{"id":"459de9a1-0274-4e74-ad4f-f383b4c150a0","arxiv_id":"2412.08974","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A Fourier-space simulated-annealing framework constructs 3D two-phase media with prescribed spectral densities, yielding the first 3D antihyperuniform realizations and estimates of their spreadability and permeability.","lead":"Researchers built 3D digital materials with precisely controlled disorder, including the first 3D examples of a rare antihyperuniform class with wildly mixed cluster sizes. The work offers a design route for composites with tailored diffusion and fluid flow behavior.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim of first 3D antihyperuniform media is unverified because the realized arrays' spectral density and local volume-fraction variance are never measured; finite-L matching of a diverging target may not produce the required R^{-2} scaling.","rationale":"The reader's weakest assumption and my stress-test converge on the same gap: the paper asserts realization of antihyperuniform 3D media but provides no direct measurement of the constructed arrays' spectral density or local volume-fraction variance. This is load-bearing because antihyperuniformity is defined by the scaling of sigma_V^2(R) and by a diverging spectral density at k=0, neither of which can be inferred solely from convergence of a finite set of constrained k-values in a periodic box. The finite-L issue is concrete: at k=0 the spectral density of a fixed-volume-fraction periodic binary array is identically zero, and the smallest nonzero k is 2*pi/L, so the 1/k divergence can only be probed through the scaling trend across available low-k modes and through the real-space variance scaling. Without that verification, the central novelty claim is conditional at best. I do not see an internal inconsistency that would justify rejection; the method is a plausible extension of established 2D constructions, and the visual realizations are suggestive. The typographical issues are real but do not undermine the framework once corrected. Therefore I recommend keeping the reader's CONDITIONAL verdict: the concern is addressable by a straightforward numerical check, but until that check is reported, the strongest claim should not be taken as established.","tokens_in":23471,"tokens_out":3778,"duration_ms":40951,"concrete_test":"Compute the actual angular-averaged spectral density tilde-chi_V(k) of each final binary array from Eq. (29) by FFT, independently of the annealing objective, and compare with the target tilde-chi_V^0(k) for all k, not just constrained k < K*. Then compute sigma_V^2(R) over spherical windows for R up to L/2 for the L=128 arrays and repeat for L=256; fit the large-R tail. For the antihyperuniform class, the realized media must show sigma_V^2(R) decaying as R^{-2} (slower than R^{-3}), and the measured tilde-chi_V(k) must grow toward small k following ~k^{-1} as far as the finite-size k=2*pi/L allows. If either test fails, the construction does not realize antihyperuniform media and the transport results should be recomputed from measured spectra.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on the assumption that the finite L=128 voxel arrays produced by the simulated-annealing algorithm actually realize the target spectral density, especially its small-k divergence. The paper never measures the spectral density of the constructed arrays nor the local volume-fraction variance sigma_V^2(R), which is the defining diagnostic of antihyperuniformity (Eq. 17). This matters because the antihyperuniform target is singular as k->0 (Eq. 47, with tilde-chi_V ~ 2*pi^2/k), while any periodic binary array has exactly zero spectral density at k=0 and only discrete k values down to 2*pi/L. Matching constraints at those discrete points does not guarantee the infinite-volume limit has a diverging spectral density; the inverse problem is underdetermined (N_Omega < N), so many arrays can satisfy the finite constraint set without exhibiting the required R^{-(d+alpha)} variance scaling. A typo in Eq. (48) (Si defined with cos rather than sin) and the apparent outlier in Table I (0.09539 versus 0.9603/0.9218) are secondary, but the missing direct verification is the load-bearing gap: if the realized arrays do not show sigma_V^2(R) ~ R^{-2} (for alpha=-1 in d=3), the 'first realization' claim is not established, and transport properties computed from the target functions describe target models, not the constructed media.