{"id":"0cb3eb84-3fe9-4ed2-a28a-be0b325468ce","arxiv_id":"2501.05047","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new analytical model reproduces numerically computed energy barriers and switching attempt frequencies for magnetization reversal in finite Co and Fe chains on Pt(332), enabling coercive force estimates.","lead":"This paper computes the energy barriers and attempt frequencies for flipping the magnetization of short cobalt and iron atomic chains on a platinum surface, and builds an analytical model that captures the numerical trends. The model allows estimating the coercive field of such chains over a wide range of lengths, temperatures and sweep rates, which is relevant for atomic-scale magnetic memory and spintronics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central analytical claim is conditional on the true MEP remaining inside the trial manifold (XY for Co; tilted uniform-phi0 for Fe), but this is only checked at the reference point; Eq. (31) as printed is undefined for Table I parameters, making the Fe branch non-reproducible.","rationale":"The reader's conditional verdict is appropriate, and this stress-test reinforces it rather than overturning it. The paper does demonstrate genuine qualitative agreement with GNEB and TST at the reference Hamiltonian (N up to 100, By = 1 T, Table I parameters), so the central claim has real numerical support where directly tested. The weakest point is the extrapolation: the analytical formulas are built from a restricted set of trial configurations, and the same ansatz is then used to predict coercive forces over wide ranges of J, K-|E|, and Dz in Fig. 7(d-f). Those predictions have not been checked numerically, so the breadth of the central claim is exactly where the load-bearing assumption sits. The printed Eq. (31) is an additional concrete defect: with Table I parameters the radicand exceeds 1, making the arccos undefined, while the quoted phi0 value implies a different arrangement of the same dimensionless ratios. This does not necessarily invalidate the physical idea, because the numerical phi0 is consistent with a corrected formula, but it makes the Fe analytical branch non-reproducible as written. A variable-phi0 GNEB test at reference and extreme parameters would settle whether the manifold assumption actually holds. None of this requires changing the verdict from conditional: the paper's own limitations, including invalid prefactor peaks near mechanism switching, factor-of-two prefactor discrepancies for Fe, and 20-25% coercivity overestimates, already support a conditional accept.","tokens_in":25833,"tokens_out":16254,"duration_ms":155006,"concrete_test":"Re-run the GNEB calculation for an N=100 Fe chain at By=1 T, and at one extreme each of the Fig. 7(d-f) sweeps (for example Dz = 0.5 and 3.5 meV), using initial paths in which phi_i is a free function of position rather than fixed at the uniform phi0 of Eq. (31), and compare the lowest barrier and TST prefactor to Eqs. (38)-(39) and (49). If the barrier decreases by more than about 1 meV or the prefactor changes by more than a factor of two, the restricted-manifold assumption does not hold outside the reference point and the extrapolated coercivities require caveats.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing condition for the paper's central claim is that the true minimum-energy paths for all chain lengths, fields, and Hamiltonian parameters stay inside the trial manifold used for the analytic saddles: in-plane rotation for Co and uniform-phi0 tilted-plane wall-plus-end-twist superpositions for Fe (Eqs. (18)-(21) and (29)-(36)). This is verified only by GNEB near the reference point (N <= 100, By = 1 T, Table I parameters); the Fig. 7(d-f) coercivity sweeps then extrapolate the same ansatz over wide intervals of J, K-|E|, and Dz with no numerical benchmark. The Fe branch is additionally fragile as printed: Eq. (31) evaluated with Table I parameters has a radicand of about 7.5, so the arccos is undefined, whereas the quoted phi0 about 80.4 degrees would follow only from a differently arranged formula. An independent reader therefore cannot currently reproduce the Fe analytical curves. If a lower-energy path exists outside this ansatz (e.g., a non-uniform phi_i wall or a spiral ground state at large Dz), the analytic barriers and prefactors, and hence the predicted coercive fields, would shift.