{"id":"307a734c-9941-489b-ac33-e78101f9edc4","arxiv_id":"2507.15052","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":6,"one_line_summary":"Ni doping at 2% in TiCoSb raises the thermoelectric power factor by 269%, which the authors attribute to a local disorder-to-order transition, but the evidence supports only a non-monotonic composition trend.","lead":"A study of nickel-doped TiCoSb thermoelectric samples reports a 269% increase in power factor and claims a disorder-to-order local structural transition at 2% Ni doping. A specialist might read it to see whether a small dopant can simultaneously raise thermopower and conductivity, but the structural-transition claim is not supported by the presented data.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim requires the x=0.02 extrema in EXAFS sigma^2 and Rietveld Biso to be real; with no error bars, no replicate samples, and the worst Ti K-edge fit at x=0.02, the 'disorder-to-order transition' is not established.","rationale":"I read the paper as an experimental report of a large PF increase at x=0.02 plus a proposed mechanism: an EXAFS/XRD-inferred local structural transition. The transport data are internally coherent and the PF enhancement is a credible experimental observation; the vulnerability is the explanatory bridge. The reader's weakest assumption identifies the same bridge: the x=0.02 anomaly is inferred from extrema in error-free fitted parameters, not from a demonstrated transition. My independent check of the SI and Table 2 strengthens that concern: the Biso differences are at the 0.01-0.02 Å^2 level, host-phase changes are sub-0.5%, and the Ti K-edge fit is worst at x=0.02. Those internal facts make the 'transition' reading particularly fragile. I do not see an ad hominem issue or a need to challenge the measured S, rho, or PF values. If the requested error-propagation/replicate check were supplied and the minima survived, a revised mechanism claim could become plausible; absent that, the central causal claim should not be accepted as established. Therefore I keep the reader's REJECT verdict rather than moving it.","tokens_in":24412,"tokens_out":5252,"duration_ms":57129,"concrete_test":"Independently re-reduce the raw Co and Ti K-edge EXAFS data for all six samples: vary the k-range and R-fitting window, float S02, ΔE0, Δr, and sigma^2, and obtain per-point uncertainties from the covariance matrix or from repeated scans; also run the same fitting pipeline on two or more independently synthesized batches at x=0.01, 0.02, and 0.03. If the 95% confidence intervals for sigma^2 at x=0.02 overlap those of neighboring compositions in either edge, or if the minimum is not reproduced in replicate batches, then the claimed disorder-to-order transition and its causal link to the power-factor enhancement are refuted. Report the same error propagation for Rietveld Biso and phase fractions to test whether their x=0.02 extrema survive.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim—that the ~269% room-temperature power-factor increase in TiCo0.98Ni0.02Sb is caused by a 'disorder-to-order' local structural transition—rests entirely on x=0.02 being a genuine minimum in EXAFS sigma^2 (Section 4.3, Figs. 7(c), 8(c)) and in Rietveld Biso (Fig. 5(c)). No uncertainties are reported for any of these fitted parameters, and each composition was measured once. The Rietveld support is internally weak: from SI Fig. S2 the Biso values are 0.675, 0.69, 0.67, 0.69, 0.69, and 0.71 Å^2, differences comparable to typical refinement precision, and the host-phase fraction changes by less than 0.4% across the series, from 99.44% at x=0 to 99.85% at x=0.02. The EXAFS support is self-contradictory: Table 2 shows the best Co K-edge fit at x=0.02 (R-factor 0.00498) but the worst Ti K-edge fit at x=0.02 (R-factor 0.03933, lowest reported 'happiness' 84.25), so the second edge, invoked as confirmation, is least reliable exactly where the anomaly is claimed. The 'drastic change in slope' of sigma^2 versus x is an eye-of-the-beholder judgment on six points, with no confidence intervals, no bootstrap, no replicate synthesis, and no statistical test separating a minimum from noise. DFT (Section 4.1) was computed only at 0%, 3%, and 6% Ni, so it does not sample the claimed transition composition. Even a true sigma^2 minimum would not by itself prove a phase transition, since sigma^2 is a mean-square displacement rather than an order parameter and no symmetry change or transition signature is shown. The measured PF enhancement may be real, but the load-bearing structural-transition mechanism proposed to explain it is unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the synthesis and characterization of arc-melted TiCo_{1-x}Ni_xSb (x = 0, 0.01, 