{"id":"403de8cc-dae6-4bbb-8281-0f27b22849f8","arxiv_id":"1908.00956","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Anapole scattering zeros are not eigenmodes and require external illumination, while ideal embedded eigenstates are provably unexcitable by external light due to Lorentz reciprocity.","lead":"This paper clears up a long-standing confusion in nanophotonics: anapole 'dark modes' are not self-sustained modes of a scatterer but scattering zeros that need external illumination to exist. It proves that true nonradiating eigenmodes, such as embedded eigenstates, cannot be excited by external light at all, and it shows numerically that anapoles radiate away once illumination stops.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Zero-excitation proof relies on the eigenmode field vanishing exactly outside the scatterer; generic BICs have evanescent tails, so the claim is proven only for strictly confined embedded eigenstates.","rationale":"The central reciprocity argument is clean and internally consistent under the stated premise that a nonradiating source produces exactly zero field outside V_s. The reader's weakest assumption identifies precisely this premise. In an open homogeneous medium, a source-free exterior solution at real frequency with no radiation is identically zero, which is why the proof works for the spherical layered-sphere example with a zero-permittivity shell. However, the paper's abstract and introduction extend the conclusion to embedded eigenstates and BICs in open cavities generally, and many BICs reside in periodic or structured claddings where the exterior field is evanescent but nonzero. For those modes, nonradiating means zero radiated power, not zero field, and the identity ∫ J1·E2 dV = 0 is no longer forced by Lorentz reciprocity. This is a real scope limitation rather than an internal contradiction. Separately, the modal-expansion formula (5) is relegated to an absent Supporting Information section and is garbled in the arXiv text, so the derivation cannot be checked; the stated '2 µs' decay time for a 100 nm sphere is also almost certainly a unit typo. These issues do not overturn the conceptual distinction between nonradiating scattering states and nonradiating eigenmodes, but they do mean the headline claim requires qualification. A concrete test on a photonic crystal slab BIC would settle whether the general claim survives or should be restricted to strictly confined embedded eigenstates.","tokens_in":9166,"tokens_out":10159,"duration_ms":118652,"concrete_test":"Use a full-wave eigenmode solver to compute a known photonic crystal slab BIC (e.g., the periodic dielectric slab of Hsu et al., Nature 2013) and directly evaluate the exterior field magnitude just outside the slab and the overlap S = ∫_{V_s} (ε-ε0) E_BIC · E_inc dV for a normally incident plane wave and for a point dipole in the near field. If |E_ext| ≠ 0 or S ≠ 0 for any causal source, Eq. (3) does not apply to this class and the paper's headline claim must be restricted to strictly confined embedded eigenstates with zero exterior field.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The most load-bearing step is identifying the embedded eigenmode's induced polarization current with a nonradiating source in the strict Wolf sense, i.e., one whose field vanishes identically in V\\V_s (Eqs. 1-3). This identification is what turns the overlap integral in Eq. (5) into the vanishing integral (3). The paper's own example (Fig. 2b) satisfies this because the outer shell has near-zero permittivity and the text asserts the external fields are zero. But the abstract and introduction claim the conclusion for embedded eigenstates and BICs in open cavities generally. Many BICs, such as photonic crystal slab modes or waveguide-array modes, have nonzero evanescent fields in the exterior cladding and are nonradiating only in the sense of zero power flux at infinity, not zero field. For such modes the right-hand side of Eq. (1) includes the integral of J2 with the nonzero exterior field E1, so Eq. (2) need not follow and the orthogonality proof does not apply. The conclusion that 'any external causal illumination' gives a_n=0 is therefore not established for this broader class. The missing Supporting Information derivation of Eq. (5) prevents checking whether this confinement assumption enters the modal-expansion argument elsewhere, and the printed Eq. (5) is garbled by OCR, making the formula impossible to verify from the arXiv text.