← All papers

Signal-Worker Boundary Confinement: A Corrected Ontology of Surface vs Bulk Transport

DOI: 10.5281/zenodo.21974194
Published: 2026-08-14

Author: Quni-Gudzinas, Rowan Brad (QNFO) ORCID: 0009-0002-4317-5604 Date: 2026-08-17 Version: v0.3 WBS: QNFO.INM.001 · Slug: signal-worker-boundary-confinement Status: Published v0.3 (2026-08-17 — DOI 10.5281/zenodo.21974194; concept 10.5281/zenodo.21931224)

Abstract

The Signal-Worker (S-W) ontology — boson = signal (the delocalized field instruction), fermion = worker (the localized state that performs work) — is the QNFO corpus's proposed decomposition of the wave–particle duality fog. This paper delivers the red-team-hardened correction of that ontology's boundary-confinement reading, following a 3-slot adversarial audit (2026-08-14) that found 5 HARD defects. The correction is a taxonomy, not a renaming: only mode-confinement phenomena (topological insulators, the quantum Hall effect, and the non-Hermitian skin effect) confine fermionic transport to the material boundary; the Meissner effect and the AC skin effect expel the field/current density while the electrons (the "workers") continue to flow through the bulk. The mapping also fails on composite bosons (Cooper pairs — two fermions condensing into one boson) and self-conjugate particles (Majorana zero modes), and the corpus's own flagship substrate (TaAs, a Weyl semimetal with a gapless conducting bulk) contradicts the "worker excluded from the bulk" claim. Every claim carries an epistemic label and a falsifiability condition; the ontology itself is explicitly labeled an unconfirmed internal proposal. Why a reader should care: the corrected taxonomy determines where "the worker is confined to the surface" is physically true (topological insulators, quantum Hall, non-Hermitian skin effect) and where it is a category error (Meissner, skin effect) — a distinction that decides which boundary phenomena can carry quantized transport and which cannot, with direct consequences for interpreting conductance measurements and for the corpus's superconductivity program. Premise-depth disclosure: the TERRITORY claims are as deep as the published physics they cite (bulk–boundary correspondence, spin-statistics, non-Hermitian topology); the MAP claims are derived from the ontology's own primitives (signal = delocalized field instruction, worker = localized state that performs work) and the imported corpus records [7–15]; the framework's premises END at the pre-registered falsifiability conditions P1–P3 (§10) — no deeper postulate is claimed.

v0.2 (2026-08-16). This newversion adds a classical-electrodynamics companion section (§3.6: bundling practice, skin-depth thresholds, and the quantum origin of the bulk–insulator distinction) and completes the deposited source set with references.bib and citation-audit.md. No changes to the v0.1 taxonomy or falsifiability register.

v0.3 (2026-08-17, published). Engages the NHSE literature in depth — the consolidated review [16] and the experimental realization [17] — naming non-reciprocal skin localization as the third member of the surface trichotomy; embeds the KIF-60 pre-registered predictions (P1–P3) in the falsifiability register (§10); closes the zero-in-text citation gaps on refs [6], [13], [15]. Red-team remediation (2026-08-17, commit 66764cd) scoped PRE-REG-1 with the point-gap/line-gap boundary condition and added the quantized-transport boundary literature [18,19]. Published 2026-08-17 as 10.5281/zenodo.21974194 (newversion of 10.5281/zenodo.21969297; concept 10.5281/zenodo.21931224).

1. Introduction and Positioning

The Signal-Worker ontology originates in five QNFO corpus records: Unifying Photosynthetic Energy Transduction and Ambient Superconductivity [7], Structural versus Driven Quantum Coherence [8], Quantum Architectonics [9], Gauge-Invariant Field Theory of Signal-Worker Interactions [10], and Quantum Abacus [11]. It proposes that bosons (photons/phonons) act strictly as informational signals directing localized fermions (electrons/excitons) to perform work, and that this functional decomposition is "non-dualistic."

A 3-slot CMD RED TEAM SUB (Accuracy / Completeness / Dependency, 2026-08-14) audited the ontology's central reading — "the signal orders where the worker may act; in topologically-protected phases the worker is excluded from the bulk and confined to the boundary." Verdict: the underlying physics claims are accurate (Accuracy: 0 HARD), but the unifying reading is overgeneralized (Completeness: 5 HARD) and the corpus carries a terminology collision (Dependency: LCI used for two different concepts). This paper formalizes the corrected position.

