Gap-Resolved Approach Overview
- Gap-resolved approach is a versatile strategy that isolates intrinsic gap features from extraneous effects via controlled experimental or algorithmic contrasts.
- It is applied across fields such as superconductivity, semiconductor spectroscopy, and optimization, enabling precise identification of gap-related phenomena.
- By distinguishing true intrinsic properties from background noise, this method enhances interpretive specificity and improves experimental and computational accuracy.
A gap-resolved approach is a methodological strategy in which a quantity called a “gap” is isolated by deliberately separating it from confounding structure. Across published uses of the term, the resolved object can be a superconducting gap, an electronic band gap, a gap function, an optimality gap, or a physical geometric gap. In bilayer nickelate ultrathin films, the phrase denotes a transfer-controlled STM/STS protocol that separates the intrinsic superconducting gap from oxygen-loss-induced spectral distortions (Wang et al., 14 May 2026). In ARPES, trARPES, field-angle thermodynamics, and terahertz spectroscopy, it denotes explicit resolution of gap structure in momentum, time, or field orientation (Lee et al., 2022). In optimization and equilibrium-constrained control, it denotes reformulations that drive a gap function or an optimality gap to zero or to a near-zero certificate (Lin et al., 2024).
1. Conceptual scope
The term is not tied to a single formalism. A plausible unifying description is that a gap-resolved approach replaces an averaged, indirect, or ambiguity-prone observable by an experimental or algorithmic construction that makes the target gap identifiable under controlled contrasts.
| Setting | Resolved quantity | Resolving mechanism |
|---|---|---|
| Bilayer nickelate STM/STS (Wang et al., 14 May 2026) | Intrinsic superconducting gap | Compare rapid cryogenic UHV transfer with longer room-temperature UHV exposure |
| Confined monolayer Ag ARPES/trARPES (Lee et al., 2022) | Occupied and unoccupied states across the band gap | Combine static ARPES with $6$ eV trARPES and compare to DFT-GW |
| Superconducting spectroscopies (Huang et al., 2012) | , , or field-angle-dependent gap structure | ARPES, trARPES, TDTS, and field-angle-resolved specific heat |
| Optimization and equilibrium problems (Panthee et al., 21 Aug 2025) | Optimality gap or gap-function residual | Exact reformulation, spatial branch-and-bound, or gap-function descent |
This domain dependence matters because the word “gap” changes meaning across fields. In superconductivity it is the pairing gap or the low-energy density-of-states suppression; in semiconductor spectroscopy it is the separation between band edges; in variational analysis it is a merit function that vanishes on the solution set; in global optimization it is the difference between incumbent and best bound.
2. Transfer-controlled isolation of the intrinsic nickelate gap
In "Atomically resolved intrinsic superconducting gap in (La,Pr)NiO films" (Wang et al., 14 May 2026), the gap-resolved approach is defined operationally by comparing atomically thin $1.5$-unit-cell films grown on under two transfer conditions that differ mainly in oxygen history. The rapidly transferred condition used rapid cooling below $150$ K, a low-temperature UHV suitcase, and transfer into the STM in less than 0 min. The longer-transfer condition involved more than 1 min of UHV exposure without cooling during room-temperature stages. The key empirical point is that the same nominal film, the same general structural order, and a superconducting transport onset can nevertheless yield sharply different local spectra when the post-growth oxygen history differs.
The rapidly transferred films showed atomically flat terraces, atomic-resolution images of the reconstructed lattice, and an ordered reconstructed surface. The abstract describes this as an ordered 2 surface, whereas the detailed extraction reports a clear 3 surface reconstruction (Wang et al., 14 May 2026). Under these conditions, atomic-resolution STS line scans showed a fully opened U-shaped gap, strongly suppressed zero-bias conductance, and two sets of coherence peaks within one spectrum. The characteristic gap scales were reported as approximately 4–5 meV for the inner gap and approximately 6 meV for the outer gap, with Dynes-fit values 7 meV and 8 meV (Wang et al., 14 May 2026).
By contrast, longer UHV exposure without cooling produced V-shaped spectra even though the surface reconstruction remained visible and transport still showed a superconducting onset in the 9–0 K range, with the abstract noting an onset above 1 K (Wang et al., 14 May 2026). The longer-transfer spectra showed a gap scale around 2 meV and more substantial zero-bias conductance. The paper therefore argues that surface structural order and bulk-like transport onset are not sufficient indicators that a local tunnelling spectrum is intrinsic. The line shape is more sensitive to oxygen stoichiometry than either reconstruction or transport onset.