0.02, 0.03, 0.04, 0.06) half-Heusler alloys, combining DFT band-structure calculations, Rietveld refinement of powder XRD, Williamson-Hall analysis, Co and Ti K-edge XAS, and temperature-dependent resistivity and thermopower measurements. The central claim is a ~269% room-temperature power-factor enhancement at x = 0.02, attributed to a disorder-to-order local structural transition that simultaneously increases the Seebeck coefficient and electrical conductivity. The authors support this by reporting extrema at x = 0.02 in lattice strain, dislocation density, host-phase fraction, Debye-Waller factor, and EXAFS mean-square relative displacements, and they interpret the transport anomalies through the same structural picture.","tokens_in":24968,"tokens_out":2866,"duration_ms":31942,"significance":"If the claimed disorder-to-order transition were robustly established, the work would be of interest to the half-Heusler thermoelectric community as a rare example of simultaneous enhancement of S and σ by small doping, and it would motivate further study of local structural order as a tuning parameter for power factor. The paper deserves credit for direct transport measurements, a systematic doping series, and the use of complementary XRD and XAS probes. However, the significance is contingent on the structural anomaly at x = 0.02 being real, and that premise is not currently supported by the reported data: the fitted structural parameters have no uncertainties, the composition series has no replicates, and the EXAFS fit quality is worst exactly at the claimed transition composition.","major_comments":[{"comment":"The central structural claim rests on minima in the EXAFS disorder parameter σ² at x = 0.02, yet no uncertainties or replicate measurements are reported for any σ² value, and each composition was measured once. The 'drastic change in slope' of σ² versus x is asserted from six points without a statistical test, bootstrap, or confidence interval that would distinguish a genuine minimum from scatter. Moreover, the Ti K-edge fit at x = 0.02 has the worst R-factor (0.03933) and lowest 'happiness' (84.25) of the entire series, so the second edge invoked as confirmation is least reliable exactly at the claimed anomaly. Without error bars, the EXAFS data do not establish a local structural transition.","section":"Section 4.3, Figs. 7(c) and 8(c), Table 2"},{"comment":"The Rietveld support for a long-range anomaly at x = 0.02 is internally weak. The reported Biso values are 0.675, 0.69, 0.67, 0.69, 0.69, and 0.71 Ų, a spread of ~0.04 Ų that is comparable to typical refinement precision, and the x = 0.02 pattern is not the best fit (Rwp = 22.7%, Gof = 3.62; the largest Gof in the series). The lattice parameter varies only between 5.88270 and 5.8832 Å, and the host-phase fraction changes by less than 0.4% across the series. These differences are within plausible systematic and statistical uncertainty, so the claimed minimum in Biso and the maximum host phase do not constitute demonstrated anomalies.","section":"Section 4.2, Fig. 5(c), and SI Fig. S2"},{"comment":"The DFT calculations were performed only for Ni concentrations of 0%, 3%, and 6%, so they do not sample the claimed transition composition x = 0.02. The computed Fermi-level shift is a monotonic doping trend and cannot provide evidence for an anomaly at x = 0.02; it is therefore not a valid supporting pillar for the structural-transition claim.","section":"Section 4.1, Fig. 2"},{"comment":"There is a circularity in the argument: the 'disorder-to-order transition' is inferred from the minima in Biso, σ², and strain at x = 0.02, and the same transition is then used to explain the transport anomalies at that composition. The Lorentz number is derived from the measured thermopower using a single parabolic band model, and the resulting L(T) behavior is cited as confirming the structural transition. This does not provide independent evidence; at most it is consistent with the authors' interpretation, and it cannot validate the existence of a structural transition.","section":"Sections 4.3 and 4.4, Eqs. (6)-(8)"}],"minor_comments":[{"comment":"The Williamson-Hall equation is written as βcosθ = k_B λ / D + 4ε sinθ with k_B identified as the Boltzmann constant; the standard form uses the Scherrer shape factor K, not the Boltzmann constant. This should be corrected to avoid confusion.","section":"Section 4.2, Eq. (2)"},{"comment":"There are multiple typos, including 'anazlyzed' instead of 'analyzed', 'smples' instead of 'samples' in Section 4.2, and 'XENES' instead of 'XANES' in the caption of Fig. 6. The manuscript needs careful proofreading.","section":"Section 4.4"},{"comment":"The Introduction states 'local structural rearrangement in