{"id":"eb5044a5-b4e4-40f7-9036-d44d5865a251","arxiv_id":"2501.17057","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Inhomogeneous electric fields can deterministically stabilize and switch polar textures in PbTiO3 and drive domain walls at speeds up to about 3000 m/s.","lead":"Using atomistic simulations, this paper shows that specially shaped electric fields that vary in space and time can create, erase, and move tiny polarization patterns in the ferroelectric PbTiO3. Because these patterns could serve as nanoscale information carriers, the control scheme is a step toward practical ferroelectric nanoelectronics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed intrinsic ~3000 m/s speed limit is inferred from a single moving-field condition in 16 u.c. periodic cells; no amplitude/temperature scan or finite-size check is presented, so the universality claim is unsupported.","rationale":"The paper's core demonstration—that modulated electric fields stabilize stripes, skyrmion lattices, and bubbles in bulk PbTiO3 and reversibly switch them—is credible within the second-principles model; the figures show metastable textures after field removal, which is the key evidence for deterministic control. I do not object to that part. The risk concentrates on the dynamics section: the ~3000 m/s limiting velocity is presented as an intrinsic upper bound, independent of field amplitude and temperature, but the manuscript shows no systematic scan. This is not an external-consensus disagreement; it is an internal evidence gap. The finite-cell issue compounds it: with lambda = 16 u.c. and a commensurate supercell, the wall circulates on a ring and the fragmentation threshold may reflect periodic-image interactions or phase-locking of the imposed potential rather than an intrinsic material velocity. A targeted scan in a larger cell at varied A and T would settle whether the universal claim holds. Since the reader already conditioned the verdict on exactly this kind of evidence, I recommend keeping the conditional verdict rather than moving to accept or reject.","tokens_in":11098,"tokens_out":7879,"duration_ms":75851,"concrete_test":"Run the moving-field protocol (Eq. 3) with lambda_x = 16 u.c. at two additional amplitudes (A = 1 and 4 MV/cm) and two additional temperatures (T = 100 and 200 K), in a supercell with Lx = 64 u.c. while keeping the field wavelength fixed, and measure the wave velocity at which the domain texture first fragments. If the fragmentation threshold shifts with A or T, or differs by more than ~10% from the value in the Lx = 16 u.c. cell, the claim of an intrinsic, parameter- and finite-size-independent speed limit fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The control/nucleation results are plausible, and the qualitative diffuse-boundary picture is supported by the snapshots. The load-bearing weakness is the speed-limit claim. In Section III, 'Dynamics of ferroelectric domains', the paper states that with lambda_x = 16 u.c. and A = 2 MV/cm the limiting velocity is ~3000 m/s, and immediately asserts that this value 'remains unaffected by changes in the electric field magnitude or temperature' and constitutes an 'intrinsic upper bound.' No figure, table, or supplementary scan is provided to support the amplitude/temperature independence; the text reports essentially one trajectory at one amplitude and at most a small set of temperatures for the lag, not for the fragmentation threshold. In addition, the simulation cell is commensurate with the modulation (lambda_x equal to the cell period), so the domain wall moves around a ring and interacts with its periodic image; a threshold for losing phase-locking on a 16 u.c. ring could be a property of the driving protocol and finite cell rather than an intrinsic material limit. Because the abstract and conclusions advertise an 'upper limit' and 'intrinsic upper bound' as central results, this unsupported universality claim is the main risk to the paper's headline dynamics.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports second-principles molecular dynamics simulations of bulk PbTiO3 subjected to spatially and temporally modulated electric fields. The authors show that cosine-modulated fields (Eq. 1) stabilize stripe domains, skyrmion lattices, vortex/antivortex lattices, and bubble domains, and that these textures can be switched reversibly by changing the field profile (Fig. 3). They then study domain-wall motion under a moving cosine field (Eq. 3), reporting a lag at the onset of motion that disappears at steady state, an immediate stop upon field removal, diffuse-boundary nucleation, behavior they describe as consistent with creep dynamics, and an 'intrinsic upper bound' on domain-wall