{"id":"1a8fc91d-3cc1-4737-9059-ebee9861d166","arxiv_id":"1908.00603","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Scaled demagnetization effective-medium models are shown to predict magnetic susceptibility, resonant, and relaxation frequencies of high-susceptibility magnetic-particle composites better than Maxwell-Garnett and Bruggeman models.","lead":"This paper compares a family of effective-medium models against measured magnetic properties of polymer-particle composites and adds volume-fraction-corrected models for susceptibility, resonant frequency, and relaxation frequency. A generalist should read it to see whether simulation can replace trial-and-error formulation of magnetic composite materials used in RF devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"ScEMT's claimed improvement is anchored to a single fitted ferrite whose data are included in the qualitative validation; out-of-sample transferability is untested, and no quantitative cross-validation supports the universal gamma.","rationale":"The reader's weakest assumption is the transferability of the ScEMT scaling function, which is indeed a central issue. I agree with that concern and sharpen it: the same NiZnCu ferrite used for fitting is also included in the validation set, so part of the reported agreement is in-sample. For the relaxation frequency model, only the exponent gamma survives from the fit, making the universality of gamma even more load-bearing. The paper also lacks any quantitative error metric and explicitly admits reading error in the extracted data, so the comparative claims are not rigorously established. These considerations do not overturn the reader's CONDITIONAL verdict, because the conditions the reader requires (quantitative comparison and independent test data) are exactly what would resolve the concern. Thus the verdict remains unchanged, with the understanding that the paper is not yet acceptable without such validation.","tokens_in":12291,"tokens_out":10340,"duration_ms":106004,"concrete_test":"Perform a leave-one-out cross-validation over the material systems in Table 1: for each target material, fit A0, A1, gamma to the remaining systems (or, more strictly, keep the original NiZnCu fit and exclude NiZnCu from the evaluation), then compute mean absolute log-ratio errors for susceptibility, f_rc/f_r0, and f_dc/f_d0 for ScEMT, MGT, and CMA using digitized source data. Report the win rates and error margins, and compare the margins to the estimated digitization uncertainty (5-10%). If ScEMT does not win on the out-of-sample materials by a margin exceeding the digitization uncertainty, the central improvement claim is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that ScEMT and ScEMT-modified Schlomann improve on MGT/CMA for susceptibility, resonant, and relaxation frequency is supported by visual comparisons, but the ScEMT scaling function Ap(Pp) is fixed by fitting A0=0.975, A1=0.923, gamma=1.210 to the low-frequency susceptibility of one NiZnCu ferrite composite ([3],[4]), and that same material's susceptibility, resonance, and relaxation data appear in the evaluation figures (e.g., Figs. 5b, 11). Hence part of the agreement is in-sample, not predictive. For the relaxation model Eq. 26 the fitted A0/A1 are replaced by 1.0, so only gamma=1.210 is carried over; no evidence is given that this exponent is universal across particulate chemistry, size, or frequency. The paper explicitly states data read from graphs contain author reading error and 'should not be taken as absolutes,' and duplicate formulations differ by 10-20%, yet no error metric or confidence interval is reported. Without a quantitative, out-of-sample evaluation, the claimed improvement over MGT/CMA could be an artifact of the single fitted dataset and digitization noise.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the author's earlier Scaled Effective Medium Theory (ScEMT) to predict the magnetic susceptibility, resonant frequency, and relaxation frequency of magnetic-particle composites over a wide range of particulate susceptibilities (about 5 to 4000) and volume fractions (1-100%). After reviewing the ScEMT equations, the paper derives a low-volume-fraction correction for BEMT and ScEMT, then presents a resonant-frequency model (Eq. 19) based on a Snoek-like scaling assumption (Eq. 18) and a relaxation-frequency model (Eq. 26) that uses ScEMT-type volumetric scales and Schlomann's linewidth broadening. The models are compared to literature data collated from many sources; the central claim is that ScEMT-based models improve agreement with measurement over MGT and CMA, particularly for large-susceptibility particulates, while MGT and CMA remain accurate for modest susceptibilities.","tokens_in":12587,"tokens_out":8739,"duration_ms":83232,"significance":"If the claims hold, the paper would provide a simple and useful engineering tool for predicting the high-frequency magnetic response of composites, which is relevant to EMI suppression and antenna substrate