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies magnetization reversal in finite-length Co and Fe atomic chains on Pt(332) using a classical Heisenberg model with exchange, MAE, DMI, Zeeman, and dipole-dipole interactions. GNEB calculations give energy barriers and harmonic-TST calculations give frequency prefactors for chains of 5-100 atoms in a field along the easy axis. The authors identify two reversal regimes: coherent reversal for short chains and domain-wall-mediated reversal for long chains, with distinct wall structures for Co and Fe. They then develop a continuum analytical framework for the barriers and prefactors, based on domain-wall-pair ans\\\"atze, and use these to predict coercive forces as functions of chain length, temperature, sweep rate, and Hamiltonian parameters. The analytical barrier formulas agree well with GNEB except near the mechanism-change lengths; the prefactor formulas reproduce the asymptotic trends, with the Fe prefactors about a factor of two off.","tokens_in":26026,"tokens_out":21406,"duration_ms":201843,"significance":"If the analytical framework is correct, it provides a parameter-free route to estimate switching fields of one-dimensional magnetic chains without case-by-case simulation. The paper supplies the first TST-prefactor results for finite chains with DMI and magnetic field, and the formulas are derived from the same Hamiltonian parameters used in the numerics rather than fitted to the barriers or prefactors. The main strengths are the systematic GNEB/TST study, the explicit analytical comparison, and the open discussion of the harmonic-TST breakdown at the peaks. The principal weaknesses are the non-reproducibility of the Fe analytical branch as printed (Eq. (31) is undefined for Table I parameters) and the unvalidated extension of the saddle-manifold ansatz over wide parameter ranges in the coercive-force sweeps.","major_comments":[{"comment":"For the Table I Fe parameters, the radicand in Eq. (31) is (K-|E|)/(2|E|) * (16J|E|/(pi^2 D_z^2) - 1) ~ 0.4719 * 15.89 ~ 7.50, so the arccosine is undefined; the reported value phi0 = +/-80.43 deg cannot follow from Eq. (31) as printed. Since phi0 enters the Fe saddle construction (34)-(36) and the barriers (38)-(39), an independent reader cannot reproduce the Fe analytical curves. Please correct the formula and verify the quoted angle.","section":"III B, Eq. (31)"},{"comment":"As written, these equations are not defined for chi > 0: the logarithm in Eq. (14) has argument (1+sqrt(1+chi))/(1-sqrt(1+chi)) < 0, and Eq. (17) additionally contains arccos(-2chi-1) with argument less than -1. In the actual applications EDW is evaluated at -By (chi < 0), so the numerical comparisons may be unaffected, but the printed formulas are stated for general By. Please restrict the domain or provide the correct branch (e.g., with an absolute value or a sign-dependent definition).","section":"III B, Eqs. (14) and (17)"},{"comment":"The coercive-force predictions are based on the same analytical ansatz (in-plane reversal for Co; uniform-phi0 tilted-plane walls for Fe) that is validated by GNEB/TST only at the reference point (N <= 100, By = 1 T, Table I parameters). In Fig. 7(d-f) the formulas are used over wide ranges of J, K-|E|, and D_z without any numerical benchmark. If, for example, a large D_z drives the ground state into a DMI spiral or the Fe wall develops a non-uniform phi_i, the predicted barriers, prefactors, and B_c would change. The paper should state the expected validity domain of the ansatz or provide GNEB checks at a few off-reference parameter sets before claiming applicability across a wide range of model parameters.","section":"III D, Fig. 7(d-f)"},{"comment":"The analytical prefactor formulas capture the linear short-chain growth and the exponential growth of nu_01, but they do not reproduce the sharp peaks in nu_0(N) near the mechanism-change lengths (N ~ 17 and 24 for Co, N ~ 16 for Fe). The text correctly notes that harmonic TST is invalid at those peaks, but the abstract and conclusion state that the approach qualitatively describes the numerical prefactor dependencies; this should be qualified to the asymptotic regimes outside the peaks.","section":"III C, Fig. 5(c,d)"}],"minor_comments":[{"comment":"For Fe (E = -0.89 meV), Eq. (45) gives Nmax = 2*pi/arccos((J-K+E)/J) ~ 18.5, not 32.78 as stated. The reported value appears to have been obtained with |E| in the numerator. Please reconcile the formula with the quoted Nmax.","section":"III C, Eq. (45)"},{"comment":"The discussion that the function F in Eq. (44) must remain positive and that Nmax is the maximal chain length is imprecise: because the factors for n and N-n are equal, negative factors can appear in pairs, so F can remain positive beyond the first zero of a single factor. The operative condition is that the coherent-reversal configuration has exactly one unstable mode, which is what Nmax should represent.","section":"III C"},{"comment":"The word \"ether\" in the figure captions should be \"either\".","section":"Figs. 6 and 7"},{"comment":"The sign convention for E is stated in the text (positive for Co, negative for Fe), but a brief summary after Eq. (3) would help readers keep track of the |E| that appears in later formulas.","section":"II A"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope. The main concern is reproducibility of the Fe analytical branch because Eq. (31) as printed is undefined for the reported parameters; this looks like a typographical or sign error rather than a conceptual flaw. I would ask the authors to correct the equation, reconcile the Nmax value for Fe, and add a short statement about the expected validity range of the saddle-manifold ansatz in the parameter sweeps. The numerical study and the analytical comparison are valuable and the central idea is defensible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid numerical study plus a genuinely useful analytical approximation for energy barriers and TST prefactors of Co and Fe chains on Pt(332). The new piece is the harmonic-TST prefactor calculation with finite fields, including the non-monotonic length dependence; I believe that is the first for these systems. The GNEB and TST numerics look carefully done, and the analytical formulas match them well over N = 5–100, especially for barriers. The authors also say clearly where the harmonic approximation fails, which is more honest than most papers in this area.