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses a controversy in nanophotonics about whether radiationless states can be excited by external illumination. It distinguishes two classes: anapole states, which are scattering zeros of an open cavity and therefore require an incident field to exist, and embedded eigenstates (bound states in the continuum), which are true nonradiating eigenmodes. The central claim is that an ideal embedded eigenstate has identically zero excitation amplitude under any external causal illumination, because its induced polarization current is a nonradiating source, so Lorentz reciprocity (Eqs. 1-3) forces the spatial overlap integral in the modal expansion (Eq. 5) to vanish. The paper supports this with Mie-theory plots, field distributions, and transient simulations, and concludes that anapole-based lasing is not directly possible.","tokens_in":9386,"tokens_out":3899,"duration_ms":41824,"significance":"If the central claim holds, the paper provides a useful conceptual clarification in a literature where the terms 'anapole mode' and 'embedded eigenstate' are often conflated. The reciprocity argument is elegant, parameter-free, and grounded in Wolf's classical nonradiating-source theorem, and the distinction between a scattering zero and a nonradiating eigenmode is likely to be influential. The paper also gives a clear physical picture for why anapoles radiate after the incident field is switched off. However, the significance is tempered by the fact that the proof, as written, applies to strictly confined embedded eigenstates (fields vanishing exactly outside the scatterer) rather than to the broad class of open-cavity BICs with evanescent tails that the introduction claims.","major_comments":[{"comment":"The central conclusion that a_n = 0 for an embedded eigenstate rests on the modal-expansion coefficient formula (5), but the printed equation is corrupted by OCR artifacts (e.g., the garbled '2( ) ) ( () ( ), np iT it n nn nnV n ic e e d iat' fragment) and its derivation is relegated to a Supporting Information document that is not included in the arXiv posting. This makes the load-bearing step unverifiable from the main text. The authors should provide the full derivation of Eq. (5) and a clean, legible equation in the paper or in an accessible supplement.","section":"Role of Lorentz Reciprocity, Eq. (5)"},{"comment":"The orthogonality argument requires that the nonradiating eigenmode's field vanish identically in the exterior volume V\\V_s, so that the induced polarization current is a nonradiating source in the strict Wolf sense. Many bound states in the continuum, such as photonic-crystal slab modes or waveguide-array modes, have evanescent tails in the exterior cladding and are nonradiating only in the sense of zero power flux at infinity, not zero field. For such modes, the right-hand side of Eq. (1) includes a contribution from the exterior field, and Eq. (3) need not vanish. The abstract and introduction claim the conclusion for embedded eigenstates and BICs in open cavities generally, but the proof as written establishes it only for strictly confined embedded eigenstates such as the zero-permittivity-shell example in Fig. 2(b). Please either restrict the claim to strictly confined eigenstates or extend the proof to modes with nonzero exterior tails.","section":"Eqs. (1)-(3), Fig. 2(b)"}],"minor_comments":[{"comment":"The transient simulation in Fig. 4 is described only qualitatively; please provide simulation parameters such as pulse duration T_p, polarization, mesh resolution, and boundary conditions so that the radiative-decay timescale can be reproduced.","section":"Fig. 4 caption and surrounding text"},{"comment":"The sign and time-harmonic convention for the induced polarization current J_1 = -iω(ε(r)-ε0)E_int are not defined in Eq. (1)-(3); the convention is mentioned only in the Fig. 1 caption. Please state the convention explicitly where J is first introduced.","section":"Eq. (3) and general notation"},{"comment":"The text contains numerous typesetting and OCR artifacts (e.g., '1 1TMc' for the Mie coefficient, 'ite ω−' in the Fig. 1 caption, and the missing closing parenthesis in reference [39]). A careful copyediting pass is needed.","section":"Throughout"},{"comment":"The statement that an anapole state 'cannot be used, directly, to achieve lasing' is presented as a firm conclusion; consider softening this to a conjecture or supporting it with a more detailed pole-zero analysis, since the proposal in Ref. [15] may involve gain and nonlinearities not covered by the present linear scattering argument.","section":"Section 'Anapoles vs. Embedded Eigenstates', final paragraph"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely correct under its explicitly confined-eigenstate assumption, and the conceptual clarification is valuable. However, the missing Supporting Information for Eq. (5) is a serious issue for a Letter, and the scope of the claim in the abstract and introduction ('embedded eigenstates and bound states in the