Scope. This paper does NOT re-derive the S-W ontology or the topological-material corpus (see the deep-due-diligence report, companion artifact). It corrects the boundary-confinement claim, the composite-particle mapping, and the epistemic status of the ontology, and it fixes the LCI terminology collision. v0.2 adds §3.6 — the classical-EM companion that applies this correction to engineering practice — plus the two missing provenance files.

2. The Signal-Worker Ontology, Stated

RoleCarrierWhat it doesCorpus label
Signalboson (photon / phonon)carries the instruction — a delocalized field modifierinformation
Workerfermion (electron / exciton)performs the work — a localized state vectoraction

Statement 1 [MAP — interpretive, proposed]. Wave–particle duality is functionally decomposable into a bosonic instruction role and a fermionic action role in driven non-equilibrium systems. Status: internal proposal; unconfirmed; ambient superconductivity has never been achieved. The framing is hierarchical (the signal orders the worker), so it is not non-dualistic in the strict sense — it privileges one pole.

3. The Established Physics: Surface vs Bulk — a Taxonomy

The counterintuitive surface-vs-bulk phenomena the ontology cites are real, but they are mechanistically heterogeneous. They split into three classes: (i) mode confinement — boundary-localized transport with gapped/inert bulk (topological insulators §3.1, quantum Hall effect §3.2, non-Hermitian skin effect §3.5); (ii) field expulsion — the field leaves the bulk while the carriers stay (Meissner §3.3); (iii) current-density redistribution — no confinement at all (AC skin effect §3.4). The trichotomy is fixed by mechanism, not by scale (§3.6, §4).

3.1 Mode confinement (topological insulators) — TERRITORY

A 3D Z₂ topological insulator has a gapped (insulating) bulk and hosts gapless, metallic surface states [1,2]. The surface states are protected by time-reversal symmetry and the bulk Z₂ index via the bulk–boundary correspondence. Electron transport is genuinely confined to the boundary. This is a mode-confinement phenomenon: the bulk does not conduct, the surface does, and the surface modes are spin-momentum locked (helical).

3.2 Mode confinement (quantum Hall effect) — TERRITORY

In a 2D electron gas at strong magnetic field, the quantized Hall conductance is carried by chiral edge channels; the bulk is inert [2]. Again, transport is genuinely boundary-localized (chiral, in this case).

3.3 Field expulsion (Meissner effect) — TERRITORY

A superconductor actively expels the magnetic field: B ≈ 0 in the bulk, with the field decaying over the London penetration depth (λ_L ≈ 10–100 nm) [2]. This is active expulsion, distinct from a perfect conductor's flux-trapping. Crucially, the electrons do NOT leave the bulk — they flow through it as a dissipationless supercurrent. The London depth is the field's decay length, not an electron-localization length. The "worker" stays; the "field" leaves. The corpus's own superconductivity-validation record — Superconductivity Quadrangle [13] — is consistent with this reading: tensor-locked resilience in topological substrates concerns the condensate and its screening response, not boundary confinement of the carriers.

3.4 Current-density redistribution (AC skin effect) — TERRITORY

At high frequency, AC current density crowds toward the conductor surface over the skin depth δ ∝ 1/√f. This is a current-density redistribution within the conductor — no particle confinement, no boundary-localized eigenmodes.

3.5 Mode collapse (non-Hermitian skin effect) — TERRITORY (external, recent)

Under non-Hermiticity (gain/loss or asymmetric hopping), the non-Hermitian skin effect (NHSE) drives all bulk eigenmodes to collapse onto the boundary (Yao–Wang 2018) [3]. This is the literal "bulk → boundary" analog and the most on-point mode-confinement phenomenon for the ontology's vocabulary — and it is absent from the S-W corpus (a Completeness finding of the 2026-08-14 audit). [v0.3] The mechanism is now consolidated: the review literature [16] systematizes the modified non-Bloch bulk–boundary correspondence — spectral-winding and point-gap invariants replace the Hermitian Z₂/Chern classification — and the effect has been experimentally realized: an ultrafast topological non-Hermitian skin mode bound to a frequency-jump interface inside a gain-switched semiconductor laser, with direct intensity sampling of the skin modes (583 ± 16 fs FWHM) [17]. NHSE is therefore the third member of the surface trichotomy — non-reciprocal skin localization — distinct from both classical field redistribution (§3.4) and Hermitian mode confinement (§3.1–3.2). Its defining quantity is the point-gap winding number, an integer count, not a length scale. Quantized conductance in non-Hermitian systems arises only from line-gap/biorthogonally protected edge channels — demonstrated for non-Hermitian Chern insulators (Yu & Zhai 2018 [18]) — never from skin-localized bulk modes alone; NHSE can even localize chiral edge states outright (Liu et al. 2024 [19]). PRE-REG-1 (§10) codifies this boundary condition.

