{"id":"84019fec-aaa7-4ab2-b1c2-0db6eaaa1f1b","arxiv_id":"1908.04492","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Non-PT-symmetric photonic crystals with Dirac-like cone dispersions can exhibit real spectra and behave as complex conjugate media when the average non-Hermiticity in the unit cell is zero.","lead":"This paper shows that photonic crystals without parity-time symmetry can still have real frequency bands when the loss and gain in the unit cell average to zero for the relevant modes. The authors use this to realize a medium with a real refractive index but complex permittivity and permeability, which can produce lasing.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'always' universality claim hinges on a deferred proof that F_Omega,++ and F_Omega,-- always cross; the main text shows only one tuned geometry and assumes k-independent tau and kappa.","rationale":"The model construction is credible and partially independently supported: the pseudo-Hermiticity relation in Eq. (11) is explicit, Eq. (9) follows by direct diagonalization, and the COMSOL comparison in Fig. 2 validates the analytic dispersion for the chosen parameters. The EMT extraction and the S-matrix lasing predictions in Figs. 3-4 are internally consistent and offer concrete, checkable scattering singularities at d/a=17 and d/a=9. The soft spot is the generalization from one fitted example to the word 'always': the pseudo-Hermitian condition is imposed by tuning l_r at a single k-point where F_A,++ = F_A,--, and the existence of such a crossing for all Dirac-like-cone PCs is asserted with proof deferred to the supplementary notes. This is precisely the weakest assumption the reader identified, so the CONDITIONAL verdict is appropriate. I do not see a demonstrated internal inconsistency, and the numerical evidence is strong enough that rejection would be too harsh; the concern is missing or deferred support for the universality claim, not a proven counterexample.","tokens_in":10365,"tokens_out":10751,"duration_ms":119246,"concrete_test":"Independently verify the deferred crossing proof by scanning a family of two-component square-lattice PCs that each host an accidental monopole/dipole Dirac-like cone (e.g., vary rc over 0.18a-0.22a and epsilon_A over 10-15, retuning for degeneracy) and computing F_A,++(k) - F_A,--(k) from the Hermitian eigenmodes; check whether a sign change occurs for every member before the EP-ring radius. In the same calculation, compare the 2x2 Hamiltonian eigenvalues using the actual k-dependent tau_m(k) and kappa(k) against the constant-parameter prediction of Eq. (9) at gamma=0.367, and record whether the real-spectrum boundary shifts by more than a few percent of kc.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that any Dirac-like-cone PC can be made pseudo-Hermitian with real spectra rests on the assertion in Section III A that F_Omega,++ and F_Omega,-- necessarily cross as k moves away from Gamma, with the proof deferred to Supplementary Note 6. The main text demonstrates a single intersection at kxa/2pi=0.019 for one geometry (rc=0.1999a, epsilon_A=12.5) and then treats Eq. (12) as a universal construction. Since Eq. (12) requires F_A,++ = F_A,-- for both bands simultaneously, the pseudo-Hermitian Hamiltonian (7) and eigenvalue formula (9) both depend on that crossing being generic. The manuscript also assumes tau_m and kappa are k-independent near the Dirac-like cone, calling them 'almost constant' in Fig. 1B; however, the crossing point lies inside the EP ring (kc=0.019), tau_m is not exactly zero at Gamma, and no error bound is given for the constant-parameter approximation. If the deferred band-inversion proof fails for some Dirac-like-cone PCs, or if k-dependence of tau_m or kappa shifts the real-spectrum region materially, the 'always' claim reduces to the single fitted example.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies two-dimensional photonic crystals (PCs) with a loss-gain distribution that is not PT-symmetric. The authors build a two-band non-Hermitian Hamiltonian for TM polarization near a Dirac-like cone, show that when the average non-Hermiticity in the unit cell vanishes the Hamiltonian becomes pseudo-Hermitian, and derive a real band spectrum outside a ring of exceptional points. They then extract effective permittivity and permeability from the real bands, finding a complex conjugate medium with real refractive index, and demonstrate S-matrix poles (lasing) for a slab of this effective medium, with COMSOL simulations for one geometry. The central assertion is that such real spectra can always be obtained in two-component PCs carrying Dirac-like cones formed by monopolar and dipolar resonances, because the overlap functions F_Omega,++, F_Omega,-- necessarily cross near the