speed of about 3000 m/s that is claimed to be independent of field amplitude and temperature. The abstract and conclusions advertise the speed limit as a central result.","tokens_in":11346,"tokens_out":3980,"duration_ms":38653,"significance":"If the control scheme works as described, the paper offers a simple and potentially useful route to deterministic, in-situ manipulation of polar textures in a prototypical ferroelectric, which is of interest for nanoelectronic applications. The static and switching results are the strongest part: they are grounded in a DFT-fitted second-principles model (3644 configurations, DFPT harmonic terms) and are supported by qualitative real-space snapshots for several distinct field profiles. The dynamic claims, especially the universal speed limit, are the most novel but are not backed by quantitative evidence in the current manuscript. The paper would be significantly strengthened by systematic parameter scans and finite-size checks, and by a quantified creep-law analysis or a clearly softened interpretation.","major_comments":[{"comment":"The central dynamical claim, that the limiting velocity is an intrinsic upper bound of about 3000 m/s that is independent of field amplitude and temperature, is not supported by the data presented. The text states that with λx = 16 u.c. and A = 2 MV/cm the limiting velocity is approximately 3000 m/s and immediately asserts that this value 'remains unaffected by changes in the electric field magnitude or temperature' and constitutes an 'intrinsic upper bound', but no figure, table, or supplementary scan of amplitude or temperature is provided. The conclusion appears to rest on a single trajectory at one amplitude and one periodicity; please provide a systematic scan of A and T, define the fragmentation threshold quantitatively, and report uncertainties.","section":"Section III, 'Dynamics of ferroelectric domains'"},{"comment":"The identification of creep dynamics is qualitative. The text says that doubling the domain periodicity requires doubling the field to restore mobility and that this 'is consistent with the predictions of the creep law', but no creep-law fit is presented: there is no measurement of velocity versus field or temperature, no extraction of the creep exponent μ, and no analysis of pinning. Either provide such a quantitative analysis or explicitly label the statement as a qualitative analogy rather than a demonstrated creep-law behavior.","section":"Section III, 'Dynamics of ferroelectric domains'"},{"comment":"The simulation setup is commensurate with the applied modulation: the moving cosine field has λx = 16 u.c. and, based on the methods and figure description, the supercell period is also 16 u.c., so the domain walls propagate around a ring and interact with their periodic images. The threshold for losing phase-locking under this protocol may therefore be a property of the finite cell and the driving waveform rather than an intrinsic material limit. Please perform finite-size checks (for example, 32 and 64 u.c. cells with the same λx) and, if possible, compare with a protocol in which the domain wall is not wrapped around the cell.","section":"Section III and Fig. 4"},{"comment":"The asymmetric inertial response is asserted mainly on the basis of visual inspection of snapshots. The text reports a lag of 4 ps at T = 5 K, reduced by half at 100 K and negligible at 200 K, but no quantitative time series of domain-wall position versus the field node, no velocity curves, and no statistical measure over independent runs are provided. To make the asymmetry claim (onset lag vs. immediate stop) robust, please provide time-resolved position data for both the turn-on and turn-off transients, with multiple thermal seeds.","section":"Section III, 'Dynamics of ferroelectric domains'"}],"minor_comments":[{"comment":"The word 'strucutre' should be 'structure'.","section":"Fig. 3 caption"},{"comment":"'computational approches' should be 'computational approaches'.","section":"Introduction"},{"comment":"The Gaussian field expression uses µ−r−vt without clarifying whether r and vt are vectors; please define the norm or vector operation explicitly.","section":"Eq. (2)"},{"comment":"It would be helpful to state the thermostat used for the conventional MD simulations and the number of independent trajectories or thermal seeds for the dynamical runs.","section":"Methods"},{"comment":"The statement that for λ = 8 u.c. the dipoles 'tilt towards the direction perpendicular to the modulation' is qualitative; a quantitative order parameter (e.g., the average in-plane polarization angle) would strengthen the description.","section":"Section III, 'Stabilization of stripe