The wide-energy-range spectra are central to this distinction. In the U-shaped, oxygen-sufficient condition, the broad asymmetric suppression had edges roughly at 3 meV and 4 meV. In the V-shaped, oxygen-loss-prone condition, the wide-energy dip changed to edges near 5 meV and 6 meV. These broader scales were interpreted as comparable to density-wave-related spectral features in oxygen-sensitive 7 systems, leading to the conclusion that oxygen loss mixes density-wave-related spectral weight into the low-energy tunnelling response (Wang et al., 14 May 2026).
3. Line shape, fitting, and symmetry inference
The decisive spectroscopic feature in the nickelate study is the extended flat zero-conductance bottom of the U-shaped spectrum. The paper treats this as strong evidence for a fully opened, nodeless superconducting gap because the local density of states is strongly depleted over a finite interval, not merely linearly suppressed near zero bias (Wang et al., 14 May 2026). This line-shape argument is strengthened by the transfer-time contrast: the V-shaped spectrum does not simply represent a nodal superconducting state, since it can be generated by oxygen-deficient handling while the reconstruction survives and transport still indicates superconductivity.
The temperature evolution was analyzed by extracting peak energies from the negative second derivative of symmetrized normalized 8. The characteristic temperatures obtained in this way were about 9 K and 0 K, comparable to the transport superconducting onset near 1 K (Wang et al., 14 May 2026). The paper states that the temperature dependence follows BCS-like behavior.
For modelling, the spectra were fit with a two-gap Dynes density of states. The tunnelling conductance was written as
2
Here 3 is the Fermi-Dirac distribution, 4 is the quasiparticle lifetime broadening, and 5 is the momentum-dependent superconducting gap. The authors report that a nodeless two-gap model reproduces the data well, whereas fits with a dominant nodal component fail, particularly because they cannot reproduce the flat zero-bias bottom (Wang et al., 14 May 2026).
The interpretive consequence is narrow but important. The paper does not claim that every V-shaped spectrum implies nodal pairing, and it explicitly argues the opposite for these films: V-shaped spectra can be composite spectra in which degraded superconductivity is mixed with an oxygen-loss-induced background tied to density-wave physics. In that sense, the gap-resolved approach functions as a symmetry filter. It distinguishes intrinsic nodeless behaviour from line shapes that could otherwise be overinterpreted as direct evidence for 6-wave or nodal pairing.
4. Momentum-, time-, and field-angle realizations
In momentum-resolved superconducting spectroscopy, the gap-resolved approach appears most explicitly in ARPES work on Fe-based superconductors. There the relevant object is not a single scalar gap but a momentum-dependent gap function 7 measured separately on each Fermi-surface sheet. The review "Angle-resolved photoemission studies of the superconducting gap symmetry in Fe-based superconductors" argues that ARPES generally finds nodeless and nearly isotropic gaps with small anisotropy, often consistent with global momentum-space forms such as 8, and distinguishes the true pairing gap 9 from an effective gap 0 inferred by probes sensitive to residual density of states and scattering (Huang et al., 2012).
In nonequilibrium ARPES, the same logic becomes time- and momentum-resolved. In superconducting Bi2212, trARPES showed that the near-nodal gap inside the normal-state Fermi arc collapses once the pump fluence exceeds about 1, while the far-off-nodal gap remains open up to at least 2–3. The response amplitude is therefore strongly momentum dependent, yet the recovery dynamics are nearly momentum independent across the measured angles (Smallwood et al., 2014). This use of “gap-resolved” does not isolate an intrinsic-versus-extrinsic line shape, as in the nickelate case; instead, it resolves gap dynamics across momentum space and after photoexcitation.
Time-domain terahertz spectroscopy on NbN provides a direct gap-resolved approach in the frequency domain. Because the superconducting gap lies in the low-THz range, the experiment measured 4, fit the conductivity to a BCS model, and extracted 5 explicitly. The gap suppression occurred on a 6 ps timescale and recovery on a 7 ps timescale, with Rothwarf–Taylor analysis yielding 8 ps, 9 per unit cell, and 0 (Beck et al., 2011).
Field-angle thermodynamics constitutes another major branch. In FeSe, field-angle-resolved specific heat separated three gaps 1, 2, and 3, assigning the smallest gap 4 meV to the electron-type 5 band and inferring two vertical-line nodes or gap minima along 6 from a four-fold azimuthal oscillation with sign change and an anisotropy-inverted polar-angle dependence (Sun et al., 2017). In CeRu7, by contrast, the low-8, low-9 disappearance of the fourfold oscillation below a characteristic field $1.5$0 supports an anisotropic $1.5$1-wave state without nodes, with gap minima along $1.5$2 (Kittaka et al., 2013). Theoretical work on polar-angle-resolved zero-energy density of states generalizes this logic, identifying a MAX structure, in which the global maximum of $1.5$3 shifts from anti-nodal to nodal directions with field, and a MIN structure, in which local minima appear near the nodal angle except for linear point nodes (Tsutsumi et al., 2016).