Gamma point, enabling a real loss-gain ratio l_r through Eq. (12).","tokens_in":10628,"tokens_out":5211,"duration_ms":55022,"significance":"If the universality claim holds, the paper offers a conceptually new route to real spectra in non-Hermitian photonic systems beyond PT symmetry and connects it to an effective complex-conjugate medium with real refractive index. The analytical Hamiltonian model and the boundary-field-averaging effective medium calculation agree with full-wave COMSOL results for the presented structure, and the reproduction of S-matrix poles by the effective slab is a clean demonstration. The connection between pseudo-Hermiticity, Dirac-like cones, and CCM behavior is potentially useful for designing gain-loss photonic devices. However, the paper's main theoretical novelty is the 'always' statement, and that statement is the part that is least supported in the main text.","major_comments":[{"comment":"The universality claim rests on the assertion that F_Omega,++ and F_Omega,-- necessarily cross as k moves away from the Gamma point for any two-component Dirac-like cone PC whose bands arise from monopolar/dipolar resonances. The main text shows a single crossing for rc=0.1999a and epsilon_A=12.5 at kxa/2pi=0.019, and the general proof is deferred to Supplementary Note 6. Since Eq. (12) and the pseudo-Hermitian Hamiltonian (7) require this crossing to realize tau_plus=tau_minus=0 simultaneously for both bands, the 'always' claim is load-bearing. The main text should present the proof or at least the precise hypotheses of the supplementary proof, and an independent numerical example with different material parameters would substantially strengthen the claim.","section":"Section III A, Eq. (12) and paragraph after Fig. 1B"},{"comment":"The condition for a single real l_r to make both tau_+ and tau_- vanish is F_A,++/F_B,++ = F_A,--/F_B,--, which is a ratio equality. The text states that 'we must tune the system parameters so that the system satisfies F_Omega,++ = F_Omega,--' and then uses Eq. (12). Equality of the F's is sufficient (given the orthonormality relation, Eq. (3)) but is not necessary. The manuscript should clarify whether the supplementary proof establishes the stronger equality or only the ratio condition, because the universality statement concerns whichever condition is actually required.","section":"Section III A, Eq. (12)"},{"comment":"The analytical band structure Eq. (9) is derived under the exact conditions tau_plus=tau_minus=0 and k-independent tau_m and kappa. In the actual system l_r is fixed at the crossing point, so at k values away from that point tau_plus and tau_minus are only approximately zero; the text calls them 'almost constant' and 'approximately fulfilled' without a quantitative error bound. The COMSOL agreement in Fig. 2 for gamma=+0.367 supports the approximation for this one structure, but the k-independence and the small-tau assumption are assumptions, not derived results. Please state the range of k and gamma over which the approximation is expected to hold and, ideally, provide an error estimate for the real parts of the eigenfrequencies.","section":"Section III A, paragraph after Fig. 1C; Eq. (9)"},{"comment":"The abstract's claim that such PCs 'can always exhibit real spectra as long as the average non-Hermiticity strength... is zero' is conditional: the real spectra follow by construction once l_r is chosen from Eq. (12) to force tau_plus=tau_minus=0. The nontrivial assertion is the existence of a real l_r that realizes this condition for any Dirac-like cone PC of the stated class. The manuscript should be explicit that the prediction is the existence of this parameter and the occurrence of real spectra in a finite k-region, rather than real spectra appearing without parameter tuning. As written, a reader could mistake a designed condition for an emergent phenomenon.","section":"Abstract and Section IV"}],"minor_comments":[{"comment":"The vertical dashed line in Fig. 1C is described as denoting the intersection F_Omega,++ = F_Omega,--, but the figure shows the average non-Hermiticity tau_m; please clarify on the figure itself which quantity crosses and at which k value, and define the domain Omega in the caption.","section":"Fig. 1 caption and Section III A"},{"comment":"There is a typo 'tunning' in Section III A near 'by tunning l_r'; it should read 'tuning'.","section":"Throughout"},{"comment":"The definition of H as H_1^{-1} H_2 and the role of the scalar beta would be clearer if the text explicitly noted that beta is complex in general but positive real when tau_plus=tau_minus=0, since the pseudo-Hermitian discussion uses that case.","section":"Section II, Eq. (4)"},{"comment":"In the caption of Fig. 4, the labels for panels (D) and (F) are correct, but the main