{"id":"10f7a233-b384-47b0-83a4-98ac0627d989","arxiv_id":"2412.18253","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Direct experimental observation of linear light localization in two-dimensional photonic quasicrystals, with a localization threshold that decreases as the discrete rotational symmetry order increases.","lead":"This paper shows that light can be trapped, or localized, in clean photonic quasicrystals without any nonlinearity or disorder, provided the optical potential is deep enough. The required depth decreases as the rotational symmetry order of the quasicrystal grows.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Finite-window and finite-length effects may shift the reported localization thresholds; the sharp 'all eigenmodes extended below ELDT_N' claim needs a finite-size scaling check.","rationale":"The reader's weakest assumption is that the finite-window numerical spectrum and 2 cm propagation experiments capture the infinite-system physics. My concern is essentially the same, sharpened to the missing numerical-window details and the heuristic slope-change criterion for ELDT_N. This is load-bearing because the paper's central novelty is not just that light appears localized in quasicrystals, but that there is a sharp, symmetry-ordered localization-delocalization transition with a well-defined threshold below which all eigenmodes are extended. If the threshold is actually a finite-size crossover, the scaling trend with N may be an artifact of the chosen computational window, and the abstract's unqualified 'all eigenmodes extended' claim would be unsupported. The experiments alone cannot settle this because a 2 cm sample cannot distinguish slow diffraction from true localization. The proposed finite-size scaling test would directly check whether the thresholds converge as the system grows. Since this is the same condition the reader identified, and the paper's current evidence is insufficient to remove it, the appropriate verdict remains CONDITIONAL, so no change from the reader's verdict is needed.","tokens_in":10731,"tokens_out":6873,"duration_ms":68399,"concrete_test":"Re-compute the largest-β eigenmode for N=5 and N=8 at E0 values spanning the reported thresholds using finite-difference windows of linear size L = 40, 80, 160, 320, keeping the same k, A, φ, and boundary conditions as the paper, which must be stated explicitly. Plot χ_L(E0) for each L. If curves for different L cross at a common E0 and the below-threshold χ_L scales as 1/L, the transition is genuine; if the crossing shifts monotonically with L or below-threshold χ_L saturates at a window-independent value, the reported ELDT_N is a finite-size crossover and the 'all eigenmodes extended' claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that the slope change in the eigenmode form factor χ(E0) computed in a finite-difference window marks a true localization-delocalization transition of the infinite quasicrystal. The paper does not state the numerical window size, discretization, or boundary conditions for the eigenvalue problem, so finite-size effects are uncontrolled. In a finite window every eigenstate is normalizable, and the highest-β state transitions gradually from a 1/L-extended state to a localized state as E0 grows; the 'slope change' may occur where the localization length drops below the window, not where an eigenmode becomes truly square-integrable. The experimental support is a 2 cm propagation, which likewise cannot distinguish a slowly diffracting beam from a genuinely stationary mode. Thus the abstract's assertion that below ELDT_N all eigenmodes are extended is not established by the provided evidence. The symmetry-ordering of thresholds could survive as a statement about finite-sample diffraction suppression, but the sharp spectral-transition picture requires a finite-size scaling test.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Using optically induced photorefractive photonic quasicrystals formed by N interfering plane-wave pairs, the authors report experimental and numerical evidence that a probe beam remains localized on propagation when the lattice depth exceeds a threshold E_LDT^N, for N=5,7,8,9,10,11,12, while for the periodic N=6 lattice it never localizes. They find that the threshold decreases with N separately for odd and even symmetries and that central and off-center excitations give nearly the same threshold. They interpret a slope change in the numerically computed form factor of the highest-β eigenmode as a localization-delocalization transition, with all eigenmodes extended below threshold.","tokens_in":10923,"tokens_out":9024,"duration_ms":85730,"significance":"The experimental observation is significant: previous photonic-quasicrystal localization experiments involved nonlinearity or additional disorder, and the symmetry systematics reported here are new. The paper has a strong set of experimental controls: a periodic N=6 control, central versus off-center excitation, five repeated measurements at