Outside superconductivity, "Confined monolayer Ag as a large gap 2D semiconductor and its momentum resolved excited states" uses a gap-resolved approach to map both sides of a semiconductor gap. Static ARPES located the valence-band maximum at $1.5$4 about $1.5$5 eV below $1.5$6, while $1.5$7 eV trARPES located the conduction-band minimum at $1.5$8 about $1.5$9 eV above 0, yielding an indirect gap of about 1 eV. The occupied valence-band dispersion matched DFT-GW well, but the conduction band showed an anomalously large effective mass 2, much larger than the GW value 3, with exciton-dressed dispersion preferred over electron-plasmon coupling as the explanation (Lee et al., 2022).
5. Gap functions, optimality gaps, and reformulation-based algorithms
In optimization and variational analysis, the term changes meaning but preserves the same structural ambition: the hidden object is a solution set, and the gap-resolved approach makes proximity to that set measurable and certifiable.
For three-phase infeasibility analysis in power grids, the original TPIA problem is a non-convex NLP that can trap local solvers at local minima or saddle points. The gap-resolved strategy reformulates the problem exactly as a non-convex bilinear program, applies spatial branch-and-bound, and uses sequential bound tightening to strengthen McCormick envelopes. The reported result is near-zero optimality gap, with gaps below 4 treated as “globally optimal for practical purposes,” and solution of cases above 5k nodes with runtime reductions up to 6 from the presolve bounding routine (Panthee et al., 21 Aug 2025).
For optimal control problems with equilibrium constraints, the key object is a variational-inequality gap function. A regularized gap function
7
allows the VI condition to be replaced by 8 together with 9, yielding a more concise and differentiable reformulation than KKT-based MPCC systems (Lin et al., 2024). A related development uses Auchmuty’s generalized primal gap and generalized D-gap functions together with a relaxation parameter 0 and a semismooth Newton flow. The resulting dynamical-system approach solves a sequence of gap-constraint-based reformulations and achieves local exponential convergence under standard assumptions (Lin et al., 2024).
A further generalization appears in variational inequalities, mixed variational inequalities, and equilibrium problems through a scaled Fukushima or Mastroeni regularized gap. There the decisive identity is that the scaled projection step is exactly the maximizer or minimizer that defines the regularized gap, so the search direction is simultaneously the algorithmic step and the gap-defining direction. Combined with a modified non-monotone Armijo line search, this yields global convergence for VI, MVI, and EP, and an 1-linear rate for strongly monotone Lipschitz VI/MVI under fixed or controlled-change variable metrics (Alshahrani, 22 Jun 2026).
6. Geometric extensions, interpretive cautions, and general significance
Some uses of the term are neither spectral nor variational, but geometric. In electrical-machine simulation, the reformulated air-gap element is not a conventional mesh-resolved finite-element discretization of the air region. Instead, it is a spectral macro-element approximation that resolves air-gap harmonics analytically, evaluates the coupling by FFT, and applies the operator matrix-free inside a Conjugate Gradient iteration, while incorporating rotation, skewing, and eccentricity (Gersem et al., 2023). In Hele-Shaw Faraday instability, a gap-resolved approach directly models the transverse gap flow and contact-angle dynamics rather than averaging them away. This yields a modified damping dependent on the static contact angle and hysteresis range and a combined amplitude equation in which capillary hysteresis is added to oscillatory Stokes-flow damping (Li et al., 11 Sep 2025).
A methodological caution follows from these examples. “Gap-resolved” does not guarantee that the inferred gap is fundamental unless the experimental or computational nuisance variables are also controlled. The nickelate STM study shows that a V-shaped low-energy spectrum can survive together with preserved reconstruction and superconducting transport onset, yet still fail to represent the intrinsic pairing state because oxygen loss mixes density-wave-related spectral weight into the tunnelling response (Wang et al., 14 May 2026). The Fe-based ARPES review makes an analogous distinction between the intrinsic pairing gap 2 and a smaller operational gap 3 that can emerge in probes sensitive to scattering-broadened residual density of states (Huang et al., 2012). In both cases, the central issue is not merely resolution, but interpretive specificity.
The broader significance is therefore methodological rather than terminological. A plausible implication is that a gap-resolved approach becomes valuable when the conventional observable conflates several mechanisms: intrinsic superconductivity with oxygen deficiency, quasiparticle gaps with pseudogap filling, band edges with excited-state many-body renormalization, solution quality with local-stationary-point artefacts, or geometric confinement with gap-averaged surrogates. The approach then proceeds by building a contrast—transfer condition, momentum cut, time delay, field orientation, auxiliary variable, bilinear lifting, or direct transverse model—that isolates the target gap from the background that would otherwise obscure it.