3.6 Engineering companion: bundling, skin-depth thresholds, and the quantum bulk–insulator boundary [v0.2]

A recurring misreading holds that conductors are bundled because of the skin effect. The engineering record says otherwise — and the distinction is a live instance of the category discipline of §4. [TERRITORY — claimed identity; disconfirmed if the engineering record shows skin-effect mitigation as the dominant bundling motive at mains frequency (it does not: stranded = flexibility, HV bundled = corona suppression, Litz = high-frequency only)]

Bundling practice. Three distinct practices are conflated under "bundling":

  • Stranded conductors (ordinary copper cable) exist for mechanical flexibility. At 50/60 Hz the skin depth in copper is $\delta = \sqrt{2\rho/\omega\mu} \approx 9.2$ mm at 50 Hz and $\approx 8.4$ mm at 60 Hz (computed with $\rho = 1.678\times10^{-8}\ \Omega\text{m}$, $\mu = \mu_0$; BP-10 recompute 2026-08-17), so stranded wires below ~17 mm overall diameter conduct through essentially the full cross-section at mains frequency.
  • Bundled conductors on high-voltage lines (2–4 subconductors per phase, usually aluminum conductor steel-reinforced — ACSR, not copper) exist primarily to suppress corona discharge by reducing the surface electric-field gradient. Electrically, bundling lowers the series inductance (larger effective geometric mean radius) and raises the shunt capacitance (larger equivalent radius), which raises the surge-impedance loading and the power-transfer capability.
  • Litz wire (individually insulated strands) is the only bundling scheme whose purpose is the skin/proximity effect — and it exists for high-frequency use, not mains.

So the AC skin effect is real, but almost none of the wire we see is bundled because of it. The error is a category slip: a field/current-density phenomenon (skin effect) is misattributed as the motive for a structural practice (bundling) — the same slip the corrected taxonomy of §4 exists to prevent.

Where quantum effects actually enter the bulk–insulator boundary. The conductor/insulator distinction itself is quantum mechanical: copper conducts because its band structure provides a partially filled band of delocalized Bloch states, while an insulator presents a filled valence band plus a gap. The AC skin effect, by contrast, is classical Maxwell electrodynamics — no quantum input is required at macroscopic scales. Quantum physics becomes explicit for surface-vs-bulk questions when a length scale approaches the Fermi wavelength ($\lambda_F \approx 0.46$ nm in copper) or the electron mean free path (~40 nm at room temperature): conductance quantization in integer multiples of the conductance quantum $G_0 = 2e^2/h$, ballistic transport, quantum confinement, and — genuinely quantum and genuinely boundary-localized — the topological surface states of §3.1, protected by bulk topology. The skin effect and the topological boundary are therefore not the same kind of "surface": the former is a classical field redistribution inside a conductor; the latter is a mode-confinement effect. The §4 category distinction survives the nanoscale.

S-W reading. In Signal-Worker vocabulary: under the AC skin effect, the signal (the electromagnetic field) is redistributed toward the surface while the workers (the conduction electrons) remain distributed through the bulk; in a topological insulator, the boundary modes carry the work. Nothing in this companion section changes the v0.1 taxonomy — it supplies the engineering and quantum-threshold context in which that taxonomy is applied.

4. The Category Error and Its Correction

Finding [HARD]. The S-W reading "the worker is excluded from the bulk and confined to the boundary" conflates field/current-density expulsion (Meissner, skin effect) with mode confinement (topological insulator, quantum Hall). They are different physical mechanisms with different consequences for where the "worker" may act.