\n\nThe main soft spot is the Fe branch. Equation (31) as printed is not computable for the Table I parameters: the quantity under the outer square root is about 7.5, so the arccos is undefined. The quoted phi0 of about 80.4 degrees would follow from a differently arranged formula, but a reader cannot reproduce the Fe analytical curves from the paper alone. This matters because the whole analytical claim for Fe hinges on those saddle-point configurations. The numerical GNEB results support the trial manifold at the reference point, but the parameter sweeps in Fig. 7(d–f) extrapolate that ansatz over wide ranges with no benchmark, so a lower-energy path outside the manifold is a real, unexcluded possibility.\n\nThe other weaknesses are less severe. The Fe prefactor estimates are off by about a factor of two, though the authors admit they only need order-of-magnitude accuracy. The sharp prefactor peaks at mechanism-switch lengths are explicitly invalid within harmonic TST, and the authors say so. No code or data is deposited, so the numerics are not directly reproducible. None of this kills the paper, but the Fe formula issue needs to be fixed before the analytical claims can be taken as fully supported.\n\nCredit where due: the paper does not fit parameters to the reported barriers or prefactors, the reuse of the earlier zero-field results is properly cited, and the analytical framework does capture the qualitative and often quantitative behavior. The coercive-force estimates are clearly labeled as approximate, with the error budget explained.\n\nWho should read this: anyone working on atomic-chain magnetism, DMI domain walls, or TST for nanoscale spin systems. It deserves a serious referee, not a desk reject, but the referee should be asked to verify and repair the Fe branch and ideally to share the numerical data. I would accept it conditionally, with the Fe formula fix as a hard requirement.","headline":"Useful, honest subfield advance on 1D chain reversal rates; the analytical framework is promising, but the Fe branch as printed does not reproduce, so referee should demand a fix before endorsement.","tokens_in":26598,"tokens_out":1590,"would_cite":true,"duration_ms":18800,"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":"The paper claims that an analytical model built from domain-wall-pair solutions reproduces the numerically computed energy barriers and frequency prefactors for magnetization reversal in finite Co and Fe chains on Pt(332), and that this…","keywords":["magnetization reversal","atomic chains","Dzyaloshinskii-Moriya interaction","energy barriers","frequency prefactors","transition state theory","geodesic nudged elastic band","coercive force"],"falsifier":"Run a geodesic nudged elastic band calculation for a Co chain of $N=30$ atoms at $B_y=2$ T and compare the converged minimum-energy path and barrier with the $XY$-plane domain-wall-pair predictions of Eqs. (22)-(25), and for an Fe chain do the same against the tilted-plane predictions of Eqs. (38)-(39); if the path leaves the assumed plane or the barrier differs outside the mechanism-crossover window, the ansatz is wrong.","tokens_in":25531,"feed_emoji":"🧲","tokens_out":8073,"duration_ms":73744,"temperature":0.7,"pith_summary":"The paper tries to establish that magnetization reversal in finite Co and Fe atomic chains on Pt(332) can be captured by a compact analytical model, not just by case-by-case simulation. It first maps the reversal mechanisms numerically: short chains reverse by coherent rotation of all moments, while long chains nucleate domain walls whose structure differs between Co and Fe because of the Dzyaloshinskii-Moriya interaction. Using geodesic nudged elastic band calculations for chains of 5 to 100 atoms, it obtains energy barriers, and harmonic transition state theory gives frequency prefactors that depend strongly and non-monotonically on chain length and field. The paper's central move is to approximate ground, local-minimum, and saddle configurations as superpositions of twisted or untwisted domain-wall-pair solutions of a continuous model, which yields closed-form barriers and prefactors that qualitatively match the numerics. If correct, this gives a way to estimate coercive forces of one-dimensional magnetic chains across lengths, temperatures, sweeping rates, and Hamiltonian parameters without rerunning expensive simulations.","feed_headline":"Switching of Co and Fe chains is now analytically predictable","feed_subtitle":"Short chains flip together, long chains flip via