelectric systems and disordered lattices to active matter and magnetic interfaces—offering practical insights for controlling interfacial stability and transport.

Field-induced interface pinning denotes a family of phenomena in which an applied field, a field-defined boundary condition, or a field-coupled bias stabilizes, localizes, or arrests an interface, interfacial polarization, domain wall, or contact region. Across the literature, the phrase does not refer to a single microscopic mechanism. In ferroelectric tunnel junctions, a pinned interface dipole is chemically or structurally locked and the applied field only switches the ferroelectric barrier and the opposite interface; in disordered lattice free fields, a pinning field directly rewards contacts with a substrate and drives a localization transition; in active spin systems, a spatial reversal of the external field traps flocks at an interface and arrests transport (Wu, 2014, Giacomin et al., 2019, Karmakar et al., 26 Mar 2026).

1. Field-induced pinning as a cross-disciplinary concept

A recurring structure is the competition between a driving force and an interfacial energy landscape. In some systems the field creates the localization mechanism directly. The disordered lattice Gaussian free field uses an external pinning field $h$ through the Hamiltonian

$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$

with $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$, so localization occurs when the reward for contacts exceeds the entropic cost of keeping the interface near the substrate (Giacomin et al., 2019). In other systems the field reveals or modulates an already existing pinned structure. In ferroelectric tunnel junctions, the pinned interface dipole is assumed fixed by interfacial bonding, chemistry, atomic termination, and structural asymmetry; the applied field reshapes screening and depolarization fields but does not reorient the pinned dipole (Wu, 2014).

This distinction removes a common ambiguity. “Field-induced” does not always mean that the field creates the pinning center. The graphene contact literature makes this especially explicit: charge-density pinning at a metal/graphene interface is a field-suppressed condition caused by screening from the metal and spill-out electrons, whereas depinning occurs when an interfacial oxide weakens screening and allows the back-gate field to tune the carrier density under the contact (Nouchi et al., 2010). Likewise, high-field Fowler–Nordheim tunneling through h-BN does not create Fermi level pinning at the h-BN/metal interface; rather, the high-field regime exposes the pinned alignment and enables extraction of the pinning factor (Hattori et al., 2018).

Taken together, these works suggest that field-induced interface pinning is best understood as a field-coupled control problem. The interface may be pinned by chemistry, quenched disorder, exchange anisotropy, anisotropy gradients, lattice periodicity, or screening asymmetry, while the field determines whether that pinning is activated, overcome, or made experimentally visible.

2. Ferroelectric interfaces: pinned dipoles and microstructure-specific wall pinning

In ferroelectric tunnel junctions with asymmetric interfaces, one interface dipole can be switchable and collinear with the ferroelectric barrier polarization, while the opposite interface dipole is pinned and nonswitchable. The electrostatics are set by displacement continuity,

$$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$

with depolarization fields

$$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$

and tunneling electroresistance defined as

$\mathrm{TER}=(G_R-G_L)/G_L.$

The key result is that a nonswitchable interface dipole is sufficient to induce nonzero TER even with symmetric electrodes, because one interface switches and the other does not. The paper reports that TER increases monotonically with the magnitude of the switchable interfacial polarization $|P_{iL}|$, whereas TER is comparatively insensitive to the magnitude of the pinned dipole $P_{iR}$ for fixed direction. Orientation matters strongly: when the pinned dipole points toward the ferroelectric film, large TER can be achieved, and lower interface dielectric constants substantially enhance TER, with decreasing $\epsilon_{iR}$ increasing TER by about three orders of magnitude in the large-$|P_{iR}|$ case (Wu, 2014). A related misconception is explicitly excluded in this model: the pinning is intrinsic, not field-created.

Field-driven pinning also appears at ferroelectric and ferroelastic domain walls, but here the local microstructure sets the pinning potential. In epitaxial PbTiO$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$0 on cubic KTaO$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$1, automated Piezoresponse Force Microscopy was used to analyze about 1500 switching events and quantify wall displacement with the metric $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$2, the mean absolute difference expressed as a percentage of a $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$3 patch area. Using $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$4 as a practical pinning criterion, the study found two mobility fingerprints. Uniform ferroelastic structures (Classes I–III) are essentially static at $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$5, show single-step activation at $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$6 with $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$7–$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$8, and increase further to $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$9–$\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$0 at $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$1. Heterogeneous boundaries (Classes IV–V) remain largely static at $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$2 and $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$3, then depin above about $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$4 and exceed $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$5 at $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$6 (Barakati et al., 29 May 2025).