design. The paper's strengths are its transparent algebraic derivations, the breadth of the compiled experimental dataset (12 composites, multiple chemistries), and its explicit and honest acknowledgment of data limitations (graph-reading error, duplicate-data scatter, and the need for controlled validation). However, the central evidence is qualitative: no error bars, goodness-of-fit metrics, or uncertainty propagation are reported, and part of the ScEMT evaluation uses the same material whose susceptibility was used to fit the model parameters. The significance paragraph of the abstract overstates the results relative to the evidence presented.","major_comments":[{"comment":"The central claim that ScEMT and ScEMT-modified Schlomann improve over MGT and CMA is supported only by visual comparison. The manuscript itself states, after Table 1, that 'the reader should assume that there is some level of author reading error' and 'A full validation of models must await a series of carefully controlled experiments.' No quantitative metric (e.g., relative error, R^2, RMSE, or confidence intervals) is provided for any model-measurement comparison, and the text notes that duplicate formulations differ by 10-20% (Figure 2b and the Fe data in Figure 4). With such scatter, the claimed improvement may be within the digitization and measurement uncertainty. The paper should provide a quantitative error analysis that separates the fitted material from truly out-of-sample predictions and that propagates the stated reading error into the model comparison.","section":"Validation sections (after Table 1, Figures 2-13)"},{"comment":"The ScEMT parameters A0=0.975, A1=0.923, and gamma=1.210 were fitted in prior work to the susceptibility of NiZnCu ferrite composites (Refs. [3],[4]). The same material appears in the resonant-frequency evaluation (Figure 5b) and in the relaxation-frequency evaluation (Figure 11, the NiZnCu entries with susceptibilities 839 and 863). These agreement plots are therefore partly in-sample and do not independently validate the model. For the other composites the parameters are assumed transferable, but no quantitative out-of-sample statistics are given. A leave-one-material-out analysis, or at least separate error metrics for the fitting material versus all other materials, is needed to support the universality claim.","section":"ScEMT Review, Eqs. (5)-(6), and Figures 5b and 11"},{"comment":"The 'ScEMT modifications to Schlomann' used in Eq. (26) do not actually employ the ScEMT parameters fitted to susceptibility; the text sets A0=A1=1.0 and C=1, and the alternative 'Volume Ratio' scale sets x=1 while noting that simulations suggested x near 3/2. Thus the relaxation model contains at least two additional adjustable choices beyond the original ScEMT fit. With these adjustments, the 'overall best agreement' reported for the relaxation model is not a test of the fixed ScEMT theory. The paper should either use the same parameters as the susceptibility fit, or treat C and x as fitted parameters with a documented model-selection criterion and uncertainty estimates.","section":"Relaxation Frequency Model, Eq. (26)"},{"comment":"The derivation of the composite resonant frequency assumes a Snoek-like relation, chi_c f_rc = gamma P_p (2/3) 4 pi M_s, for the composite. This proportionality is not justified for metallic particulates (Fe, NiFe), where Snoek's law in the form of Eq. (17) is in general not valid because of eddy currents and different resonance mechanisms. Equation (19) depends only on the ratio of the Snoek products, so the model can still be tested against data, but the paper should either verify that the bulk particulate properties satisfy Eq. (17) or restrict the resonance predictions to ferrites. The poor Fe comparison noted by the authors (Figure 7b) may be a symptom of this invalid assumption; the paper should discuss this possibility explicitly.","section":"Eqs. (17)-(19), resonant frequency model"}],"minor_comments":[{"comment":"There is a missing period after 'composites are presented' in the abstract; the sentence should end before 'ScEMT predictions'.","section":"Abstract"},{"comment":"The caption writes 'Coe~1.0' but the parameter is denoted C in the text; please correct the notation.","section":"Figure 9 caption"},{"comment":"The caption is confusing: it describes two NiZnCu ferrites with a vertical line, then mentions a MnO-ZnO ferrite. Please clarify which symbols correspond to which material and which vertical line separates the two NiZnCu data sets.","section":"Figure 11 caption"},{"comment":"The term 1.38/chi_p^2 in Eq. (24) has units of inverse susceptibility squared, which may be dimensionally confusing since chi_p is dimensionless in this paper; please rewrite the expression to make the cancellation of chi_p explicit (e.g., 1.38 f_r / chi_p).","section":"Equation (24)"},{"comment":"The paper derives a low-fraction correction for ScEMT, but then concludes that the correction is not sufficient and that ScEMT should not be