text refers to 'Figure 4B' for the PC slab schematic while Fig. 4B is also used for the EM/PC transmission and reflection plot; please renumber or refer to the schematic explicitly to avoid ambiguity.","section":"Fig. 4 caption"}],"recommendation":"major_revision","confidential_remarks":"The main text repeatedly defers the key proof of the universality claim to Supplementary Note 6, but the supplement was not available for the review. If the supplement contains a rigorous proof of the crossing of F_Omega,++ and F_Omega,-- for the full class of Dirac-like cone PCs, then the major concerns are largely presentational and could be resolved in a revision. If the supplement only provides a heuristic argument or an additional example, the abstract's 'always' claim should be weakened. The paper is otherwise a sound and interesting study of a single, carefully analyzed system, and the COMSOL-validation part is convincing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper earns its keep as a design recipe, not as a universality proof. The two-band Hamiltonian is clean, the COMSOL agreement is convincing, and the effective-medium prediction of lasing is a nice payoff. The abstract's 'always' claim is the soft spot: it rests on a band-inversion crossing argument deferred to the supplement, and the main text shows one geometry, one gamma, and assumes k-independent tau and kappa without error bounds.\n\nWhat's actually new: prior work had real spectra in non-PT potentials and EP rings from Dirac cones, but the explicit route here—enforce tau_plus = tau_minus = 0 by adjusting the loss-gain ratio l_r, yielding a pseudo-Hermitian Hamiltonian with an EP ring and real bands outside—is a new and useful application. The effective medium analysis is well done: the complex-valued k(omega) band structure, the boundary-averaging retrieval, and the slab scattering all line up, including the S-matrix pole that corresponds to lasing. That's real evidence.\n\nWhere it's soft: the universality claim is the weakest link. Equation (12) determines l_r from F_A,++ = F_A,--, and the paper asserts this crossing always happens for Dirac-like-cone PCs due to band inversion, pointing to Supplementary Note 6. If that proof is solid, the architecture is general; if not, it's a single tuned example. The k-independence of tau_m and kappa is taken as 'almost constant' with no error estimate. These are fixable in revision—show the crossing for a second geometry or parameter set, and give a bound on the k variation. The central trick is admittedly designed (l_r is solved to impose tau = 0), but that is a feature, not a flaw: it is a construction recipe, and the lasing prediction validates the machinery. I would not call it circular in any damaging sense.\n\nBottom line: this is a competent, well-written paper that belongs in the literature. A serious referee should have the supplement in hand to check the crossing proof and the k-independence; with those confirmed, the paper's claims stand mostly as stated, with the word 'always' needing moderation. I'd send it to review.","headline":"A clean design recipe for real-spectra non-PT photonic crystals, but the 'always' universality claim is broader than the evidence shown.","tokens_in":11163,"tokens_out":2535,"would_cite":true,"duration_ms":23533,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A photonic crystal without PT symmetry can have real bands and act as a complex conjugate medium.","keywords":["non-Hermitian photonics","pseudo-Hermiticity","Dirac-like cone","photonic crystals","exceptional points","complex conjugate medium","effective medium theory","lasing"],"falsifier":"Compute $F_{\\Omega,++}-F_{\\Omega,--}$ along a $k$-line away from $\\Gamma$ for a family of two-component photonic crystals that all have monopole/dipole Dirac-like cones, varying filling fraction and permittivity contrast; a crystal in which this difference never changes sign breaks the 'always' claim. Alternatively, realize the rod-in-air crystal with $\\ell_r=-0.15235$ and $\\gamma=0.367$ and look for the predicted real band and the lasing singularity at $\\omega a/2\\pi c=0.5416$ with $d/a=17$; if transmission and reflection stay finite while the effective parameters match, the complex-conjugate-medium mapping is wrong.","tokens_in":10122,"feed_emoji":"💡","tokens_out":12532,"duration_ms":113639,"temperature":0.7,"pith_summary":"This paper shows that a non-Hermitian photonic crystal need not be PT-symmetric to have real frequency bands. In a two-component crystal whose band structure has a Dirac-like cone at the Brillouin-zone center, one can choose a single loss-gain ratio so that the average non-Hermiticity of each participating eigenstate inside the unit cell is zero; the two-band Hamiltonian then becomes pseudo-Hermitian and its