TiCo_{1-x}Ni_xSb (0 < x < 0.6)' but the doping range studied is 0 < x < 0.06; the upper limit should be corrected.","section":"Section 1"},{"comment":"The modified Williamson-Hall derivation contains 'Millar indices' (should be Miller) and the meaning of the 'Happiness of fit' metric used in Table 2 is not defined anywhere; it should be defined or replaced with a standard statistical measure.","section":"SI, Section 1"}],"recommendation":"reject","confidential_remarks":"The transport measurements themselves may be of interest, but the paper's headline claim—a disorder-to-order transition at x = 0.02 causing a 269% PF enhancement—is not supported by the evidence as presented. The absence of error bars and replicates, combined with the worst EXAFS fit at the claimed transition composition and the inadequacy of the DFT doping grid, means that the central conclusion cannot be accepted. Establishing or refuting the claim would require additional measurements (e.g., replicate syntheses, error propagation, temperature-dependent diffraction or pair-distribution-function analysis), which is beyond the scope of a revision of the current manuscript. I see no indication of misconduct, only over-interpretation of under-characterized data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid transport-measurement paper with an interpretive overreach. The new experimental series — TiCo1-xNixSb at x = 0.01, 0.02, 0.03, 0.04, 0.06 — is not in the prior Ni-doping literature, and the simultaneous rise in |S| and conductivity at x = 0.02 is a genuine, interesting observation. The ~269% room-temperature power-factor increase is a direct measurement and worth taking seriously. The DFT at 3% and 6% showing metallic-like DOS with a Fermi-level shift is consistent with the transport trend, and the transport analysis itself is mostly standard.\n\nThe soft spot is the central claim: a 'disorder-to-order local structural transition' at x = 0.02. That claim rides entirely on minima in Rietveld Biso, EXAFS sigma^2, and strain/dislocation density at that composition. No uncertainties are reported on any of those fitted parameters, and each composition was measured once. From the SI, the Biso values are 0.675, 0.69, 0.67, 0.69, 0.69, 0.71 Å^2 — differences in the noise of typical refinement. The host-phase fraction changes by less than 0.4% across the series. On the EXAFS side, the Co edge fits best at x = 0.02 but the Ti edge — invoked as confirmation — fits worst at exactly that composition (R-factor 0.039, happiness 84.25). So the 'drastic change in slope' of sigma^2 is an eye-of-the-beholder judgment on six points without error bars or replicates. Sigma^2 is not an order parameter, and no symmetry change or transition signature is shown. DFT was computed only at 0, 3, 6%, so it does not cover the claimed transition composition. The Lorentz-number analysis derives L from the measured S via a single-parabolic-band model, so it is not independent evidence. Thus the measured PF can be real while the proposed mechanism remains unproven.\n\nAlso missing: direct carrier-concentration or Hall data, which would have been the natural check. Citation pattern is fine; prior Ni-doping work is cited and the new series is distinct.\n\nWho this is for: experimentalists working on half-Heusler thermoelectrics will get useful data points and a cautionary tale about overinterpreting fitted structural parameters. It deserves a serious referee: the measurements are a legitimate contribution, and a referee could push for error bars, replicate samples, and a much softer claim about a transition.","headline":"Measured PF jump at x=0.02 is likely real; the disorder-to-order transition used to explain it is not established by the data.","tokens_in":25559,"tokens_out":2579,"would_cite":false,"duration_ms":27489,"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 that 2 percent nickel doping raises the power factor of the half-Heusler thermoelectric TiCoSb by roughly 269 percent by making thermopower and electrical conductivity rise together at the same composition.","keywords":["TiCoSb half-Heusler","Ni doping","power factor","disorder-to-order transition","EXAFS","Rietveld refinement","thermopower","electrical resistivity"],"falsifier":"Measure EXAFS $\\sigma^2$ on a finer series with several samples per composition and explicit uncertainties. If the minimum at $x=0.02$ and its slope discontinuity do not reproduce, or if the values at neighboring compositions overlap within error, the disorder-to-order transition is not supported. A second check is Hall-effect carrier concentration: if carrier density rises monotonically with Ni content while thermopower peaks at $x=0.02$, then the thermopower peak is not a simple carrier-concentration