freshly rewritten lattice locations, low probe power, and the bias field turned off during probing. The central physical claim is plausible and likely reproducible. However, the sharp spectral-transition interpretation rests on a finite-window numerical eigenmode analysis whose finite-size control is not reported, and the 2 cm propagation length cannot by itself distinguish a stationary eigenmode from a very slowly diffracting wave packet.","major_comments":[{"comment":"The claim that at E0 < ELDT_N all eigenmodes of H are extended is not established by the numerical data shown. The paper does not report the finite-difference window size, grid spacing, or boundary conditions used to solve Hw = -beta w. Since every eigenstate of a finite-window discretization is normalizable, the slope change in chi(E0) used to define ELDT_N can occur when the localization length becomes comparable to the window rather than at a true spectral transition. Please add a finite-size scaling analysis (e.g., chi(E0; L) for increasing L at fixed discretization and boundary conditions) and show that below the quoted threshold chi tends to zero as L grows and that the extracted threshold converges. The 2 cm propagation experiments, while strongly suggestive, cannot by themselves distinguish a genuinely stationary mode from a wave packet whose diffraction length exceeds the sample; presenting output width or form factor versus propagation distance would help. In the absence of such tests, the abstract's assertion that all eigenmodes are extended below ELDT_N should be softened to a statement about the absence of localized modes in the probed windows.","section":"Fig. 2(a) and the paragraph 'The central result...' (pp. 6-7)"},{"comment":"There is an internal inconsistency in the description of the N=5 threshold. The text reports ELDT_5 approximately 300 V/mm and then states that Fig. 5(b), taken at E = 300 V/mm, is 'substantially lower than the critical field ELDT_5'. If the threshold is about 300 V/mm, the N=5 panel is at threshold, not clearly below it. Please give the precise threshold estimate with uncertainty and select a field that is unambiguously below the threshold, or relabel the panel as an at-threshold case, because this panel is used to illustrate the ordering of thresholds at fixed field.","section":"Experimental results section, p. 13 and Fig. 5(b)"},{"comment":"The numerical thresholds are reported as dimensionless E0 values (e.g., ELDT_5 about 3.6 and ELDT_8 about 4.1) while the experimental thresholds are reported as applied fields (about 300 and 360 V/mm), but the manuscript does not give the value of the transverse length unit D used to convert E0 to the physical applied field, nor the mapping between the numerical lattice parameters k=2, A^2=2.24 and the experimental writing-beam configuration. Without stating D and the conversion used to compare Fig. 2(a) with Fig. 5(a), the claimed agreement between the numerical and experimental threshold systematics cannot be independently checked.","section":"Methods and Eq. (1)"}],"minor_comments":[{"comment":"Please tabulate all measured ELDT_N values with uncertainties for N=5,7,8,9,10,11,12; the text quotes only N=5 and N=8, which makes the claimed rapid-decrease trend difficult to verify quantitatively from the prose alone.","section":"Fig. 5 and accompanying text"},{"comment":"The sentence 'Meantime, for quasicrystals belonging to dihedral groups of even and odd orders the critical depths are not strictly alternating, for example, ELDT_8 > ELDT_5 > ELDT_10 > ELDT_12' mixes even and odd orders; presenting the full ordered sequence of measured thresholds would make the separate odd/even monotonicity transparent.","section":"p. 7, discussion of ordering"},{"comment":"The asymptotic formula chi_as approximately -0.2138 + 0.3162 sqrt(E0) is described as 'well approximated'; please state whether the constants are obtained by fitting and over what E0 range, or provide a derivation from the limiting Bessel potential.","section":"Fig. 2(a) and asymptotic formula"},{"comment":"The statement that the translation stage allows recording of the intensity at every distance z inside the sample would be more convincing if at least one z-scan (output width or form factor versus propagation distance) were shown above and below the threshold; the main figures show only output distributions.","section":"p. 13, propagation-distance statement"},{"comment":"Although the displayed profiles cover -20 <= x,y <= 20, the text says the calculation window is 'much larger'; please state the actual window size, grid spacing, and boundary conditions used in the eigenvalue solver, as these are needed to reproduce the numerical thresholds.","section":"Fig. 2(b,c) and numerical details"}],"recommendation":"major_revision","confidential_remarks":"The experimental protocol and controls are convincing, and the paper is within the journal's scope. The central observation is likely correct, but the sharp localization-delocalization interpretation