eigenvalues are real outside a ring of exceptional points. In the long-wavelength limit the crystal is described as a complex conjugate medium: the effective refractive index is real, while the effective permittivity and permeability are complex numbers. The same real bands can produce $S$-matrix poles, i.e. a lasing effect, at particular slab thicknesses, and flipping the sign of loss and gain changes the effective parameters by complex conjugation, leading to different scattering behavior.","feed_headline":"Real spectra need no PT symmetry in photonic crystals","feed_subtitle":"Loss and gain balanced in a Dirac-like cone give a real refractive index with complex permittivity and permeability.","key_machinery":"The load-bearing object is the two-band non-Hermitian Hamiltonian built from the Hermitian Bloch modes of the photonic crystal. The odd flat band decouples from the two even bands by symmetry, leaving a $2\\times2$ matrix whose diagonal entries carry $\\tau_\\pm$ and whose off-diagonal entries carry $\\kappa$. The condition $\\tau_\\pm=0$ makes the prefactor $\\beta$ real and renders the Hamiltonian pseudo-Hermitian with $\\eta H(\\gamma)\\eta^{-1}=H(\\gamma)^\\dagger=H(-\\gamma)$, where $\\eta=\\mathrm{diag}(W_-^{(0)},-W_+^{(0)})$. The Dirac-like cone does the constructive work: band inversion is claimed to force the overlap functions $F_{\\Omega,++}$ and $F_{\\Omega,--}$ to cross away from $\\Gamma$, so Eq. (12) fixes a real loss-gain ratio $\\ell_r$; the eigenvalues then take the square-root form whose branch cut is a ring of exceptional points, with real spectrum outside the ring.","core_discovery":"On the paper's own terms, the central discovery is that pseudo-Hermiticity, not PT symmetry, is the operative condition for real spectra in a non-Hermitian photonic crystal. Expanding the non-Hermitian operator in the basis of Hermitian Bloch modes and keeping only the two linearly dispersive bands gives a $2\\times2$ Hamiltonian whose diagonal average non-Hermiticity $\\tau_\\pm$ and off-diagonal overlap $\\kappa$ control the spectrum. Setting $\\tau_\\pm=0$ makes the Hamiltonian pseudo-Hermitian, and the eigenvalues are $W_\\pm=(1/\\beta)(W_d\\pm\\sqrt{(C_gk)^2(1+\\gamma^2|\\kappa|^2)-\\gamma^2|\\kappa|^2W_d^2})$, which are real outside a ring of exceptional points of radius $k_c=k_b(1+|\\gamma\\kappa|^{-2})^{-1/2}\\le k_b$. For a rod-type square-lattice photonic crystal with rod permittivity $12.5+i\\gamma$ in air with permittivity $1+i\\ell_r\\gamma$, choosing $\\ell_r=-0.15235$ realizes $\\tau_\\pm\\approx0$ near the Dirac-like cone, and the numerically computed bands match the model. The retrieved effective medium has $n_e^2=\\varepsilon_e\\mu_e$ real while $\\varepsilon_e$ and $\\mu_e$ are complex, and a slab of this medium reproduces the lasing poles expected of a complex conjugate medium.","pith_inferences":["If the overlap-function crossing is genuinely forced by band inversion for every monopole/dipole Dirac-like cone, the recipe would work across many lattice geometries and filling ratios; the main text states this and leaves the full proof to the supplement.","The same pseudo-Hermitian construction should carry over to acoustic or elastic metamaterials with monopole/dipole Dirac-like cones, since the argument rests on mode symmetry and the zero-average-non-Hermiticity condition rather than on the Maxwell equations specifically.","A clean experimental check would be to measure the lasing peak at the predicted frequency and slab thickness, or to map the complex band structure by angle-resolved reflectance; success would validate the effective complex-conjugate-medium picture without requiring a direct extraction of $\\varepsilon_e$ and $\\mu_e$."],"forward_implications":["A two-component photonic crystal with a Dirac-like cone formed by monopolar and dipolar modes has a real spectral region once the loss-gain ratio is tuned so that $\\tau_\\pm=0$, and this does not require PT symmetry in space.","In the long-wavelength limit the crystal behaves as a complex conjugate medium with real $n_e$ and complex $\\varepsilon_e,\\mu_e$; the real refractive index is guaranteed by the real spectrum rather than by simple loss-gain compensation.","A slab of thickness $d/a=17$ for $\\gamma_+$ (or $d/a=9$ for $\\gamma_-$) shows transmission and reflection singularities at $\\omega a/2\\pi c=0.5416$, which is the predicted lasing signature.","The conjugate-paired systems $\\gamma_+$ and $\\gamma_-$ share nearly identical real bands but have complex-conjugate effective parameters, so their scattering, impedance and lensing behavior differ even though $n_e$ is the same."],"supporting_citations":[{"why":"Establishes pseudo-Hermiticity as the necessary condition for a real spectrum, the criterion the paper exploits.","marker":"13"},{"why":"Supplies