domains'"}],"recommendation":"major_revision","confidential_remarks":"The static control and switching results are plausible and may be publishable, but the dynamic claims—especially the 'intrinsic upper bound'—need substantial additional evidence. The heavy reliance on the authors' own APEX paper (Ref. 25) and related BaTiO3 preprint (Ref. 38) is understandable but should not substitute for independent validation. There is no data availability statement; for a simulation paper, sharing input files and analysis scripts would greatly increase confidence. If the authors can supply the requested amplitude/temperature scans, finite-size checks, and quantitative creep/inertia analyses, the paper could become acceptable; otherwise the headline claims should be softened."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi [Name],\n\nQuick take on arXiv:2501.17057. The paper demonstrates that inhomogeneous electric fields, both static and time-modulated, can stabilize, reversibly switch, and translate polar textures in bulk PbTiO3 using second-principles molecular dynamics. The static control results are genuinely new and convincing: the field profiles in Figs. 1 and 2 map directly onto the stabilized stripe, skyrmion, and vortex/antivortex textures, and the reversible switching in Fig. 3 is a clean demonstration of in situ control. I want to give credit for that — it's a useful capability the field was missing.\n\nThe dynamic section is where the overreach is. The asymmetric inertial response — lag at onset, immediate stop after removal — is interesting and plausible, and the diffuse-boundary nucleation picture is consistent with the snapshots. But the paper claims the limiting wall velocity (~3000 m/s for 16 u.c. domains) is independent of field amplitude and temperature, calling it an 'intrinsic upper bound.' I can't find support for that in the manuscript. The text reports essentially one condition: A=2 MV/cm, lambda_x=16 u.c., presumably T=5 K. There's no amplitude sweep, no temperature scan for the fragmentation threshold, and no finite-size check. Worse, the simulation cell is commensurate with the modulation, so the domain wall moves on a ring and sees its own periodic image — the threshold could easily be a finite-size or protocol artifact. That's a load-bearing claim for the abstract and conclusions, and it's unsupported.\n\nThe 'creep dynamics' assertion is softer: it's inferred from mobility scaling (double the periodicity, double the field) rather than a quantitative creep-law fit. That's acceptable if labeled qualitative, but it shouldn't be presented as a confirmed mechanism.\n\nNo code or data is provided, which is a reproducibility gap but not a fatal one.\n\nOverall: the static control work is solid and worth building on, and the dynamic observations are thought-provoking. But the universal speed-limit claim needs either more data or more cautious language. I'd send this to a good referee — the core result is important enough to deserve careful review. If you handle it, ask the authors to show the amplitude/temperature and finite-size dependence of the fragmentation threshold, or replace 'intrinsic' with 'for the conditions studied.'\n\nBest,\n[Your name]","headline":"The static control of polar textures via modulated fields is solid and new, but the claimed intrinsic 3000 m/s speed limit is unsupported by the data shown.","tokens_in":11894,"tokens_out":3424,"would_cite":true,"duration_ms":31930,"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":"Spatially modulated electric fields deterministically nucleate, switch, and move ferroelectric domain textures in bulk PbTiO3, with creep-like wall motion and an intrinsic speed ceiling near 3000 m/s.","keywords":["ferroelectric domain walls","polar topological textures","inhomogeneous electric fields","second-principles molecular dynamics","domain-wall creep","skyrmion bubbles","PbTiO3","domain-wall inertia"],"falsifier":"Repeat the moving-field MD protocol in supercells of 64 or more unit cells, or with open boundaries, sweeping field amplitude and temperature over wider ranges; if the fragmentation velocity (near 3000 m/s for $\\lambda_x = 16$ u.c.) shifts with cell size, amplitude, or temperature, the claimed intrinsic bound is an artifact of the periodic supercell rather than a material property.","tokens_in":10906,"feed_emoji":"⚡","tokens_out":9651,"duration_ms":79352,"temperature":0.7,"pith_summary":"This paper argues that inhomogeneous electric fields—spatially and temporally modulated field patterns—can deterministically create, erase, and displace ferroelectric domain textures in bulk PbTiO3, something homogeneous fields cannot do with the same selectivity. Using second-principles molecular dynamics, the authors nucleate