requires a finite-size scaling test and a softening of the 'all eigenmodes extended' claim. The requested revision is well-defined and does not require new physics. I do not see grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is the first clean experimental demonstration of linear (nonlinearity-free, disorder-free) localization of light in optically induced photonic quasicrystals, and it shows the threshold potential depth drops as the rotational symmetry order N increases. The experiment is carefully done: an N=6 periodic control stays diffractive, central and off-center launches give nearly the same threshold, each lattice is rewritten five times to fresh locations, and the probe power is low enough to avoid self-action. That core observation holds up.\n\nThe genuinely new result is the systematic symmetry dependence: for N=5,7,8,9,10,11,12 the critical field decreases with N within odd and even families, and the numerics and experiment agree on the ordering. The paper also honestly notes that the experimental form-factor saturation does not follow the numerical asymptotic curve and attributes that to finite refractive-index contrast and the onset of nonlinearity. The citation of prior work—nonlinear photonic quasicrystals [5], BEC localization [11], disorder-assisted transport [12]—is accurate and gives the right context for what is new.\n\nWhere the paper is soft is the sharp spectral-transition claim. The abstract says that below the threshold ELDT_N all eigenmodes of the Hamiltonian are extended. That statement outruns the evidence. The eigenvalue problem is solved in a finite-difference window, but the paper does not report the window size, discretization, or boundary conditions, and there is no finite-size scaling. In a finite window every eigenstate is normalizable, and the slope change in the form factor chi(E0) can occur where the localization length drops below the window rather than at a true localization-delocalization transition. The experiment, with 2 cm of propagation, likewise cannot distinguish a very slowly diffracting beam from a genuinely stationary mode. The symmetry ordering of thresholds would likely survive as a statement about finite-sample diffraction suppression, but the sharp all-extended-below-threshold picture needs a convergence check. The stress-test note is on target here.\n\nA smaller but real weakness: the data and code are only available \"upon reasonable request,\" not archived. For a flagship experimental claim that is a needless barrier.\n\nThis is not a fatal flaw in the experimental contribution. The paper is honest about the finite width of the transition and about the mismatch with the asymptotic curve, and the central observation is reproducible by the described methods. It deserves serious peer review. I would send it out and push for major revision: report the numerical parameters, add a finite-size scaling analysis, soften the abstract to what the finite-window and finite-length evidence actually supports, and put the data/code in a public repository. With those changes the paper would be solid.","headline":"A credible first observation of linear localization in photonic quasicrystals with a clean symmetry trend, but the sharp 'all modes extended below threshold' claim needs finite-size support.","tokens_in":11459,"tokens_out":1991,"would_cite":true,"duration_ms":20426,"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":"Clean linear photonic quasicrystals localize light above a symmetry-dependent potential depth, with higher rotational symmetry lowering the threshold.","keywords":["photonic quasicrystals","light localization","localization-delocalization transition","discrete rotational symmetry","photorefractive crystals","optical induction","aperiodic potentials","form factor"],"falsifier":"Grow the crystal (or enlarge the numerical window) at a fixed field just below a reported $E_{LDT,N}$ and measure the output form factor versus propagation length: if the form factor increases with length, or if a direct diagonalization on a larger supercell finds a localized eigenmode below threshold, the sharp transition claim would be falsified while the ordering of thresholds by $N$ could still hold.","tokens_in":10533,"feed_emoji":"💡","tokens_out":7665,"duration_ms":64975,"temperature":0.7,"pith_summary":"Light propagating through a two-dimensional quasicrystal can become trapped by the quasicrystal's own interference pattern alone, with no disorder and no nonlinearity required. The paper demonstrates this experimentally in photorefractive crystals, using a probe beam launched at the centre or off-centre, for quasicrystals with five-fold through twelve-fold rotational symmetries. In every non-crystallographic case there is a depth of the induced optical potential above which a localized mode appears and below which the beam diffracts. The critical depth falls rapidly as the order of the rotational symmetry grows, for odd and even symmetries separately, and no such transition occurs for the periodic