Correction (this paper's primary claim):

ClassPhenomenaWhat is at the boundaryDoes the "worker" (electron) leave the bulk?
Mode confinementTI surface states, QH edge channels, NHSEElectron transport modesYes (bulk gapped/inert)
Field expulsionMeissner effectThe B fieldNo (electrons flow through bulk)
Current-density redistributionAC skin effectThe current densityNo (redistribution, not confinement)

Statement 2 [MAP — to be defended]. The boundary-confinement reading of the S-W ontology is valid only for the mode-confinement class. A corrected framing must distinguish field-expulsion from mode-confinement. Falsifiability condition C2: the corrected taxonomy is disconfirmed if it yields no new observable consequence beyond the standard bulk–boundary correspondence — i.e., if it is pure relabeling (the criterion the Prime Valuation Depth follow-on applied to itself).

5. The Composite-Boson Problem

Finding [HARD]. The ontology's flagship example is superconductivity, yet the superconducting charge carrier is a bosonic Cooper pair — two fermionic "workers" condensing into a "signal-like" boson. The mapping boson = signal / fermion = worker is not closed under composite particles:

  • Cooper pairs (two fermions → one boson): the superconducting "worker" is a boson; in the condensate gauge invariance is spontaneously broken and the photon acquires mass (Anderson 1963 [6]) — a pairing-level fact the functional split does not generate. The corpus's own fluxonium record (Ab Initio Architectonics [15]) applies the S-W vocabulary to exactly such Cooper-pair devices, where the operative "worker" is the pair, not the electron.
  • Phonons: bosons that serve as both the pairing glue (the signal) and a lattice excitation — the instruction/action split blurs.
  • Majorana zero modes (self-conjugate, neither boson nor fermion): fall outside both labels. The corpus itself works this thread (ZBW-Majorana, [14]).

Correction. The functional split is a regime-specific interpretation, not a universal classification. It must explicitly exclude composites and self-conjugate particles.

6. Spin-Statistics Engagement

Finding. "boson = signal / fermion = worker" is a teleological gloss over the real boson/fermion distinction — the spin-statistics theorem (integer vs half-integer spin). The ontology re-describes a kinematic fact without deriving a new observable.

Correction. The corrected framing must engage spin-statistics directly: the kinematic distinction (integer/half-integer spin → symmetric/antisymmetric statistics) is the ground truth; the functional signal/worker roles are a valid interpretation only where field modifiers and localized carriers are mechanistically distinct (driven non-equilibrium regimes).

7. The Weyl-Semimetal Counterexample

Finding [HARD]. The corpus's own flagship substrate is TaAs, a Weyl semimetal — with a gapless, conducting bulk and Fermi-arc surface states (Quantum Abacus [11]; Xu et al. 2015 [5]; Wan et al. 2011 [4]). The corpus additionally contains Hamiltonian Engineering of Topological Deconfinement in Weyl Semimetals [12], which already engages Weyl physics directly. Weyl semimetals are not Z₂ topological insulators: the bulk conducts. Therefore "the worker is excluded from the bulk" is false for the corpus's own material.

Correction. The paper must distinguish: (a) gapped-bulk mode confinement (TI/QH), (b) gapless-bulk Weyl/Dirac semimetals (Fermi arcs — boundary states on top of a conducting bulk). The S-W vocabulary may describe Fermi arcs as boundary-enhanced work, but not as bulk-excluded work.

8. The LCI Acronym Collision

Finding [HARD — dependency]. The corpus uses LCI for two different concepts:

RecordLCI =
Gauge-Invariant Field Theory [10] (10.5281/zenodo.18466522)Logical Cloning Prohibition (Ward identity of the Signal gauge field)
Structural vs Driven [8] (10.5281/zenodo.18441402); Quantum Architectonics [9] (10.5281/zenodo.18515458)Lossless Complexity Index

Correction. This paper uses LCI only for Logical Cloning Prohibition [10] and spells out Lossless Complexity Index in full wherever it appears [8,9]. Any cross-corpus citation must disambiguate.

9. The Corrected Signal-Worker Framing

The corrected framing is a three-part statement:

  1. [TERRITORY — claimed identity; disconfirmed if a boundary-localized transport phenomenon falls outside all three classes (mode confinement / field expulsion / current-density redistribution), or if fermionic boundary transport is observed in a gapless-bulk geometry without bulk conduction] The boson/fermion distinction is the spin-statistics theorem (kinematic). Surface-vs-bulk transport splits into mode confinement (TI, QH, NHSE — boundary-localized transport, gapped/inert bulk), field expulsion (Meissner — field leaves, electrons stay), and current-density redistribution (skin — no confinement).
  2. [MAP] The S-W "instruction vs action" split is a valid functional interpretation in driven non-equilibrium regimes, and a valid mode-confinement statement only for the mode-confinement class. It is not a universal decomposition: it fails on composite bosons (Cooper pairs), self-conjugate particles (Majorana zero modes), and gapless-bulk Weyl semimetals.
  3. [EPISTEMIC] The ontology is an unconfirmed internal proposal. Ambient superconductivity has never been achieved; the Ward-identity no-cloning derivation (LCI [10]) is flagged speculative and must be independently reproduced. Boundary confinement is a transport statement — it does NOT imply topological-QC fault tolerance at nonzero temperature (the corpus's own evidence synthesis refutes the FCI alternative via thermal anyon proliferation; see deep-due-diligence companion).