The most frequently contrasted geometries are the single-variant boundary $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$7 (Class I) and the polydomain twin boundary $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$8 (Class IV). The former is already activated at $\delta_x=\mathbf{1}_{[-1,1]}(\phi(x))$9, whereas the latter stays pinned up to a certain level of bias magnitude and changes only marginally as the bias increases from $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$0 to $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$1 until the higher threshold is exceeded (Barakati et al., 29 May 2025). Orientation dependence persists even at high field: displacement maxima occur near about $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$2 and $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$3, while pronounced minima occur near about $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$4, $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$5, and $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$6. This supports a microstructure-specific picture in which elastic incompatibility, electrostatic mismatch, and local defect structure determine whether the applied electric field produces coherent motion, creep, or effective pinning.

3. Disordered lattice fields: localization transitions near criticality

In the lattice Gaussian free field and related disordered free-field models, field-induced interface pinning is a genuine localization transition of an effective interface near a substrate. The field acts directly through contact rewards. In the renormalized convention used by Giacomin and Lacoin, the disordered pinned measure contains the weight

$$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$7

with $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$8. For $$D=\sigma_s=\epsilon_{iL}E_{iL}+P_{iL}=\epsilon_fE_f+P_f=\epsilon_{iR}E_{iR}+P_{iR},$$9, the quenched critical point equals the annealed or pure value, $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$0, but the critical behavior differs sharply: the pure model has linear free energy near criticality, while the quenched free energy is quadratically small, $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$1 for small $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$2, and for Gaussian disorder this quadratic behavior is sharp (Giacomin et al., 2015).

The later criticality analysis makes this statement exact in the unrenormalized convention. Writing $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$3 with

$$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$4

the free energy obeys

$$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$5

This identifies a second-order transition with an explicit disorder-dependent prefactor (Giacomin et al., 2019). The contact fraction is asymptotically linear in $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$6, and the typical field magnitude is

$$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$7

on all but a vanishing fraction of sites as $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$8 (Giacomin et al., 2019).

These results clarify an important point about disorder relevance. The critical point may remain unchanged while the critical exponent changes. In $$E_{iL}^d=(\sigma_s-P_{iL})/\epsilon_{iL},\quad E_f^d=(\sigma_s-P_f)/\epsilon_f,\quad E_{iR}^d=(\sigma_s-P_{iR})/\epsilon_{iR},$$9, the quenched and annealed thresholds coincide in the renormalized convention, yet disorder rounds the transition from first order in the pure model to quadratic in the quenched model (Giacomin et al., 2015). The mechanism is the balance between entropic repulsion and sparse energetic gains at favorable disorder sites. The upper bound

$\mathrm{TER}=(G_R-G_L)/G_L.$0

and the expansion

$\mathrm{TER}=(G_R-G_L)/G_L.$1

show how the quadratic free-energy law emerges from the second-order Jensen gap rather than from a simple extensive contact density (Giacomin et al., 2019).

4. Active matter: field-reversal boundaries and arrested transport

In the 4-state Active Potts Model, field-induced interface pinning arises when a weak symmetry-breaking field reorganizes coexistence and interfacial transport in a nonequilibrium flocking system. Under a homogeneous unidirectional field, the standard coexistence between an apolar gas and a polar liquid is replaced by phase separation between two field-aligned polar phases: a low-density, weakly polarized background and a high-density, strongly polarized band, both moving along the field. When the field is transverse to a longitudinal lane, the lane executes treadmilling motion against the field, with

$\mathrm{TER}=(G_R-G_L)/G_L.$2

and $\mathrm{TER}=(G_R-G_L)/G_L.$3 for small $\mathrm{TER}=(G_R-G_L)/G_L.$4 (Karmakar et al., 26 Mar 2026).

The clearest realization of interface pinning occurs when the field points in opposite directions in two halves of the system. Then a field-reversal interface forms, and three regimes appear as $\mathrm{TER}=(G_R-G_L)/G_L.$5 increases: bidirectional flocking at weak field, an oscillatory interfacial state at intermediate field, and field-induced interface pinning for $\mathrm{TER}=(G_R-G_L)/G_L.$6 at strong field. In the pinned regime, particles accumulate at the boundary, the net interface-normal flux vanishes, and motion is confined to the interfacial columns with back-and-forth oscillations (Karmakar et al., 26 Mar 2026).