six-fold lattice. The result settles a long-standing question because earlier photonic-quasicrystal experiments only saw localization with added nonlinearity or disorder.","feed_headline":"Light localizes in clean linear quasicrystals — no disorder needed","feed_subtitle":"Photonic quasicrystals with higher rotational symmetry need a shallower potential to switch from diffraction to a localized spot.","key_machinery":"The load-bearing object is the optically induced quasicrystalline potential in the paraxial Schrödinger equation (1), $H = -\\frac{1}{2}\\nabla^2 + \\frac{E_0}{1 + I_N(r)}$, where $I_N(r)$ is the intensity pattern from $N$ pairs of counterpropagating plane waves with wavevectors rotated by $2\\pi/N$. The potential has $N$-fold dihedral symmetry and no translational symmetry for $N$ other than 2, 3, 4, and 6. The argument is carried by the integral form factor $\\chi = \\left(\\iint |\\psi|^4 d^2r\\right)^{1/2}/U$, an inverse participation ratio that measures how tightly a mode is confined; the localization-delocalization transition is defined as the point where $\\chi(E_0)$ changes slope. The mechanism is interference, not defect trapping: the energy $-\\beta$ of the localized fundamental mode crosses several local potential minima, yet the mode stays confined near one maximum.","core_discovery":"The paper's central claim is that a clean linear two-dimensional quasicrystal supports localized eigenmodes once the depth of its optical potential exceeds a symmetry-dependent critical value, which the authors call the localization-delocalization transition point $E_{LDT,N}$. Below $E_{LDT,N}$ all eigenmodes of the Hamiltonian are extended; above it, at least the fundamental mode becomes spatially localized, with localization persisting for probe beams launched either at the rotational centre or at off-centre lattice maxima. The transition is diagnosed by a slope change in the form factor $\\chi(E_0)$, the inverse participation ratio of the most confined eigenmode. The threshold decreases rapidly with $N$ within each parity class ($E_{LDT,5} > E_{LDT,7} > \\cdots$ and $E_{LDT,8} > E_{LDT,10} > \\cdots$), while a periodic $N=6$ lattice shows no transition for any field, ruling out disorder or defects as the cause. For deep potentials, form factors of all symmetries collapse onto a common asymptote $\\chi_{\\rm as} \\approx -0.2138 + 0.3162\\sqrt{E_0}$; in the $N\\rightarrow\\infty$ limit the potential tends to a Bessel-function form $[A J_0(2r)]^2$ and the threshold approaches a finite minimum $E_0 \\approx 0.4572$. Experimentally, the thresholds for $N=5$ and $N=8$ are about 300 V/mm and 360 V/mm, and the measured dependence on $N$ matches the numerics.","pith_inferences":["If the trend continues, quasicrystals with very high $N$ would localize light at index contrasts far below the values needed for $N=5$, making low-power optical confinement practical in other aperiodic wave systems.","The same interference-induced mechanism could be transferred to acoustic, atomic, or polaritonic systems, where the symmetry order $N$ would become a design knob for the localization threshold.","The connection the authors draw between threshold and filling fraction suggests a quantitative predictor: measure the central-to-peripheral index deviation of any aperiodic potential and one may estimate whether it will localize at a given depth.","A direct extension would be to trace localized eigenmodes below the numerical threshold in larger supercells; if a localized state appears, the sharp transition picture would need revision, but the ordering of thresholds by $N$ could still hold."],"forward_implications":["Any clean linear quasicrystal with a non-crystallographic rotational symmetry can confine light above a threshold potential depth, for both central and off-centre excitation, without disorder or nonlinearity.","Higher rotational symmetry lowers the threshold potential depth, so high-order quasicrystals ($N=9,11,12$ and beyond) localize light at applied fields where lower-order ones still diffract.","The odd- and even-order dihedral groups behave as two separate sequences, so symmetry parity, not just $N$, enters the threshold scaling.","In deeper potentials localization becomes essentially independent of $N$, with the mode width set by the central index maximum; the $N\\rightarrow\\infty$ Bessel limit sets the lowest possible threshold for this family of potentials.","Periodic lattices ($N=6$) show no localization, confirming that the quasicrystal's aperiodicity, not fabrication defects, is responsible."],"supporting_citations":[{"why":"introduces quasicrystals as metallic phases with long-range orientational order and no translational symmetry, the class of structures studied here.","marker":"[1]"},{"why":"provides the optical-induction method in photorefractive crystals and the earlier observation that photonic-quasicrystal localization required nonlinearity.","marker":"[5]"},{"why":"predicted that the lowest electronic states in five-fold