effect and the structural explanation has to carry the argument.","tokens_in":24224,"feed_emoji":"⚡","tokens_out":7310,"duration_ms":74768,"temperature":0.7,"pith_summary":"The paper reports that replacing 2% of the cobalt in the half-Heusler thermoelectric TiCoSb with nickel raises the room-temperature power factor by roughly 269%, from about 95 to 350 $\\mu$W K$^{-2}$ m$^{-1}$. The notable feature is that thermopower and electrical conductivity increase together at that composition, although they normally oppose each other. The authors trace the simultaneous gain to a local structural \"disorder-to-order\" transition at $x=0.02$, seen as a minimum in the EXAFS disorder parameter $\\sigma^2$ together with matching extrema in lattice strain, Debye-Waller factor, host-phase fraction, and dislocation density. If the attribution is right, local atomic ordering is a compositional lever for thermoelectric power factor, not just a side effect of doping.","feed_headline":"2% nickel doping lifts TiCoSb power factor by 269 percent","feed_subtitle":"A disorder-to-order switch at x=0.02 raises thermopower and conductivity together, beating their usual trade-off.","key_machinery":"The load-bearing observable is the EXAFS disorder parameter $\\sigma^2$, the mean-square fluctuation of near-neighbor bond lengths around Co and Ti atoms; its value falls to a minimum at $x=0.02$ and rises on either side, and the slope change in $\\sigma^2$ versus $x$ is interpreted as the disorder-to-order transition. The supporting machinery is the set of long-range structural probes that reproduce the same extremum: Williamson-Hall strain and dislocation density, Rietveld Debye-Waller factor $B_{\\rm iso}$, host-phase weight fraction, and unit-cell volume. Density-functional-theory supercell densities of states carry the electronic side of the argument, converting Ni substitution into a Fermi-level shift that changes the transport from semiconducting toward metallic.","core_discovery":"The central claim is that TiCo$_{1-x}$Ni$_x$Sb undergoes a local structural transition from a disordered to a more ordered atomic arrangement at $x=0.02$, and that this transition is what permits the thermopower and electrical conductivity to improve at the same time, producing a marked increase in power factor. The authors base the claim on four coordinated signatures at that composition: first-principles calculations that push the Fermi level toward the conduction band on Ni substitution; Rietveld refinement of X-ray diffraction showing maximum host TiCoSb phase, minimum Debye-Waller factor, and minimum strain and dislocation density; EXAFS at the Co and Ti K-edges showing a minimum in the bond-length disorder parameter $\\sigma^2$ with a slope change read as the transition; and transport data in which thermopower rises up to $x=0.02$ and falls beyond, resistivity acquires a metallic low-temperature branch, and Lorentz-number and scattering analysis corroborate the electronic changes. The paper's conclusion is that the structural ordering and the transport enhancement are the same event.","pith_inferences":["A direct testable extension is a finer composition grid around $x=0.02$: the transition claim predicts that $\\sigma^2$ and the Debye-Waller factor trace a sharp V-shape rather than a smooth curve.","The paper does not measure thermal conductivity, so the impact on the figure of merit $ZT$ remains open; if the same ordering reduces phonon scattering, the lattice thermal conductivity could rise and partially offset the power-factor gain.","The Lorentz number shows a slope change near 125 K, suggesting a temperature-driven structural event; temperature-dependent EXAFS across that range would test whether the local ordering is composition-specific or also thermally reversible.","Because only two DFT compositions (3% and 6%) are modelled, the theoretical argument does not resolve the special status of $x=0.02$; a calculation at or near 2% would sharpen the link between the Fermi-level shift and the structural anomaly."],"forward_implications":["If the transition picture is correct, the $x=0.02$ composition is a genuine optimization point: the power factor reaches about 350 $\\mu$W K$^{-2}$ m$^{-1}$ at 300 K, a gain of roughly 269% over the undoped sample.","The simultaneous rise of $S$ and $\\sigma$ at this composition means the usual inverse relationship between thermopower and conductivity is relaxed at the ordering point, so power factor can be improved without the standard doping trade-off.","The structural and transport signatures line up at the same composition, which makes the disorder parameter a useful indicator for searching similar peaks in other half-Heusler alloys.","For $x>0.02$, the thermopower falls as carrier concentration