stripe domains, skyrmion lattices, vortex/antivortex lattices, and individual bubble domains, then switch between these states reversibly by changing the field pattern. They further show that a traveling sinusoidal field moves domain walls through a diffuse-boundary nucleation mechanism consistent with creep dynamics, with an onset lag that depends on temperature and vanishes once steady motion is reached. The central quantitative claim is an intrinsic upper bound of about 3000 m/s for domain-wall propagation, independent of field amplitude and temperature. If true, this establishes polar textures as fast, rewritable information carriers whose speed ceiling exceeds magnetic skyrmions.","feed_headline":"Traveling electric fields push ferroelectric domains at 3000 m/s","feed_subtitle":"Simulations show reversible texture writing and a speed ceiling set by the material, not the field.","key_machinery":"The central object is the periodic, traveling electric-field waveform of Eq. (3), a cosine with independently chosen spatial periodicities $\\lambda_x,\\lambda_y,\\lambda_z$ and temporal period $T$, whose phase velocity $v = (1/\\lambda_x^2 + 1/\\lambda_y^2 + 1/\\lambda_z^2)^{-1/2}/T$ acts as a moving handle on the polarization texture. The field is coupled to the atoms through the Born effective charge tensor $Z^*$, giving each atom a force $F_{\\kappa,\\beta} = Z^*_{\\kappa,\\alpha\\beta} E_\\alpha(\\mathbf{r})$; in effect the field imposes a traveling periodic potential landscape on the material. The second-principles atomistic model, fitted to DFT data, supplies the energy landscape on which this potential acts, and the molecular dynamics then reveal how the texture responds. The machinery's job is to show that the polarization's response—nucleation of a new domain patch, wall advance, lag, and arrest—is controlled by this external waveform, and that the speed ceiling follows from the field's phase velocity rather than from material inertia.","core_discovery":"Under a space- and time-modulated electric field of the form $E(\\mathbf{r},t) = A \\cos(2\\pi x/\\lambda_x + 2\\pi y/\\lambda_y + 2\\pi z/\\lambda_z + 2\\pi t/T)$, 180° domain walls in PbTiO3 move with the wave rather than by uniform sideways advance. The paper reports that this motion proceeds by diffuse-boundary nucleation: a patch of the new domain appears ahead of the wall and the wall then follows, a signature previously identified for homogeneous fields. The response is asymmetric: a lag of about 4 ps at 5 K (for a 16 unit-cell periodicity) appears when the field is switched on, but the domains halt immediately when the field is removed, with no residual momentum. The lag shrinks with rising temperature and becomes negligible by 200 K. Doubling the domain periodicity to 32 unit cells leaves the walls immobile unless the field amplitude is also doubled, in line with the creep law. Above a wave velocity of about 3000 m/s the domain structure fragments chaotically, and this limiting velocity does not change with field amplitude or temperature; the authors read it as an intrinsic, material-related bound on domain-wall motion.","pith_inferences":["The same waveform-control strategy should transfer to other ferroelectric perovskites and to ferroelectric/dielectric superlattices, where the speed ceiling may differ because domain walls are stabilized electrostatically rather than by the external field.","The reported onset lag at 5, 100, and 200 K could be used to extract an activation energy for wall nucleation; such an Arrhenius-style analysis would test the creep interpretation quantitatively.","A phased array of electrodes or a moving laser-generated field pattern could mimic the traveling cosine wave in experiments, turning the predicted 3000 m/s ceiling into a measurable quantity.","If the ceiling is set by the domain wall's internal stiffness, strain engineering that softens the wall should shift the limit, offering a materials-design knob that the paper does not itself explore."],"forward_implications":["A sequence of modulated-field pulses can write a stripe domain, erase it with a homogeneous field, rewrite it along another direction, and convert it into a skyrmion lattice without changing growth conditions.","A traveling sinusoidal field can displace 180° domain walls and bubble domains coherently; bubble domains move isotropically by elongation and contraction, like liquid droplets.","Domain-wall speeds saturate near 3000 m/s under these conditions, so faster operation would require a different material or a modified domain-wall energy landscape.","Larger domains need proportionally larger driving fields, so field amplitude can selectively move small domains while leaving larger ones stationary.","The