The mechanism is formulated as a flux balance,

$\mathrm{TER}=(G_R-G_L)/G_L.$7

where $\mathrm{TER}=(G_R-G_L)/G_L.$8 is the interface drift velocity, $\mathrm{TER}=(G_R-G_L)/G_L.$9 is the advective current across the interface, and $|P_{iL}|$0 is the field-biased return flux produced by reorientation. Pinning occurs when $|P_{iL}|$1, the reorientation length $|P_{iL}|$2 decreases to about one lattice spacing, and transport across the interface is arrested (Karmakar et al., 26 Mar 2026). Microscopically, $|P_{iL}|$3 increases with $|P_{iL}|$4, decreases with $|P_{iL}|$5, and remains essentially independent of $|P_{iL}|$6 at fixed control parameters.

The same framework extends to quenched random field orientations. There, local field reversals generate many FIIP-like regions, global order decreases algebraically with system size, and sharp first-order signatures are smoothed in a manner consistent with Imry–Ma arguments and with the Aizenman–Wehr theorem in the diffusive limit (Karmakar et al., 26 Mar 2026). This is a useful corrective to a common interpretive error: interfacial trapping and Hallmark coexistence anomalies in active matter need not indicate static obstacles or equilibrium wetting, because the pinning can be generated dynamically by field-controlled conversion at the interface.

5. Magnetic and crystalline interfaces: anisotropy, exchange, dynamics, and lattice coupling

A canonical analytical pinning problem is the domain wall at a locally engineered anisotropy step in Pt/Co/Pt strips with perpendicular magnetic anisotropy. For an anisotropy change from $|P_{iL}|$7 to $|P_{iL}|$8 across a ramp of width $|P_{iL}|$9, the depinning field is

$P_{iR}$0

which reduces to $P_{iR}$1 for a sharp step and tends to zero for $P_{iR}$2. The barrier therefore depends not only on the anisotropy contrast $P_{iR}$3 but also on the ratio of barrier width to domain-wall width (Franken et al., 2011). Experimentally, Ga focused ion beam irradiation tunes both the anisotropy difference and the barrier width, and the boundary of the irradiated area acts as a pinning barrier whose strength increases with the anisotropy difference and decreases as the anisotropy gradient broadens (Franken et al., 2011).

Field-induced pinning also appears dynamically. In Permalloy nanowires, the transmission probability of a domain wall does not increase monotonically with external field. Instead, it rises above the depinning threshold, peaks, then drops after a higher threshold associated with Walker breakdown, and recovers at still larger fields. For the $P_{iR}$4 wire, the reported values include $P_{iR}$5 at $P_{iR}$6, $P_{iR}$7 at $P_{iR}$8, a peak around $P_{iR}$9, and a pronounced minimum around $\epsilon_{iR}$0 (0808.3446). Micromagnetic simulations attribute the drop to dynamical pinning: above the Walker threshold, vortex-core switching produces transverse motion and temporary slow-downs, increasing the probability of capture by edge roughness.

At buried ferromagnet/antiferromagnet interfaces, interfacial exchange anisotropy supplies a different pinning mechanism. In Ni$\epsilon_{iR}$1Fe$\epsilon_{iR}$2/Ir$\epsilon_{iR}$3Mn$\epsilon_{iR}$4 bilayers, the standing spin-wave thickness mode is more sensitive to interface pinning than the uniform FMR mode, and the field difference for the lowest standing mode is about twice that of the uniform resonance. The interfacial pinning parameter $\epsilon_{iR}$5 is extracted together with an effective ferromagnetic thickness $\epsilon_{iR}$6, and the increase of $\epsilon_{iR}$7 with IrMn thickness and with cooling is accompanied by a decrease of $\epsilon_{iR}$8, suggesting deformation of magnetic order near the interface (Magaraggia et al., 2010).

Interfacial coupling can also impose strong anisotropic pinning in oxide superlattices. In SrRuO$\epsilon_{iR}$9/LaCoO$|P_{iR}|$0 superlattices, out-of-plane rather than in-plane magnetic fields induce stripe-like magnetic domains, and no stripes are observed for in-plane fields up to $|P_{iR}|$1. Stripes appear around $|P_{iR}|$2 at $|P_{iR}|$3 and around $|P_{iR}|$4 at $|P_{iR}|$5, weaken above about $|P_{iR}|$6–$|P_{iR}|$7, and reappear on the return branch. No skyrmions are observed up to $|P_{iR}|$8, and the work argues that interfacial exchange overwhelms Dzyaloshinskii–Moriya interaction, producing topological-like transport anomalies without true topological spin textures (Jiang et al., 25 Feb 2026).