the Dirac-like-cone photonic crystal with zero-index behavior that serves as the starting Hermitian system.","marker":"25"},{"why":"Provides the generalized-eigenvalue formulation of non-Hermitian photonic crystals from which the two-band Hamiltonian is derived.","marker":"28"},{"why":"Shows that open Dirac cones spawn rings of exceptional points, the spectral feature the model places inside a real-band region.","marker":"26"},{"why":"Fixes the even/odd mode symmetries that decouple the flat band and justify the $2\\times2$ model near the zone center.","marker":"29"},{"why":"Gives the Bloch-mode field-averaging method used to retrieve effective permittivity and permeability from the computed bands.","marker":"31"},{"why":"Supplies the scattering-matrix and pole analysis used to identify the lasing condition for the effective medium.","marker":"23"},{"why":"Defines complex conjugate media, the effective-medium class the paper realizes.","marker":"20"}],"fun_headline_variants":["Pseudo-Hermiticity gives real spectra in lossy photonic crystals","Non-PT crystals with balanced loss show real refractive index","Complex conjugate medium from non-PT photonic crystals","Real spectra without PT symmetry via pseudo-Hermiticity","Gain-loss balance in crystals yields real spectra, no PT needed"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The broad 'always' conclusion depends on the assumption that for every two-component photonic crystal whose Dirac-like cone comes from monopolar and dipolar resonances, the overlap functions $F_{\\Omega,++}$ and $F_{\\Omega,--}$ cross somewhere away from the $\\Gamma$ point so that a single real loss-gain ratio makes $\\tau_\\pm=0$; the paper says this follows from band inversion and defers the proof to a supplement.","fun_headline_variants_meta":{"raw":{"variants":["Pseudo-Hermiticity gives real spectra in lossy photonic crystals","Non-PT crystals with balanced loss show real refractive index","Complex conjugate medium from non-PT photonic crystals","Real spectra without PT symmetry via pseudo-Hermiticity","Gain-loss balance in crystals yields real spectra, no PT needed"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000292,"raw_usage":{"total_tokens":1758,"prompt_tokens":1057,"completion_tokens":701,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":673,"completion_tokens_details":{"reasoning_tokens":615}},"tokens_in":673,"tokens_out":701,"duration_ms":6630,"temperature":1.0,"reasoning_tokens":615,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:40:50.366068+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $F_{\\Omega,++}-F_{\\Omega,--}$ along a $k$-line away from $\\Gamma$ for a family of two-component photonic crystals that all have monopole/dipole Dirac-like cones, varying filling fraction and permittivity contrast; a crystal in which this difference never changes sign breaks the 'always' claim. Alternatively, realize the rod-in-air crystal with $\\ell_r=-0.15235$ and $\\gamma=0.367$ and look for the predicted real band and the lasing singularity at $\\omega a/2\\pi c=0.5416$ with $d/a=17$; if transmission and reflection stay finite while the effective parameters match, the complex-conjugate-medium mapping is wrong.","supporting_citations":[{"cited_title":"Christodoulides, and Ulf Peschel","cited_arxiv_id":null,"evidence_quote":"Establishes pseudo-Hermiticity as the necessary condition for a real spectrum, the criterion the paper exploits."},{"cited_title":"Basiri, I","cited_arxiv_id":null,"evidence_quote":"Supplies the Dirac-like-cone photonic crystal with zero-index behavior that serves as the starting Hermitian system."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the generalized-eigenvalue formulation of non-Hermitian photonic crystals from which the two-band Hamiltonian is derived."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that open Dirac cones spawn rings of exceptional points, the spectral feature the model places inside a real-band region."},{"cited_title":"Joannopoulos, and Marin Soljačić","cited_arxiv_id":null,"evidence_quote":"Fixes the even/odd mode symmetries that decouple the flat band and justify the $2\\times2$ model near the zone center."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the Bloch-mode field-averaging method used to retrieve effective permittivity and permeability from the computed bands."},{"cited_title":"Dragoman","cited_arxiv_id":null,"evidence_quote":"Supplies the scattering-matrix and pole analysis used to identify the lasing condition for the effective medium."},{"cited_title":"All-real spectra in optical systems with arbitrary gain-and-loss distributions","cited_arxiv_id":null,"evidence_quote":"Defines complex conjugate media, the effective-medium class the paper realizes."}],"review_version":1}