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a Fourier-space simulated-annealing method for constructing three-dimensional, statistically isotropic two-phase random media from prescribed analytical spectral density functions. The target functions cover Debye random media (Eqs. 39-40), standard hyperuniform media (Eqs. 42-43), stealthy hyperuniform media (Eq. 45), and antihyperuniform media with a power-law autocovariance (Eqs. 46-47), for volume fractions 0.25, 0.5, and 0.75 and correlation lengths a=5, 10, and 25 voxels. It claims the first realization of antihyperuniform two-phase media in 3D, supported by visual renderings showing clusters of very different sizes. It then computes the diffusion spreadability from Eq. (20) and estimates fluid permeability from Eq. (28) directly from the target spectral density functions, reporting that varying the length-scale parameter produces orders-of-magnitude changes in spreadability and that antihyperuniform media have the largest dimensionless permeability.","tokens_in":23739,"tokens_out":9001,"duration_ms":97092,"significance":"If fully verified, the framework would usefully extend the Yeong-Torquato Fourier-space construction approach to 3D and provide a parameterized route to microstructures across all hyperuniformity classes, including a candidate antihyperuniform medium. The computational strategy is efficient, the phase-inversion symmetry of the constructed media is a sensible consistency check, and the reported long-time spreadability scalings agree with known theory. However, the central advertised novelty, the first 3D antihyperuniform two-phase media, is not yet established because the paper never measures the spectral density or local volume-fraction variance of the constructed arrays; the transport results are also properties of the target functions rather than of the constructed realizations. These gaps are fixable and do not invalidate the method, but they must be addressed before the central claims can be accepted.","major_comments":[{"comment":"The constructed media are never quantitatively verified against their targets. The paper reports no measured angular-averaged spectral density of the final arrays and no local volume-fraction variance sigma_V^2(R). For the antihyperuniform claim, Eq. (17) requires sigma_V^2(R) ~ R^{-(d+alpha)} = R^{-2} in d=3 for alpha=-1, and Eq. (50) requires a diverging spectral density ~1/k. A finite periodic binary array has exactly zero spectral density at k=0 and only discrete wavevectors down to 2*pi/L, so matching the target at the constrained shells in Eq. (35) does not by itself establish the infinite-volume divergence. Please show measured tilde-chi_V(k) from the final arrays for all four classes, and sigma_V^2(R) scaling for the antihyperuniform case, including at least one larger system size.","section":"Sec. IV, Figs. 1-8; Eq. (17)"},{"comment":"The definitions of Ci and Si in Eq. (48) are incorrect. As printed, Ci(x) = integral_0^x cos(t)/t dt and Si(x) = integral_0^x cos(t)/t dt are identical, and the Ci integrand is not integrable at t=0; Si should contain sin(t), and Ci is conventionally defined with the lower limit at infinity. Since Eq. (47) is the target spectral density for the central antihyperuniform construction, the formula must be corrected and the construction checked against the corrected expression. If the code actually used standard definitions, this should be stated explicitly.","section":"Eq. (48)"},{"comment":"The spreadability and permeability results are evaluated directly from the parameterized target functions, not from the constructed voxel arrays. Consequently, the observed trends, including the orders-of-magnitude variation of S(t) with a and the permeability ranking among the four classes, are inherited from the input spectral densities and do not test the construction. The abstract and title present these as properties of the constructed media. Either compute S(t) and permeability (or at least l_p^2) from the realizations themselves, or explicitly reframe Section V as analytical predictions for the target spectral-density models rather than for the constructed microstructures.","section":"Sec. V, Eqs. (20) and (28)"},{"comment":"The underdetermined nature of the inverse problem is acknowledged in Sec. III.A (N_Omega < N, no unique solution to Eq. (33)), and Sec. II.A notes that sufficient realizability conditions for autocovariance functions are an open problem. These limitations are particularly important for the singular antihyperuniform target: many binary arrays can satisfy the finite set of discrete-k constraints without exhibiting the required R^{-2} variance scaling. The paper states that L=64 and L=256 were investigated to verify resolution effects, but no results are shown. Please include this finite-size convergence data and the variance scaling evidence.","section":"Sec. III.A; Sec. II.A"}],"minor_comments":[{"comment":"The entry 0.09539 for the standard hyperuniform medium at phi1=0.5 and a=10 is an apparent outlier relative to the neighboring entries 0.9603 (a=5) and 0.9218 (a=25); please check this value and the associated calculation.","section":"Table I"},{"comment":"The phrase 'Interesting, we note' should read 'Interestingly, we note.'","section":"Sec. IV.A"},{"comment":"The sentence describing 'density fluctuations that strongly suppress scattering' appears to be a typo: for antihyperuniform media the zero-wavenumber scattering diverges, so the fluctuations should enhance, not suppress, scattering.","section":"Sec. IV.D, last