A more formal continuum realization appears in the phase-field-crystal model. Nonadiabatic coupling between slow liquid–solid envelopes and the underlying lattice yields a generalized Gibbs–Thomson relation,

$|P_{iR}|$9

which reduces in the small-slope limit to a driven sine-Gordon equation with KPZ nonlinearity. The periodic pinning term $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$00 is the continuum expression of lattice-induced front locking, and depinning occurs when $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$01 (Huang, 2012). This formulation is notable because the pinning potential is not imposed phenomenologically; it is derived from the coupling between mesoscopic amplitudes and microscopic crystalline periodicity.

6. Electronic interfaces, Wigner solids, and computational pinning formalisms

Electronic interfaces provide two distinct uses of pinning. In graphene field-effect transistors, charge-density pinning refers to the inability of the back gate to tune the carrier density beneath metal contacts because the gate field is screened by the metal. In the pinned case, the transfer curve remains roughly V-shaped with mild electron–hole asymmetry; in the depinned case, produced here by oxidation of easily oxidizable ferromagnetic metals such as Ni and Co, the gate field penetrates the contact region and the transfer characteristic develops strong distortion, kinks, or apparent double minima. The paper reports that annealing at $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$02 in Ar for $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$03 removes distortion in Ni-contacted devices, while exposure to air for $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$04 increases distortion in a Co-contacted device (Nouchi et al., 2010). This establishes contact oxidation as a route from a field-suppressed pinned interface to a gate-tunable depinned one.

High-field tunneling through h-BN/metal contacts reveals a different kind of pinning: Fermi level pinning at the interface. Systematic Fowler–Nordheim analysis across Pd, Au, Cr, and Ti shows that the electron barrier to the h-BN conduction band minimum clusters near about $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$05 and depends only weakly on metal work function. The extracted relation is

$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$06

with pinning factor $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$07, charge neutrality level $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$08, and electron affinity $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$09. The same experiments show that Fowler–Nordheim tunneling through h-BN is hole tunneling rather than electron tunneling (Hattori et al., 2018). Again, the field does not create the pinning; it moves the device into a transport regime where the pinned alignment becomes measurable.

In magnetic-field-induced Wigner solids, the field-induced aspect lies in the formation of the solid itself at low filling factor. The pinned solid exhibits a collective resonance whose frequency $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$10 measures the disorder that pins it. Across ultralow-disorder GaAs quantum wells of widths $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$11, $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$12, $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$13, and $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$14, $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$15 decreases strongly with increasing $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$16, reaching about $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$17 in the $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$18 well. The measured trend is explained by wave-function tails penetrating into alloy-disordered Al$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$19Ga$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$20As barriers, with $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$21 and with self-consistent confinement along the growth direction being imperative to reproduce the observed dependence (Freeman et al., 2024). This is an interfacial disorder problem in which the magnetic field induces the Wigner solid and thereby exposes the effect of barrier-interface disorder on the pinning mode.

A methodological formalization of field-induced interface pinning appears in equilibrium crystal-growth simulations. The interface pinning method stabilizes a two-phase crystal–liquid configuration by adding a harmonic bias field,

$$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$22

where $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$23 is a crystallinity order parameter. The method pins the interface against drift and extracts the kinetic coefficient from the terminal exponential relaxation of $$H_{\Lambda}(\phi,\omega;\beta,h) := \sum_{x\in\Lambda}(\beta\omega_x + h)\,\delta_x,$$24, treating crystal growth as a Smoluchowski process (Pedersen et al., 2014). Here the pinning field is explicitly artificial and computational, but it supplies a useful abstract template: a field conjugate to an order parameter can hold an interface in a metastable two-phase state and convert interfacial kinetics into an equilibrium fluctuation problem.

Across these domains, the common lesson is not microscopic uniformity but structural analogy. A field may reward substrate contacts, balance advective and return fluxes, tilt a domain-wall energy landscape, expose buried band alignment, or impose a harmonic constraint. What remains invariant is the central role of an interfacial coordinate whose dynamics are arrested, biased, or stabilized by the interplay of field coupling and local pinning structure.