continuum in open cavities') goes beyond what the proof actually covers. The authors should be asked to either provide the missing derivation and a legible Eq. (5), or explicitly narrow the claim to strictly confined embedded eigenstates. This is fixable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe short version: worth reading and worth sending to referees, but the central claim is proved for a narrower class than the abstract promises. The paper makes a genuinely useful distinction between radiationless scattering states (anapoles) and nonradiating eigenmodes (embedded eigenstates/BICs), and shows via Lorentz reciprocity that an ideal nonradiating eigenmode with strictly confined fields cannot be excited by external illumination. The anapole-as-scattering-zero story is classic Mie theory repackaged, but the framing is helpful, and the transient demonstration that an anapole radiates when the pump is switched off is a nice concrete illustration.\n\nThe reciprocity argument itself is sound: if the induced polarization current of the eigenmode is nonradiating in the strict Wolf sense—field identically zero outside the object—then Eq. (3) forces the overlap integral in the modal expansion to vanish, so the excitation coefficient a_n = 0. That is a clean result. The problem is the jump from this strict confinement to the general category of embedded eigenstates and BICs. Many BICs in photonic crystals and waveguide arrays have non-zero evanescent fields outside the structure; they are nonradiating in power-flux terms but do not have zero exterior fields. For those, the Wolf orthogonality argument does not directly apply, and the paper does not prove a_n = 0. The specific example in Fig. 2(b) (zero-permittivity shell) satisfies the strict condition, so the theorem holds there, but the abstract and introduction overclaim generality.\n\nAlso worth noting: the key modal expansion Eq. (5) is deferred to Supporting Information not included in the arXiv posting, and the printed formula is garbled by OCR. This is addressable, but it means the main text's central equation can't be checked from the arXiv alone. There's also a physically implausible '2 µs' decay time for a 100-nm sphere (likely a typo for ps or ns).\n\nNone of this overturns the paper. The taxonomy is important for people working in nonradiating photonics, and the reciprocity proof is a useful formal anchor. A thoughtful referee can fix the overclaim and the typos. My recommendation: send it to peer review, but ask the authors to state clearly that the zero-excitation theorem applies to strictly confined embedded eigenstates, and to clarify which BIC classes satisfy that condition.","headline":"A clean conceptual clarification—anapoles as scattering zeros, embedded eigenstates as unexcitable poles—but the zero-excitation proof is proved only for strictly confined eigenmodes, so the abstract's general BIC claim needs qualifying.","tokens_in":9956,"tokens_out":2319,"would_cite":true,"duration_ms":22687,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Ideal nonradiating eigenmodes of open cavities have zero modal amplitude under any external causal illumination, because Lorentz reciprocity forces the overlap between the nonradiating polarization current and the incident field to vanish.","keywords":["nonradiating sources","anapole modes","embedded eigenstates","bound states in the continuum","Lorentz reciprocity","Mie scattering zeros","open cavities","quasinormal modes"],"falsifier":"Construct an open reciprocal cavity whose purported embedded eigenstate is computed numerically, illuminate it with a time-domain plane-wave pulse tuned to the real eigenfrequency, and measure the field energy inside the cavity after the pulse has left: if the energy is nonzero and persists longer than the pulse, the zero-excitation theorem is violated. Alternatively, compute the overlap integral $\\int_{V_s} (\\varepsilon-\\varepsilon_0)\\mathbf E_{\\mathrm{int}}\\cdot \\mathbf E_{\\mathrm{inc}}\\,dV$ for a mode with a finite exterior tail and show it is nonzero while scattering remains zero.","tokens_in":8947,"feed_emoji":"💡","tokens_out":6912,"duration_ms":67610,"temperature":0.7,"pith_summary":"Nonradiating field distributions come in two kinds, and only one kind can be switched on from outside. The paper argues that anapole states, scattering zeros where a particle's dipolar response vanishes, are not eigenmodes of the open cavity at all: they exist only while an incident field is present, and they begin radiating as soon as the excitation is switched off. In contrast, an ideal embedded eigenstate, also called a bound state in the continuum, is a self-sustained nonradiating eigenmode of an open cavity, and Lorentz reciprocity forces its modal amplitude under