domain walls, and the formulas match full simulations for 5-100 atoms.","key_machinery":"The load-bearing object is the domain-wall-pair solution $\\Theta(x,K,B_y)$, a closed-form solution of the continuum energy functional for an infinite chain with exchange, easy-axis anisotropy, Dzyaloshinskii-Moriya interaction, and Zeeman field, representing a twisted or untwisted pair of domain walls whose separation is set by field and anisotropy. Saddle states are written as superpositions of shifted copies of $\\Theta$ with fitting parameters: two copies for Co, and for Fe a tilted-wall ansatz with uniform angle $\\varphi_0$ plus end corrections. This ansatz supplies the end-energy $E_{\\mathrm{end}}$ and domain-wall energy $E_{\\mathrm{DW}}$ of a semi-infinite chain, which are combined with a coherent-rotation formula for short chains to give energy barriers. Frequency prefactors are estimated from the single-spin transition state theory result multiplied by an effective number of reversing moments $N_{\\mathrm{eff}}$ and, for the ground-state-to-local-minimum transition, a determinant ratio that produces exponential growth with $N$. The method turns a $2N$-dimensional saddle-point search into one-dimensional curve fitting.","core_discovery":"The paper claims that for Co chains the reversal path is essentially planar rotation in the $XY$-plane, while for Fe chains the saddle configurations lie in a plane tilted by a uniform angle $\\varphi_0$, producing domain walls intermediate between Bloch and N\\'eel walls. For each system the energies of the ground state, local minimum, and the two saddle points are approximated as combinations of a twisted or untwisted domain-wall-pair solution $\\Theta(x)$ of the continuum sine-Gordon-like equation, with edge corrections from the free ends. The resulting barrier formulas reproduce the geodesic nudged elastic band barriers except in the narrow length window where the reversal mechanism switches, and the prefactor formulas reproduce the linear growth for short chains, the saturation of $\\nu_{02}$, and the exponential growth of $\\nu_{01}$ with chain length $N$. The authors state that this analytical framework qualitatively describes the numerical dependencies and enables coercive-force estimates across a broad parameter range, so it should transfer to a wide class of one-dimensional magnetic systems described by the same Hamiltonian.","pith_inferences":["One could invert the analytic coercive-force formulas against measured switching fields to extract effective values of $J$, $K$, $E$, and $D_z$ for a given surface, a possibility the paper notes but does not carry out.","The sharp peaks in the numerically computed prefactors near the crossover lengths signal a near-flat Hessian direction where harmonic transition state theory breaks down; a more complete reaction-rate theory with dissipative, dynamical prefactors would test whether the exponential growth of $\\nu_{01}$ survives beyond the harmonic approximation.","Because the ansatz treats reversal as a one-parameter family of domain-wall pairs, the same fitting strategy might transfer to artificial spin chains or nanowires with modulated anisotropy, where the continuum solution would change but the superposition method would not.","The predicted exponential prefactor growth partly compensates the linear barrier growth with $N$, so the effective reversal time versus chain length may be much flatter than barrier-only estimates suggest; this is an implication the paper leaves implicit."],"forward_implications":["For Co chains longer than about 17 to 24 atoms and Fe chains longer than about 16 atoms, reversal proceeds by domain-wall nucleation rather than coherent rotation; the analytic formulas pick the lower of the two mechanisms and so predict the crossover length.","Frequency prefactors for finite Co and Fe chains are of order $10^{12}$ to $10^{13}$ Hz, several orders above the often-used $10^9$ Hz estimate, and $\\nu_{01}$ grows exponentially with chain length while $\\nu_{02}$ saturates; this changes predicted switching rates.","Coercive force can be estimated across chain length, temperature, sweeping rate, exchange integral $J$, anisotropy difference $K-|E|$, and Dzyaloshinskii-Moriya parameter $D_z$ without further simulation, and the rough estimate $\\nu_{\\downarrow\\to\\uparrow}(B_c)\\approx dB/dt$ gives $B_c$ within tens of percent.","The coercive force of the chains decreases monotonically with Dzyaloshinskii-Moriya strength, and the blocking temperatures of about 6 K lie well below the critical temperatures of 35 to 40 K, so the transition state theory picture is internally consistent.","The analytical approach is claimed to apply not only to Co and Fe chains on Pt(332) but to any one-dimensional magnet described by the same Hamiltonian."],"supporting_citations":[{"why":"Supplies the DFT Hamiltonian parameters $J$, $K$, $E$, $D$, $\\mu$, and angles for infinite Co and Fe chains on Pt(332), which the finite-chain calculations inherit.","marker":"[45]"},{"why":"Establishes the zero-field Co-chain reversal mechanisms, end rotation angles, and N\\'eel-wall energetics