grows, so the effect is specific to the 2% composition rather than a monotonic doping trend.","The decrease in electron-phonon and electron-electron scattering coefficients up to $x=0.02$ is consistent with a more ordered lattice conducting electrons more easily."],"supporting_citations":[{"why":"Supplies prior evidence that embedded CoTi phases and defects shape TiCoSb transport, the baseline against which the x=0.02 anomaly is compared.","marker":"[14]"},{"why":"Earlier calculations establishing that Ni doping shifts the Fermi level in TiCo1-xNixSb, the electronic foundation for the semiconducting-to-metallic trend.","marker":"[15]"},{"why":"Establishes that Ni substitutes at Co and that the cell-volume decrease follows from atomic radii, anchoring the substitution claim.","marker":"[22]"},{"why":"Prior study of partial Ni substitution in TiCoSb that documents CoTi embedded phases and their transport impact.","marker":"[36]"},{"why":"Provides the EXAFS data-analysis route from which sigma^2 and bond lengths are extracted.","marker":"[42]"},{"why":"Supplies the VO2 analogue in which structural disorder drives a metal-insulator transition, the template for reading slope changes in sigma^2 as a transition.","marker":"[46]"},{"why":"Reports similar thermopower variation in Ti0.5Zr0.25Hf0.25Co1-xNixSb, the closest transport comparison for the Ni-doping trend.","marker":"[47]"},{"why":"Provides the resistivity fitting models used to extract residual resistivity and electron-phonon and electron-electron scattering coefficients.","marker":"[48]"}],"fun_headline_variants":["Nickel doping orders atoms, lifts TiCoSb power factor 269%","Local structural shift at x=0.02 drives TiCoSb power factor up 269%","Disorder-to-order switch boosts Ni-doped TiCoSb power factor 269%","Ni doping at 2% triggers structural change and 269% PF jump in TiCoSb"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire interpretation rests on the dip in EXAFS $\\sigma^2$ and the matching extrema in strain, Debye-Waller factor, and dislocation density at $x=0.02$ being a real disorder-to-order transition rather than scatter in a six-composition series plotted without error bars.","fun_headline_variants_meta":{"raw":{"variants":["Nickel doping orders atoms, lifts TiCoSb power factor 269%","Local structural shift at x=0.02 drives TiCoSb power factor up 269%","Disorder-to-order switch boosts Ni-doped TiCoSb power factor 269%","Ni doping at 2% triggers structural change and 269% PF jump in TiCoSb"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000722,"raw_usage":{"total_tokens":3259,"prompt_tokens":985,"completion_tokens":2274,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":601,"completion_tokens_details":{"reasoning_tokens":2182}},"tokens_in":601,"tokens_out":2274,"duration_ms":19884,"temperature":1.0,"reasoning_tokens":2182,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T15:41:38.228908+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure EXAFS $\\sigma^2$ on a finer series with several samples per composition and explicit uncertainties. If the minimum at $x=0.02$ and its slope discontinuity do not reproduce, or if the values at neighboring compositions overlap within error, the disorder-to-order transition is not supported. A second check is Hall-effect carrier concentration: if carrier density rises monotonically with Ni content while thermopower peaks at $x=0.02$, then the thermopower peak is not a simple carrier-concentration effect and the structural explanation has to carry the argument.","supporting_citations":[{"cited_title":"Mahakal, D","cited_arxiv_id":null,"evidence_quote":"Supplies prior evidence that embedded CoTi phases and defects shape TiCoSb transport, the baseline against which the x=0.02 anomaly is compared."},{"cited_title":"Romaka, M","cited_arxiv_id":null,"evidence_quote":"Earlier calculations establishing that Ni doping shifts the Fermi level in TiCo1-xNixSb, the electronic foundation for the semiconducting-to-metallic trend."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that Ni substitutes at Co and that the cell-volume decrease follows from atomic radii, anchoring the substitution claim."},{"cited_title":"Mahakal, D","cited_arxiv_id":null,"evidence_quote":"Prior study of partial Ni substitution in TiCoSb that documents CoTi embedded phases and their transport impact."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the VO2 analogue in which structural disorder drives a metal-insulator transition, the template for reading slope changes in sigma^2 as a transition."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports similar thermopower variation in Ti0.5Zr0.25Hf0.25Co1-xNixSb, the closest transport comparison for the Ni-doping trend."}],"review_version":1}