applied at low volume fractions because of a sign-changing artifact. This negative result is presented honestly, but it is in tension with the abstract's claim that the paper 'modifies BEMT and ScEMT for volume fractions below about 10%;' please reconcile this in the introduction or conclusions.","section":"Section 'Small Volume Fraction Correction'"},{"comment":"The paper defines 'relaxation frequency' as the full width at half maximum of the imaginary part of susceptibility, which is more conventionally called a linewidth or bandwidth. Please state this definition whenever the term is used and ensure it is consistent with the Schlomann references cited.","section":"Terminology"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest about its limitations, but the absence of any quantitative error metric makes the abstract's claims of 'improvement' hard to evaluate. The in-sample overlap between the parameter-fitting material and the validation figures should be explicitly disclosed in the main text, not only in the prior-work sections. If the author can provide a modest quantitative analysis (e.g., relative errors with the stated reading-error bounds, and a separate breakdown for the fitting material), the paper would be much more convincing. The scope fits the journal, and the collated dataset is valuable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this paper extends the author's earlier ScEMT to resonant and relaxation frequencies, and it compiles a broad set of magnetic composite data. The new pieces are the Snoek-law resonant frequency relation (Eq. 19), the low-fraction corrections (Eqs. 10-12), and the ScEMT-scaled Schlomann relaxation model (Eq. 26). Those are genuinely new combinations, not just repackaged old results. The paper is also honest: it says where MGT and CMA are better (modest susceptibility) and where ScEMT wins (high susceptibility), and it explicitly tells the reader not to treat graph-derived data as absolutes. That honesty earns credit.\n\nThe soft spot is exactly where the stress-test note lands. The Ap(Pp) scaling function was fitted to one NiZnCu ferrite in the author's earlier work, and that same material appears in the evaluation figures for resonant and relaxation frequency. So a meaningful part of the agreement is in-sample, not predictive. For the relaxation model, A0 and A1 are set to 1.0, leaving only gamma=1.210 carried over from that single fit, and no evidence shows that exponent is universal across particle chemistry, size, or frequency. There are no error bars, no RMS errors, no confidence intervals. The paper claims visual improvement over MGT/CMA, and the figures do often look better for high-susceptibility particles, but the lack of quantitative metrics makes it hard to judge how much better, especially when duplicate measurements of the same nominal composite differ by 10-20%. The low-fraction correction also turns out to have a sign-change artifact and the paper concedes it doesn't fully fix the problem, so that section is more of a side note than a contribution.\n\nThat said, the central argument is not broken. The model extensions are plausible, the algebra is straightforward, and the data compilation is useful for anyone working on magnetic composites. The citation pattern is mostly self-cites plus standard EMT and ferrite literature, which fits an author extending his own line of work. My main concern is that the claimed universality of the scaling function is an assumption, not a demonstrated result.\n\nThis paper deserves a serious referee. I would want the referee to ask for a quantitative comparison table and an out-of-sample test, or at least an explicit discussion of which predictions are in-sample. The proposed controlled experiment in the conclusions is the right kind of follow-up. With that revision, it could be a useful reference for the composites community.\n\nBring it to reading group if you want a concrete example of how effective medium models are evaluated against messy literature data; otherwise it's not urgent.","headline":"New model forms for resonant and relaxation frequency prediction, but validation is mostly visual and the scaling function's transferability is untested beyond the one material it was fitted to.","tokens_in":13057,"tokens_out":2749,"would_cite":true,"duration_ms":29707,"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 volume-scaled demagnetization correction fitted to a single ferrite composite transfers across chemistries and improves prediction of susceptibility, resonant frequency, and relaxation frequency over classical effective-medium models.","keywords":["scaled effective medium theory","demagnetization scaling","magnetic composites","susceptibility prediction","resonant frequency","relaxation frequency","Schlomann model"],"falsifier":"Prepare composites from a high-susceptibility ferrite not used in the fit, for example with susceptibility near 4000, at volume fractions of 2, 5, 15, 30, 45, and 65 percent, and compare measured susceptibility, resonant