paragraph"},{"comment":"Please specify whether the horizontal axis is time t or the dimensionless Dt/a^2. If Dt/a^2 is used, the claim of orders-of-magnitude variation with a needs to be justified beyond the explicit a^2 rescaling in Eq. (22).","section":"Fig. 9"},{"comment":"The text and Eq. (32) refer to 'pixel' and a 'square pixel,' but the construction is three-dimensional; the terminology should be voxel and cubic voxel.","section":"Sec. III.A, Eq. (32)"}],"recommendation":"major_revision","confidential_remarks":"The reader's conditional verdict and the stress-test concern are, in my reading, valid: the main gap is the absence of any direct measurement of the realized spectral density or local volume-fraction variance, which is precisely what is needed to support the 'first realization of antihyperuniform media' claim. The Eq. (48) typo is a concrete, fixable error in the central antihyperuniform formula. I do not see grounds for rejection; the missing verification and the reframing of the transport section are within the scope of a major revision. The bibliography and citation pattern appear appropriate, and I see no novelty-disclosure concern."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid, useful extension of the Fourier-space simulated-annealing construction from 2D (Refs. 64, 65) to 3D, and it delivers the first 3D realizations of several targeted spectral densities, including antihyperuniform media. The core claim is plausible but not yet quantitatively verified. The transport results are mostly evaluations of the target functions, not of the reconstructed media.\n\nWhat's good: the framework is clean, the parameter sweep across hyperuniformity classes, volume fractions, and correlation lengths is systematic, and the visual morphologies look consistent with the intended classes. The antihyperuniform construction is genuinely new as far as I know. The paper is also appropriately careful in saying the target autocovariances satisfy known necessary conditions, with sufficiency open.\n\nSoft spots: the main one is the missing direct validation of the realized arrays. The energy E in Eqs. (34)-(35) is driven below 1e-12, so the angular-averaged spectral density does match the target at the constrained discrete k points. But the paper never shows that match, and more importantly never reports the local volume-fraction variance sigma_V^2(R) for the constructed media. For the antihyperuniform claim this is the defining diagnostic: the target diverges as k^{-1}, and a finite L=128 periodic array can match a large finite value at the smallest discrete k without exhibiting the required R^{-2} variance scaling in the thermodynamic limit. So the \"first realization\" claim needs that check. Second, the spreadability and permeability are computed from the analytic target functions, not from the voxel arrays. That is legitimate if the reconstruction is faithful, but it means the observed transport trends are inherited from the input functions rather than emergent. Third, there is a typo in Eq. (48) (Si is defined with cos, should be sin), and the Table I entry 0.09539 for standard hyperuniform at phi1=0.5, a=10 is an outlier against the neighboring entries and breaks the stated monotonic trend (likely a typo).\n\nNone of this is fatal. The construction method is established, the targets are sensible, and the missing pieces are straightforward to supply. The paper deserves a serious referee, but I would send it back for major revision: add a figure showing measured vs target spectral density, compute sigma_V^2(R) for all four classes with emphasis on the antihyperuniform systems, and fix the typos and the outlier. For a reader working on microstructure reconstruction or hyperuniform media, this paper is worth reading.","headline":"Plausible 3D extension of the group's 2D spectral-density construction, with a first-claim for antihyperuniform media that needs direct variance-scaling evidence before it is fully established.","tokens_in":24291,"tokens_out":3829,"would_cite":true,"duration_ms":39165,"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":"The paper reports the first explicit construction of three-dimensional antihyperuniform two-phase media, together with hyperuniform and nonhyperuniform realizations, all generated from prescribed spectral density functions whose transport…","keywords":["hyperuniformity","antihyperuniform media","spectral density function","microstructure reconstruction","diffusion spreadability","fluid permeability","two-phase random media","simulated annealing"],"falsifier":"Measure the spectral density of the constructed antihyperuniform samples at the smallest resolved wavenumbers, or compute the local volume-fraction variance in large spherical windows, and check for the predicted $\\tilde{\\chi}_V(k)\\sim 2\\pi^2/k$ divergence and for a variance decay slower than $R^{-3}$; if the small-$k$ spectrum saturates to a finite value because of the $L=128$ cutoff, the claimed first realization of true antihyperuniform media in 3D would not