external illumination to be exactly zero. In other words, an exact nonradiating eigenmode of a linear reciprocal open system cannot be excited by external light, although arbitrarily close approximations with high quality factors can be. The distinction matters because it separates 'invisible under illumination' from 'trapped light that needs no illumination'.","feed_headline":"Nonradiating eigenmodes cannot be excited from outside","feed_subtitle":"Lorentz reciprocity forces zero modal amplitude; anapole states radiate as soon as the incident wave is switched off.","key_machinery":"The load-bearing mechanism is the nonradiating-source orthogonality theorem combined with the quasinormal-mode expansion of fields inside open cavities. The theorem states that any continuous nonradiating current distribution confined to a finite volume is orthogonal, under a volume integral, to any field in that volume that satisfies Maxwell's equations; applied to the induced polarization current $(\\varepsilon(\\mathbf r)-\\varepsilon_0)\\mathbf E_{\\mathrm{int}}(\\mathbf r)$ with the incident field as the test field, it makes the overlap integral that determines modal excitation vanish. The companion piece is the complex-frequency pole picture: eigenmodes of open cavities sit at poles of Mie coefficients in the upper half of the complex-frequency plane, whereas an embedded eigenstate has a purely real eigenfrequency, and an anapole is a zero, not a pole, along the real axis. The time-domain expansion of the internal field into these eigenmodes converts the vanishing overlap into zero modal amplitude.","core_discovery":"The paper's central claim is that radiationless field distributions split into two classes that behave differently under external excitation. A zero of a Mie scattering coefficient, such as the anapole of a dielectric sphere, is an induced polarization pattern that happens not to radiate: it is forced by the incident wave, does not satisfy boundary conditions on its own, and therefore radiates away the moment the incident field is removed. An embedded eigenstate, by contrast, is a genuine eigenmode with real eigenfrequency despite the open nature of the cavity; its internal field satisfies the continuity conditions with zero external field. The paper proves, via Lorentz reciprocity and a time-domain modal expansion, that for such an ideal embedded eigenstate the overlap integral between the nonradiating polarization current and any causal incident field vanishes, so the modal amplitude is $a_n=0$ for every excitation frequency, including at resonance. Thus an exact nonradiating eigenmode of a linear reciprocal open cavity cannot be excited from outside, and the energy stored in it under external causal illumination is identically zero.","pith_inferences":["If the orthogonality argument holds, reported 'excitations' of bound states in the continuum in open reciprocal systems are probably excitations of nearby high-Q states, not of the exact eigenstate; a direct test is to measure the dark time after switching off illumination.","The reciprocity argument suggests a design strategy: in nonreciprocal or time-modulated systems, the vanishing-overlap obstruction may be bypassed, potentially allowing direct external excitation of exact embedded eigenstates.","The same orthogonality integral could be used as an inverse-design objective, searching over permittivity profiles for field distributions whose overlap with all possible incident fields vanishes, creating cavities with arbitrarily long confinement without mirrors.","One could classify any proposed nonradiating mode operationally by its transient response: anapole-type states radiate a burst after the source turns off, whereas true eigenstates remain dark; this avoids relying on steady-state scattering measurements that cannot distinguish the two."],"forward_implications":["An anapole-based scatterer is invisible only under steady monochromatic illumination; when the incident field is switched off, the stored energy is released as radiation on a time scale set by the nearby complex eigenfrequencies.","At an exact embedded eigenstate, the incident field cannot deposit any energy into the mode; the modal amplitude $a_n$ and the stored modal energy are identically zero.","Near, but not at, an ideal embedded eigenstate, external excitation can produce very large internal fields with quality factors that diverge as the eigenstate is approached.","A laser based on the anapole cannot operate on the anapole itself, since lasing requires a pole of the scattering matrix; gain must instead bring one of the complex poles near the anapole onto the real axis.","The same classification applies to acoustic, elastic, and matter-wave scattering, since the argument uses only linear