frequency, and relaxation frequency against ScEMT and Equation 26 predictions using the fixed coefficients. If the coefficients must be refit for each chemistry, or if the predicted susceptibility increase near 10% volume fraction appears in carefully measured data, the transferability claim fails.","tokens_in":12050,"feed_emoji":"🧲","tokens_out":6539,"duration_ms":60237,"temperature":0.7,"pith_summary":"The paper argues that the failure of classical effective-medium models for magnetic composites is largely a volume-fraction-dependent demagnetization effect. It extends a previously developed scaled effective medium theory (ScEMT) in which the usual demagnetization factor is replaced by a function of particle volume fraction, fitted once to one NiZnCu ferrite composite, and applies it to susceptibility, resonant frequency, and relaxation frequency. Treated this way, ScEMT matches measured susceptibility and resonance better than Maxwell-Garnett and coherent-model approximations for particles with large susceptibility, while ScEMT-modified Schlomann models give the best relaxation-frequency agreement. The payoff would be a predictive design tool for magnetic composites that reduces the usual formulation-measurement-formulation iteration.","feed_headline":"Magnetic composites: scaled demagnetization beats classic models","feed_subtitle":"One scaled demagnetization function improves fits for susceptibility, resonance, and relaxation in high-permeability composites.","key_machinery":"The central object is the volume-scaled demagnetization coefficient $A_p(P_p)$, which replaces the fixed demagnetization constant $d-1$ in the Bruggeman effective-medium equation. It is defined by $A_c = A_0\\{1-A_1(1-P_p)^\\gamma\\}$ and $A_p = (1-A_c)/A_c$, with fitted values $A_0=0.975$, $A_1=0.923$, and $\\gamma=1.210$; this function encodes percolation-like clustering and chaining of high-susceptibility particles. The same demagnetization scaling is reused in the Snoek-law-based resonant frequency formula, Equation 19, and in the modified Schlomann relaxation frequency formula, Equation 26, and that reuse is what carries the argument from static susceptibility to frequency-dependent composite properties.","core_discovery":"The central claim is that a single nonlinear demagnetization scaling, $A_p(P_p) = (1-A_c)/A_c$ with $A_c = A_0\\{1-A_1(1-P_p)^\\gamma\\}$, fitted once to a NiZnCu ferrite composite with $A_0=0.975$, $A_1=0.923$, and $\\gamma=1.210$, transfers to a broad family of magnetic-particle composites. The paper applies this ScEMT scaling to predict DC susceptibility, resonant frequency via Snoek's law, and relaxation frequency by inserting the same scaling into Schlomann's 1969 model. Across a wide range of particulate chemistries, volume fractions, and susceptibilities, ScEMT predictions agree better with measurement than CMA and MGT for high-susceptibility particles, and ScEMT-modified Schlomann models show the best relaxation-frequency agreement. The paper also derives a low-volume-fraction correction for BEMT and ScEMT by comparing their small-fraction expansions to MGT, while noting that the correction has a sign-change artifact near 10% and that ScEMT remains best used at mid-to-high fractions.","pith_inferences":["Inference: the same volume-scaled demagnetization idea could be tested on electric permittivity composites, where percolation and clustering analogies exist, but the paper reports no such test.","Inference: the sign-change artifact near 10% volume fraction could be used as a calibration handle, since measured deviations in that range would constrain $A_1$ and $\\gamma$ independently of the high-fraction fit.","Inference: the paper's own proposed experiment with ferrites of susceptibility roughly 20, 800, and 4000, particle sizes 1, 10, and 40 microns, and volume fractions from 2% to 65% would directly test whether the fitted universal coefficients hold or need chemistry-dependent values.","Inference: the fitted exponent $\\gamma = 1.210$ could be compared with cluster statistics from micromagnetic or Brownian-dynamics simulations to give the scaling function a mechanistic derivation rather than a purely empirical one."],"forward_implications":["If the universal ScEMT scaling holds, composite susceptibility and resonant frequency can be predicted from bulk particulate susceptibility and resonance alone, without iterative formulation-measurement cycles.","ScEMT and its Schlomann modification become the preferred models for high-susceptibility particles, roughly above susceptibility 100, while CMA and MGT remain most accurate for modest susceptibility below about 100.","The low-volume-fraction corrections extend the usable range of BEMT slightly, though ScEMT's validity there is bounded by a sign-change artifact near 10% volume fraction.","Designers of EMI suppressors and antenna substrates could screen particulate chemistries and volume fractions computationally before committing to lab formulations.","Measured data sets show that