10. Falsifiability Register

|

| Claim | Type | Falsifiability condition | Status |

|:--|:------|:-----|:--------------------------|:-------| | C1 | TI/QH confine fermionic transport to the boundary; Meissner/skin expel the field/current | established | — | established | | C2 | The corrected taxonomy (field-expulsion vs mode-confinement) carries content beyond relabeling | MAP | no new observable consequence → relabeling | OPEN | | C3 | LCI (Logical Cloning Prohibition) as a Ward identity, exponential-in-N scaling | MAP-speculative | independent derivation + reproduction fails | OPEN, flagged | | C4 | Composite bosons (Cooper pairs) break the boson=signal/fermion=worker mapping | MAP | a closed mapping under composites exists | OPEN | | P1 | Skin-localized bulk modes in the pure point-gap (NHSE) regime do NOT themselves carry quantized conductance G = nG0 and are NOT classical field redistribution; quantized conductance in non-Hermitian systems requires line-gap/biorthogonally protected edge channels [18], and NHSE can localize such chiral edge channels [19] | MAP (pre-registered, KIF-60) | observation of G = nG0 quantization carried by skin-localized bulk modes absent a line-gap protected edge channel, or demonstration that NHSE localization is classical field redistribution | PRE-REGISTERED 2026-08-16; REVISED 2026-08-17 | | P2 | Cooper-pair boundary transport does NOT obey the original boson=signal mapping (composite-boson exception C4 holds) | MAP (pre-registered, KIF-60) | Cooper-pair boundary transport obeys boson=signal with no exception | PRE-REGISTERED 2026-08-16 | | P3 | The surface/bulk distinction becomes experimentally indistinguishable as kBT → Δgap (edge quantization vanishes with T/Tgap → 1) | MAP (pre-registered, KIF-60) | quantized edge conductance survives at kBT >> Δ_gap | PRE-REGISTERED 2026-08-16 |

[v0.3] Pre-registration (KIF-60, 2026-08-16). Rows P1–P3 are the risky predictions pre-registered in the Phase 1b Bayesian Evidential Weight gate (artifacts/bayesian-evidential-weight.md, commit 03ac213, sha256 064e1ee6…). C1 is established physics (zero novelty weight); C2 is [RETRODICTION] until P1–P3 are tested; P1 is the only claim with a pre-registered experimental falsification path (platform: [17]). P1 was revised 2026-08-17 (commit 66764cd) with the line-gap/biorthogonal scoping and boundary-condition refs [18,19]; the original 2026-08-16 wording is preserved in the BEW file's revision record.

11. Conclusion

The counterintuitive surface physics of topological conductors is real and correctly described — but it is narrower than the Signal-Worker framing claimed. "Electron confined to the surface" holds for topological insulators, the quantum Hall effect, and the non-Hermitian skin effect (gapped/inert bulk → boundary transport). It does not hold for the Meissner effect or the AC skin effect (field/current expulsion — electrons still flow through the bulk), and it does not hold for the corpus's own Weyl-semimetal substrate (conducting bulk with Fermi arcs). The Signal-Worker ontology, corrected, is a regime-specific functional interpretation, explicitly unconfirmed, with its composite-particle, spin-statistics, and terminology gaps closed. The physics survives the audit; the ontology, corrected, survives it too — as a falsifiable MAP, not as established doctrine. The falsifiability register now carries three pre-registered risky predictions (P1–P3, §10): experimental confirmation of any of them would give the corrected ontology positive evidential weight; until then its MAP claims remain explicitly unconfirmed.

Declarations

Funding. This work received no external funding. Competing interests. The author declares no competing interests. Data availability. No experimental data were generated or analyzed for this work. All literature evidence is cited with persistent identifiers. Code availability. No code was required for this work. Ethics approval. Not applicable. Preprint policy. This manuscript is posted as a preprint; it has not been submitted to any journal.