that the field-dependent model extends.","marker":"[43]"},{"why":"Provides the zero-field Fe-chain domain-wall structure and the tilted-plane angle $\\varphi_0$ formula that the present work carries into finite field.","marker":"[50]"},{"why":"Introduces the geodesic nudged elastic band method used to compute the energy barriers between ground, local-minimum, and saddle states.","marker":"[65]"},{"why":"Gives the harmonic-approximation transition state theory expression for the frequency prefactor, Eq. (7), used for all numerical prefactor results.","marker":"[66]"},{"why":"Supplies the continuous-model energy of a pair of domain walls, the building block of the twisted and untwisted solution $\\Theta(x)$.","marker":"[70]"},{"why":"Provides the statistical-mechanics treatment of domain-wall-pair fluctuations in a field, the basis for the analytic barrier and prefactor estimates.","marker":"[71]"},{"why":"Underpins the transition state theory rate formula and the caution that harmonic TST overestimates prefactors, which the paper uses to frame its coercive-force estimates.","marker":"[63]"},{"why":"Gives the experimental reference point for ferromagnetic Co chains and the commonly used $10^9$ Hz prefactor estimate that the paper's TST values supersede.","marker":"[13]"}],"fun_headline_variants":["Analytic theory unlocks magnetization reversal predictions for atomic chains","Co and Fe chain reversal now analytically predictable across lengths","New analytic approach captures reversal barriers and prefactors in atomic chains","Short chains flip together, long chains via domain walls: analytic prediction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the true magnetization-reversal path stays inside a restricted manifold—in-plane rotation for Co and uniformly tilted-plane rotation for Fe—with saddles built from domain-wall-pair solutions, so that if the real minimum-energy path leaves that manifold at some chain length or field, the analytic barriers and prefactors would be wrong.","fun_headline_variants_meta":{"raw":{"variants":["Analytic theory unlocks magnetization reversal predictions for atomic chains","Co and Fe chain reversal now analytically predictable across lengths","New analytic approach captures reversal barriers and prefactors in atomic chains","Short chains flip together, long chains via domain walls: analytic prediction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000844,"raw_usage":{"total_tokens":3707,"prompt_tokens":1008,"completion_tokens":2699,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":624,"completion_tokens_details":{"reasoning_tokens":2631}},"tokens_in":624,"tokens_out":2699,"duration_ms":16741,"temperature":1.0,"reasoning_tokens":2631,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:19:50.140200+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a geodesic nudged elastic band calculation for a Co chain of $N=30$ atoms at $B_y=2$ T and compare the converged minimum-energy path and barrier with the $XY$-plane domain-wall-pair predictions of Eqs. (22)-(25), and for an Fe chain do the same against the tilted-plane predictions of Eqs. (38)-(39); if the path leaves the assumed plane or the barrier differs outside the mechanism-crossover window, the ansatz is wrong.","supporting_citations":[{"cited_title":"Schweflinghaus, B","cited_arxiv_id":null,"evidence_quote":"Supplies the DFT Hamiltonian parameters $J$, $K$, $E$, $D$, $\\mu$, and angles for infinite Co and Fe chains on Pt(332), which the finite-chain calculations inherit."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the zero-field Co-chain reversal mechanisms, end rotation angles, and N\\'eel-wall energetics that the field-dependent model extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the zero-field Fe-chain domain-wall structure and the tilted-plane angle $\\varphi_0$ formula that the present work carries into finite field."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the geodesic nudged elastic band method used to compute the energy barriers between ground, local-minimum, and saddle states."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the harmonic-approximation transition state theory expression for the frequency prefactor, Eq. (7), used for all numerical prefactor results."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the continuous-model energy of a pair of domain walls, the building block of the twisted and untwisted solution $\\Theta(x)$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the statistical-mechanics treatment of domain-wall-pair fluctuations in a field, the basis for the analytic barrier and prefactor estimates."},{"cited_title":"H¨ anggi, P","cited_arxiv_id":null,"evidence_quote":"Underpins the transition state theory rate formula and the caution that harmonic TST overestimates prefactors, which the paper uses to frame its coercive-force estimates."},{"cited_title":"Gambardella, A","cited_arxiv_id":null,"evidence_quote":"Gives the experimental reference point for ferromagnetic Co chains and the commonly used $10^9$ Hz prefactor estimate that the paper's TST values supersede."}],"review_version":1}