formulation-measurement repetition can differ by 10-20%, so model-measurement agreement within that spread is the realistic target for any effective-medium prediction."],"supporting_citations":[{"why":"Introduces ScEMT and the fitted demagnetization scaling $A_p(P_p)$ that is the core object reused in this paper.","marker":"[3]"},{"why":"Demonstrates earlier ScEMT improvements for susceptibility and early resonant frequency results that this paper extends.","marker":"[4]"},{"why":"Supplies Schlomann's 1969 relaxation-frequency model with anisotropy and magnetization terms that Equation 26 modifies with volumetric scaling.","marker":"[8]"},{"why":"Provides the original void-fraction relaxation broadening expression that the 1969 analysis and this paper build on.","marker":"[7]"},{"why":"Gives the MGT- and CMA-based resonant and relaxation frequency formulas used as baselines for comparison.","marker":"[9]"},{"why":"Contains measured NiZnCu ferrite composite data used to fit the ScEMT coefficients $A_0$, $A_1$, and $\\gamma$.","marker":"[20]"},{"why":"States Snoek's law connecting susceptibility and resonance frequency, which is used to derive the ScEMT resonance formula.","marker":"[27]"},{"why":"Provides percolation theory that motivates the nonlinear dependence of demagnetization on volume fraction.","marker":"[19]"}],"fun_headline_variants":["One scaling law predicts composite magnetic response","Scaled demagnetization improves susceptibility and frequency fits","Single fitting function upgrades composite magnetic models","ScEMT scaling beats classics for high-permeability composites"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's improvements rest on the assumption that a demagnetization scaling fitted to a single NiZnCu ferrite composite transfers unchanged to every other particulate chemistry, particle size, and volume fraction, and to the frequency-dependent models for resonance and relaxation.","fun_headline_variants_meta":{"raw":{"variants":["One scaling law predicts composite magnetic response","Scaled demagnetization improves susceptibility and frequency fits","Single fitting function upgrades composite magnetic models","ScEMT scaling beats classics for high-permeability composites"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000383,"raw_usage":{"total_tokens":2119,"prompt_tokens":1124,"completion_tokens":995,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":740,"completion_tokens_details":{"reasoning_tokens":945}},"tokens_in":740,"tokens_out":995,"duration_ms":7799,"temperature":1.0,"reasoning_tokens":945,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:44:20.705771+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Prepare composites from a high-susceptibility ferrite not used in the fit, for example with susceptibility near 4000, at volume fractions of 2, 5, 15, 30, 45, and 65 percent, and compare measured susceptibility, resonant frequency, and relaxation frequency against ScEMT and Equation 26 predictions using the fixed coefficients. If the coefficients must be refit for each chemistry, or if the predicted susceptibility increase near 10% volume fraction appears in carefully measured data, the transferability claim fails.","supporting_citations":[{"cited_title":"Moore, AIP Advances 9, 3, (8 March 2019)","cited_arxiv_id":null,"evidence_quote":"Introduces ScEMT and the fitted demagnetization scaling $A_p(P_p)$ that is the core object reused in this paper."},{"cited_title":"Moore, J","cited_arxiv_id":null,"evidence_quote":"Demonstrates earlier ScEMT improvements for susceptibility and early resonant frequency results that this paper extends."},{"cited_title":"Schlomann, Phys","cited_arxiv_id":null,"evidence_quote":"Supplies Schlomann's 1969 relaxation-frequency model with anisotropy and magnetization terms that Equation 26 modifies with volumetric scaling."},{"cited_title":"Schlomann, Conf","cited_arxiv_id":null,"evidence_quote":"Provides the original void-fraction relaxation broadening expression that the 1969 analysis and this paper build on."},{"cited_title":"Tsutaoka, et.al., 2013 IEEE EMC International Symposium Digest,.545, IEEE-978-1-4799-0409-9/13","cited_arxiv_id":null,"evidence_quote":"Gives the MGT- and CMA-based resonant and relaxation frequency formulas used as baselines for comparison."},{"cited_title":"Moore, Electromagnetic Composites Handbook: Models, Measurement and Characterization, Chapter 12, McGraw Hill ISBN: 978- 1-25-958504-3 2016T","cited_arxiv_id":null,"evidence_quote":"Contains measured NiZnCu ferrite composite data used to fit the ScEMT coefficients $A_0$, $A_1$, and $\\gamma$."},{"cited_title":"Snoek, Physica, 14, 207 (1948)","cited_arxiv_id":null,"evidence_quote":"States Snoek's law connecting susceptibility and resonance frequency, which is used to derive the ScEMT resonance formula."},{"cited_title":"Clerc, G","cited_arxiv_id":null,"evidence_quote":"Provides percolation theory that motivates the nonlinear dependence of demagnetization on volume fraction."}],"review_version":1}