scattering theory and reciprocity."],"supporting_citations":[{"why":"Supplies the nonradiating-source orthogonality theorem that the paper generalizes; it is the logical core of the zero-excitation proof.","marker":"[28]"},{"why":"The recent claim that a radiationless anapole mode can be excited under reciprocity; the paper's target and the source of the specific sphere parameters and tailored excitation.","marker":"[12]"},{"why":"Provides the layered metallo-dielectric sphere that supports the embedded scattering eigenstate used as the nonradiating eigenmode example.","marker":"[20]"},{"why":"Supplies Mie theory, used to compute the scattering coefficient zeros and poles that distinguish anapoles from eigenmodes.","marker":"[29]"},{"why":"Defines bound states in the continuum, the class of radiationless eigenmodes the paper contrasts with anapoles.","marker":"[17]"},{"why":"Establishes completeness and orthogonality of quasinormal modes in open cavities, justifying the modal expansion used to derive the vanishing amplitude.","marker":"[31–38]"},{"why":"Shows embedded photonic eigenstates in zero-index metamaterials, supporting the claim that embedded eigenstates exist without external fields.","marker":"[21]"}],"fun_headline_variants":["Nonradiating eigenmodes resist external excitation","Lorentz reciprocity forbids exciting nonradiating eigenmodes","Anapoles are excitable; embedded eigenstates are not","Radiationless eigenmodes can't be driven from outside","Embedded eigenstates stay dark under external illumination"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof assumes the embedded eigenmode has a strictly confined induced polarization current inside a finite volume and exactly zero exterior field, so if any realistic candidate state has an exterior tail or surface current the orthogonality integral no longer vanishes and external excitation may become possible.","fun_headline_variants_meta":{"raw":{"variants":["Nonradiating eigenmodes resist external excitation","Lorentz reciprocity forbids exciting nonradiating eigenmodes","Anapoles are excitable; embedded eigenstates are not","Radiationless eigenmodes can't be driven from outside","Embedded eigenstates stay dark under external illumination"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000713,"raw_usage":{"total_tokens":3198,"prompt_tokens":926,"completion_tokens":2272,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":542,"completion_tokens_details":{"reasoning_tokens":2195}},"tokens_in":542,"tokens_out":2272,"duration_ms":15346,"temperature":1.0,"reasoning_tokens":2195,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:26:43.410500+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct an open reciprocal cavity whose purported embedded eigenstate is computed numerically, illuminate it with a time-domain plane-wave pulse tuned to the real eigenfrequency, and measure the field energy inside the cavity after the pulse has left: if the energy is nonzero and persists longer than the pulse, the zero-excitation theorem is violated. Alternatively, compute the overlap integral $\\int_{V_s} (\\varepsilon-\\varepsilon_0)\\mathbf E_{\\mathrm{int}}\\cdot \\mathbf E_{\\mathrm{inc}}\\,dV$ for a mode with a finite exterior tail and show it is nonzero while scattering remains zero.","supporting_citations":[{"cited_title":"Non-radiating monochromatic sources and their fields,","cited_arxiv_id":null,"evidence_quote":"Supplies the nonradiating-source orthogonality theorem that the paper generalizes; it is the logical core of the zero-excitation proof."},{"cited_title":"Excitation of the radiationless anapole mode,","cited_arxiv_id":null,"evidence_quote":"The recent claim that a radiationless anapole mode can be excited under reciprocity; the paper's target and the source of the specific sphere parameters and tailored excitation."},{"cited_title":"Embedded Photonic Eigenvalues in 3D Nanostructures,","cited_arxiv_id":null,"evidence_quote":"Provides the layered metallo-dielectric sphere that supports the embedded scattering eigenstate used as the nonradiating eigenmode example."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies Mie theory, used to compute the scattering coefficient zeros and poles that distinguish anapoles from eigenmodes."},{"cited_title":"Bound states in the continuum,","cited_arxiv_id":null,"evidence_quote":"Defines bound states in the continuum, the class of radiationless eigenmodes the paper contrasts with anapoles."},{"cited_title":"Trapping Light in Plain Sight: Embedded Photonic Eigenstates in Zero-Index Metamaterials,","cited_arxiv_id":null,"evidence_quote":"Shows embedded photonic eigenstates in zero-index metamaterials, supporting the claim that embedded eigenstates exist without external fields."}],"review_version":1}