References

  1. M. Z. Hasan and C. L. Kane. Colloquium: Topological insulators. Reviews of Modern Physics, 82:3045, 2010. DOI 10.1103/RevModPhys.82.3045.
  2. X.-L. Qi and S.-C. Zhang. Topological insulators and superconductors. Reviews of Modern Physics, 83:1057, 2011. DOI 10.1103/RevModPhys.83.1057.
  3. S. Yao and Z. Wang. Edge states and topological invariants of non-Hermitian systems. Physical Review Letters, 121:086803, 2018. DOI 10.1103/PhysRevLett.121.086803.
  4. X. Wan, A. M. Turner, A. Vishwanath, and S. Y. Savrasov. Topological semimetal and Fermi-arc surface states in the electronic structure of pyrochlore iridates. Physical Review B, 83:205101, 2011. DOI 10.1103/PhysRevB.83.205101.
  5. S.-Y. Xu et al. Discovery of a Weyl fermion semimetal and topological Fermi arcs. Science, 349:613, 2015. DOI 10.1126/science.aaa9297.
  6. P. W. Anderson. Plasmons, gauge invariance, and mass. Physical Review, 130:439, 1963. DOI 10.1103/PhysRev.130.439.
  7. QNFO. Unifying Photosynthetic Energy Transduction and Ambient Superconductivity via a Non-Dualistic Signal-Worker Ontology. Zenodo, 2026. DOI 10.5281/zenodo.18330366.
  8. QNFO. Structural versus Driven Quantum Coherence: A Proposed 'Signal-Worker' Framework for Ambient Superconductivity. Zenodo, 2026. DOI 10.5281/zenodo.18441402.
  9. QNFO. Quantum Architectonics: A Unified Framework for Substrate Engineering via Topological Genesis, Signal-Worker Dynamics, and Multi-Modal Control. Zenodo, 2026. DOI 10.5281/zenodo.18515458.
  10. QNFO. Gauge-Invariant Field Theory of Signal-Worker Interactions: Deriving the Logical Cloning Prohibition from First Principles of Quantum Architectonics. Zenodo, 2026. DOI 10.5281/zenodo.18466522.
  11. QNFO. Quantum Abacus: A Strain-Engineered Platform for Passive, Reversible Fermionic Computation. Zenodo, 2026. DOI 10.5281/zenodo.18543167.
  12. QNFO. Hamiltonian Engineering of Topological Deconfinement in Weyl Semimetals: Addressing the Thermal Scalability of Quantum Error Correction. Zenodo, 2026. DOI 10.5281/zenodo.18222365.
  13. QNFO. Superconductivity Quadrangle: Validating Tensor-Locked Resilience in Topological Substrates. Zenodo, 2026. DOI 10.5281/zenodo.18496890.
  14. R. Quni. Vanishing ZBW Signal: The ZBW-Majorana Hypothesis as a Unified Framework for Topological Fermion Distinction. Zenodo, 2026. DOI 10.5281/zenodo.21574555.
  15. QNFO. Ab Initio Architectonics: Rethinking Fluxonium Qutrits through the Signal-Worker Ontology. Zenodo, 2026. DOI 10.5281/zenodo.18447478.
  16. R. Lin, T. Tai, L. Li, and C. H. Lee. Topological non-Hermitian skin effect. Frontiers of Physics, 18(5):53605, 2023. DOI 10.1007/s11467-023-1309-z.
  17. B. Schneider, A. Dikopoltsev, M. Bestler, P. Täschler, M. Beck, D. Burghoff, O. Zilberberg, and J. Faist. Ultrafast Non-Hermitian Skin Effect. arXiv:2505.03658, 2025.
  18. C. Yu and H. Zhai. Hall conductance of a non-Hermitian Chern insulator. Physical Review B, 98:245130, 2018. DOI 10.1103/PhysRevB.98.245130.
  19. G.-G. Liu, S. Mandal, P. Zhou, X. Xi, R. Banerjee, Y.-H. Hu, M. Wei, M. Wang, Q. Wang, Z. Gao, H. Chen, Y. Yang, Y. Chong, and B. Zhang. Localization of Chiral Edge States by the Non-Hermitian Skin Effect. Physical Review Letters, 132:113802, 2024. DOI 10.1103/PhysRevLett.132.113802.