be supported by the constructed finite systems.","tokens_in":23254,"feed_emoji":"🧪","tokens_out":7154,"duration_ms":70250,"temperature":0.7,"pith_summary":"This paper establishes a computational route from a prescribed spectral density function to a three-dimensional two-phase random microstructure, and shows that the transport properties of the result can be read off from the same function. It constructs disordered realizations in four classes—stealthy hyperuniform, standard hyperuniform, nonhyperuniform, and antihyperuniform—at several volume fractions and correlation lengths, including what it reports as the first explicit three-dimensional antihyperuniform two-phase media. The paper then computes the diffusion spreadability exactly and estimates fluid permeability from the target spectral functions, finding that the length-scale parameter alone can change spreadability by orders of magnitude at intermediate and long times while leaving the asymptotic decay class unchanged. If correct, the method turns a small set of analytic autocovariance models into a practical inverse-design tool for composites with targeted diffusive and fluid-transport behavior.","feed_headline":"First 3D antihyperuniform materials built from target spectra","feed_subtitle":"A spectral-density construction method realizes all four disorder classes and links microstructure directly to transport properties.","key_machinery":"The spectral density function $\\tilde{\\chi}_V(k)$, the Fourier transform of the autocovariance $\\chi_V(r)=S_2^{(i)}(r)-\\phi_i^2$, is the object that carries the argument. Because both the diffusion spreadability and the fluid-permeability estimate are expressible directly in terms of $\\tilde{\\chi}_V(k)$, targeting $\\tilde{\\chi}_V(k)$ couples microstructure generation to transport prediction without intermediate reconstruction steps. The construction itself is driven by an energy functional $E=\\sum_{k<K^*}(\\tilde{\\chi}_V(k)-\\tilde{\\chi}_V^0(k))^2$ minimized by simulated annealing, with the generalized collective coordinate $\\tilde{J}(k)$ updated locally after each voxel swap so that the spectral density of a trial configuration is obtained in $O(1)$ time per move. The four analytic families of target functions supply the length-scale parameter $a$ that tunes morphology within each class.","core_discovery":"On its own terms, the central discovery is that a Fourier-space simulated-annealing procedure can realize explicit three-dimensional two-phase media for analytic spectral densities spanning every hyperuniformity class, including the previously unrealized antihyperuniform case. The construction minimizes the squared difference between the angular-averaged spectral density of a binary voxel configuration and a target function $\\tilde{\\chi}_V(k)$, using rapidly updated collective coordinates to make each trial move cheap. With targets given by the Debye exponential, the damped-oscillatory hyperuniform model, the stealthy zero-region model, and the power-law autocovariance model $\\chi_V(r;a)=\\phi_1\\phi_2/[1+2(r/a)+(r/a)^2]$, the paper produces media at $\\phi_1=0.25,0.5,0.75$ and $a=5,10,25$ voxels. It reports that the antihyperuniform media contain clusters of dramatically different sizes and morphologies, mimicking critical-point fluctuations. Using the exact spreadability formula and a reference-based permeability estimate, it finds asymptotic decay exponents consistent with theory and an ordering in which antihyperuniform media have the largest dimensionless permeability $k/a^2$ and stealthy hyperuniform media the smallest.","pith_inferences":["A direct test the paper does not perform is to measure the local volume-fraction variance of the constructed antihyperuniform media; if the finite-size samples fail to show the slow $R^{-2}$ decay, the realization claim would need to be restricted to finite systems.","The permeability estimates inherit the accuracy of the reference-sample scaling in Eq. (28), so the reported ordering of $k/a^2$ across classes should be checked against direct pore-scale flow simulations before being used for design.","Since sufficient realizability conditions are open, success at one resolution does not guarantee that the same target function is realizable at other resolutions or volume fractions; the construction may be selecting a particular finite-size representative of a larger equivalence class.","The same construction machinery could be applied to mixtures of basis spectral densities to engineer hybrid microstructures, a direction the paper mentions only as a possibility for parameterized targets."],"forward_implications":["Any spectral density that satisfies the known necessary conditions can now be converted into an explicit 3D two-phase microstructure, including antihyperuniform media that had only been treated analytically.","Because spreadability and permeability are computed directly from the target $\\tilde{\\chi}_V(k)$, the length-scale parameter $a$ and volume fraction $\\phi_1$ can be optimized for a desired