onset lag that grows at low temperature and the instantaneous stop on field removal identify nucleation-barrier-controlled dynamics rather than momentum-driven motion."],"supporting_citations":[{"why":"Previous method using spatially modulated acoustic phonons to tailor polar topologies; this paper extends the q≠0 modulation idea to electric fields.","marker":"[25]"},{"why":"Supplies the diffuse-boundary nucleation-and-growth mechanism used to interpret domain-wall advance.","marker":"[44]"},{"why":"Provide the creep-law scaling (temperature and field dependence of wall mobility) that the authors invoke for larger immobile domains.","marker":"[26, 27]"},{"why":"Reports the intrinsic inertial response of ferroelectric domain walls in homogeneous fields, which this work complements by showing an asymmetric onset-only lag.","marker":"[46]"},{"why":"Shows motion and teleportation of polar bubbles, supporting the Gaussian-pulse single-bubble control and the amoeba-like bubble dynamics.","marker":"[15]"},{"why":"Provides theoretical guidelines for creating and tuning electric skyrmion bubbles with AFM-like fields, grounding the Gaussian-field nucleation.","marker":"[39]"},{"why":"Supplies the DFT and second-principles simulation framework used for the PbTiO3 model.","marker":"[30]"},{"why":"Establishes the Born effective charge relation F = Z*E that links the applied field to atomic forces.","marker":"[35]"},{"why":"Recent fast magnetic skyrmion motion (900 m/s) used as benchmark showing ferroelectric walls are faster.","marker":"[29]"}],"fun_headline_variants":["Domain walls ride electric waves, hit 3000 m/s speed limit","Asymmetric inertia: ferroelectric walls lag on start, stop instantly","Wave-driven ferroelectric walls obey creep law, cap at 3000 m/s","Electric wave moves domain walls, sets 3000 m/s speed ceiling","Ferroelectric domain walls show asymmetric inertia under electric waves"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The simulations assume that periodic supercells of 16 to 32 unit cells, described by the fitted second-principles potential, capture the intrinsic bulk dynamics of PbTiO3—including the claimed speed limit.","fun_headline_variants_meta":{"raw":{"variants":["Domain walls ride electric waves, hit 3000 m/s speed limit","Asymmetric inertia: ferroelectric walls lag on start, stop instantly","Wave-driven ferroelectric walls obey creep law, cap at 3000 m/s","Electric wave moves domain walls, sets 3000 m/s speed ceiling","Ferroelectric domain walls show asymmetric inertia under electric waves"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000694,"raw_usage":{"total_tokens":3128,"prompt_tokens":924,"completion_tokens":2204,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":540,"completion_tokens_details":{"reasoning_tokens":2108}},"tokens_in":540,"tokens_out":2204,"duration_ms":13544,"temperature":1.0,"reasoning_tokens":2108,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T04:54:19.766114+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the moving-field MD protocol in supercells of 64 or more unit cells, or with open boundaries, sweeping field amplitude and temperature over wider ranges; if the fragmentation velocity (near 3000 m/s for $\\lambda_x = 16$ u.c.) shifts with cell size, amplitude, or temperature, the claimed intrinsic bound is an artifact of the periodic supercell rather than a material property.","supporting_citations":[{"cited_title":"Bastogne, F","cited_arxiv_id":null,"evidence_quote":"Previous method using spatially modulated acoustic phonons to tailor polar topologies; this paper extends the q≠0 modulation idea to electric fields."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the diffuse-boundary nucleation-and-growth mechanism used to interpret domain-wall advance."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the intrinsic inertial response of ferroelectric domain walls in homogeneous fields, which this work complements by showing an asymmetric onset-only lag."},{"cited_title":"Prokhorenko, Y","cited_arxiv_id":null,"evidence_quote":"Shows motion and teleportation of polar bubbles, supporting the Gaussian-pulse single-bubble control and the amoeba-like bubble dynamics."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides theoretical guidelines for creating and tuning electric skyrmion bubbles with AFM-like fields, grounding the Gaussian-field nucleation."},{"cited_title":"Gonze, B","cited_arxiv_id":null,"evidence_quote":"Supplies the DFT and second-principles simulation framework used for the PbTiO3 model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Recent fast magnetic skyrmion motion (900 m/s) used as benchmark showing ferroelectric walls are faster."}],"review_version":1}