quasicrystals are localized, the theoretical antecedent for linear localization.","marker":"[10]"},{"why":"observed localization of a Bose-Einstein condensate in an eight-fold symmetric quasicrystal optical lattice, the closest prior experimental demonstration in a different system.","marker":"[11]"},{"why":"reported disorder-enhanced transport in photonic quasicrystals, representing the prior photonic experiments where localization required additional disorder.","marker":"[12]"},{"why":"predicted localization-delocalization wavepacket transitions in Pythagorean aperiodic potentials, a related disorder-free aperiodic system.","marker":"[27]"},{"why":"demonstrated localization and delocalization of light in photonic moiré lattices, establishing that incommensurate but non-quasicrystalline potentials can localize light.","marker":"[28]"},{"why":"shows that weakly coupled Schrödinger operators in two dimensions generally have bound states, the result the paper argues does not apply to the slowly decaying Bessel-potential limit.","marker":"[30]"}],"fun_headline_variants":["Light traps in clean photonic quasicrystals at critical depth","Symmetry sets the switch for light localization in quasicrystals","Photonic quasicrystals: deeper potential, tighter light","Critical depth dictates light localization in quasicrystals","Higher symmetry quasicrystals need less depth for light localization"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The sharp transition picture assumes that the finite 2 cm crystal and the finite numerical window represent an infinite quasicrystal: below the reported threshold every eigenmode stays extended, and the jump in the output form factor marks a true localization-delocalization transition rather than length-dependent diffraction suppression.","fun_headline_variants_meta":{"raw":{"variants":["Light traps in clean photonic quasicrystals at critical depth","Symmetry sets the switch for light localization in quasicrystals","Photonic quasicrystals: deeper potential, tighter light","Critical depth dictates light localization in quasicrystals","Higher symmetry quasicrystals need less depth for light localization"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000655,"raw_usage":{"total_tokens":3108,"prompt_tokens":1164,"completion_tokens":1944,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":780,"completion_tokens_details":{"reasoning_tokens":1863}},"tokens_in":780,"tokens_out":1944,"duration_ms":13467,"temperature":1.0,"reasoning_tokens":1863,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T04:52:34.320351+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Grow the crystal (or enlarge the numerical window) at a fixed field just below a reported $E_{LDT,N}$ and measure the output form factor versus propagation length: if the form factor increases with length, or if a direct diagonalization on a larger supercell finds a localized eigenmode below threshold, the sharp transition claim would be falsified while the ordering of thresholds by $N$ could still hold.","supporting_citations":[{"cited_title":"& Cahn, J","cited_arxiv_id":null,"evidence_quote":"introduces quasicrystals as metallic phases with long-range orientational order and no translational symmetry, the class of structures studied here."},{"cited_title":"Quasicrystals: What do we know? What do we want to know? What can we know? Acta Crystallogr","cited_arxiv_id":null,"evidence_quote":"provides the optical-induction method in photorefractive crystals and the earlier observation that photonic-quasicrystal localization required nonlinearity."},{"cited_title":"& Grynberg, G","cited_arxiv_id":null,"evidence_quote":"predicted that the lowest electronic states in five-fold quasicrystals are localized, the theoretical antecedent for linear localization."},{"cited_title":"& Santos, L","cited_arxiv_id":null,"evidence_quote":"observed localization of a Bose-Einstein condensate in an eight-fold symmetric quasicrystal optical lattice, the closest prior experimental demonstration in a different system."},{"cited_title":"& Sutherland, B","cited_arxiv_id":null,"evidence_quote":"reported disorder-enhanced transport in photonic quasicrystals, representing the prior photonic experiments where localization required additional disorder."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"predicted localization-delocalization wavepacket transitions in Pythagorean aperiodic potentials, a related disorder-free aperiodic system."},{"cited_title":"N., Anderson localization of light, Nat","cited_arxiv_id":null,"evidence_quote":"demonstrated localization and delocalization of light in photonic moiré lattices, establishing that incommensurate but non-quasicrystalline potentials can localize light."},{"cited_title":"V., Torner, L., Konotop V","cited_arxiv_id":null,"evidence_quote":"shows that weakly coupled Schrödinger operators in two dimensions generally have bound states, the result the paper argues does not apply to the slowly decaying Bessel-potential limit."}],"review_version":1}