transport property before any microstructure is generated.","Within a fixed hyperuniformity class, changing $a$ changes the excess spreadability by orders of magnitude at intermediate and long times, so the asymptotic decay exponent alone does not determine the transport efficiency.","The dimensionless permeability $k/a^2$ decreases with increasing $\\phi_1$ and increasing $a$, and among the four classes antihyperuniform media give the largest $k/a^2$ while stealthy hyperuniform media give the smallest.","The same Fourier-space procedure can be adapted to non-isotropic target spectra and to dynamic wave properties, extending the 3D construction beyond diffusive transport."],"supporting_citations":[{"why":"Supplies the exact spreadability formula used in Eq. (20) and the asymptotic scaling laws that identify the transport class of each constructed medium.","marker":"[22]"},{"why":"Gives the two-point void bound and the reference-scaling estimate in Eq. (28) used to compute fluid permeability.","marker":"[25]"},{"why":"Defines the autocovariance and spectral density functions that the whole construction targets.","marker":"[43]"},{"why":"Establishes the simulated-annealing energy-minimization formulation that the Fourier-space construction builds on.","marker":"[44]"},{"why":"Introduces the generalized collective-coordinate Fourier-space construction method that this paper extends to 3D target spectral densities.","marker":"[64]"},{"why":"Provides the 2D spectral-density-based construction and fast update details generalized here to 3D isotropic media.","marker":"[65]"},{"why":"Lists the necessary realizability conditions the target autocovariances must satisfy, while noting sufficiency remains open.","marker":"[137]"},{"why":"Proposes the power-law autocovariance model used for the antihyperuniform targets and argues it meets all known necessary conditions.","marker":"[139]"},{"why":"Supplies the reference permeability $k_0$ of the simple-cubic hard-sphere packing used in Eq. (28) to estimate $k/a^2$.","marker":"[150]"}],"fun_headline_variants":["All four disorder classes now buildable in 3D from spectra","First 3D antihyperuniform media via spectral density targets","Spectral design links microstructure to transport in 3D","From spectra to solids: all disorder classes now in 3D"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the analytic autocovariance functions chosen as targets, in particular the power-law antihyperuniform model of Eq. (46), are genuinely realizable as three-dimensional two-phase media and that a finite $L=128$ voxel construction reproduces the diverging small-wavenumber behavior of Eq. (47); the paper notes that sufficient realizability conditions remain an open problem.","fun_headline_variants_meta":{"raw":{"variants":["All four disorder classes now buildable in 3D from spectra","First 3D antihyperuniform media via spectral density targets","Spectral design links microstructure to transport in 3D","From spectra to solids: all disorder classes now in 3D"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002353,"raw_usage":{"total_tokens":9149,"prompt_tokens":1118,"completion_tokens":8031,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":734,"completion_tokens_details":{"reasoning_tokens":7959}},"tokens_in":734,"tokens_out":8031,"duration_ms":57677,"temperature":1.0,"reasoning_tokens":7959,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T17:20:59.075278+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the spectral density of the constructed antihyperuniform samples at the smallest resolved wavenumbers, or compute the local volume-fraction variance in large spherical windows, and check for the predicted $\\tilde{\\chi}_V(k)\\sim 2\\pi^2/k$ divergence and for a variance decay slower than $R^{-3}$; if the small-$k$ spectrum saturates to a finite value because of the $L=128$ cutoff, the claimed first realization of true antihyperuniform media in 3D would not be supported by the constructed finite systems.","supporting_citations":[{"cited_title":"Gyromorphs: a new class of functional disordered ma- terials,","cited_arxiv_id":null,"evidence_quote":"Defines the autocovariance and spectral density functions that the whole construction targets."},{"cited_title":"Ensem- ble theory for stealthy hyperuniform disordered ground states,","cited_arxiv_id":null,"evidence_quote":"Lists the necessary realizability conditions the target autocovariances must satisfy, while noting sufficiency remains open."},{"cited_title":"Local volume fraction fluc- tuations in heterogeneous media,","cited_arxiv_id":null,"evidence_quote":"Proposes the power-law autocovariance model used for the antihyperuniform targets and argues it meets all known necessary conditions."},{"cited_title":"Understand- ing degeneracy of two-point correlation functions via debye random media,","cited_arxiv_id":null,"evidence_quote":"Supplies the reference permeability $k_0$ of the simple-cubic hard-sphere packing used in Eq. (28) to estimate $k/a^2$."}],"review_version":1}