{"id":"19aff55c-3743-4b87-b6b9-71bbfede1c7f","arxiv_id":"1908.01125","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Electric fields drive a reversible insulator-metal transition and lattice expansion in La2-xSrxCoO4, attributed to proliferation of high-spin Co3+ states.","lead":"Applying a voltage to the layered cobaltate La2-xSrxCoO4 turns it from an insulator into a metal and makes the crystal expand. If confirmed as non-thermal, this would be a new way to control material phases with electric fields for memory and sensor applications.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The non-thermal claim is not secured: the geometry argument does not exclude Joule heating, and the paper itself notes resistive heating during measurements.","rationale":"The reader's weakest-assumption analysis correctly identifies the exclusion of Joule heating as the load-bearing point. My review confirms this: the paper's only explicit evidence against heating is the geometry-dependence argument in the Discussion, which is logically insufficient because geometry affects current density and heat flow. The manuscript itself contains a sentence acknowledging high current and heating in the metallic phase, and no temperature measurement is reported. Therefore the central claim of a non-thermal, spin-state-driven IMT is not supported by the presented data. The observation of voltage-induced resistance switching and strain may be real, but the interpretation and the title overreach. Since the reader's verdict is REJECT and my concern is the same load-bearing gap, no change to the verdict is needed. A concrete test—time-resolved pulsed measurements with local thermometry—would settle the issue. My agreement with the reader is full: the same weakest assumption is identified.","tokens_in":5450,"tokens_out":1706,"duration_ms":22068,"concrete_test":"Perform time-resolved switching measurements on S1 and S2 with voltage pulses of varying width (e.g., 100 ns to 10 ms) at fixed ambient temperature, recording the threshold voltage and strain response. In parallel, measure the sample surface temperature with an infrared camera or a microfabricated resistive thermometer. If the threshold field decreases with pulse width in a manner consistent with a thermal activation law (e.g., threshold proportional to the square root of pulse duration), and if the measured surface temperature at switching reaches the temperature range where the resistance change would occur by heating alone, then the non-thermal interpretation is falsified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the IMT and volume expansion are non-thermal, caused by electric-field-induced proliferation of Co3+ high-spin states. The only argument against Joule heating is the geometry dependence of switching and strain (Discussion, Fig. S1/S2). This argument is not decisive: changing the bias geometry changes the current density, current path, and heat distribution, so the observed geometry dependence is exactly what Joule heating would predict. No local temperature is measured anywhere in the paper. Moreover, the text contains a direct admission of heating: after switching in Fig. 1(b), the measurement was done in the heating cycle 'because of high current flowing in metallic phase prevents cooling.' This indicates that the metallic filament carries substantial current and heats the sample, undermining the assumption that the transition is non-thermal. The microscopic CTFM images (Fig. 3) show growth of conducting regions with bias, which is also consistent with progressive Joule heating of filaments. The XRD peak shift (Fig. 4) under applied voltage could equally be thermal expansion. Because the spin-state mechanism is inferred only from the strain magnitude and the S1/S2 comparison, and no spin-state-sensitive measurement (e.g., XAS, magnetic susceptibility, or inelastic scattering) is provided, the entire 'electro-proliferation of HSS' interpretation rests on the unsupported exclusion of heating. If heating is responsible, the title and abstract overclaim both the mechanism and the non-thermal character.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports voltage-induced insulator-metal transition (IMT) and concurrent strain in polycrystalline La2-xSrxCoO4 (x = 0, S1 and x = 0.5, S2), which do not show temperature-driven IMT. The authors observe resistance switching in I-V and R-T curves, strain gauge signals, growth of conducting regions in conducting-tip force microscopy, and shifts of XRD peaks under applied voltage. They interpret these effects as a non-thermal isostructural IMT caused by electric-field-induced proliferation of Co3+ high-spin states, supported by a comparison of strain between S1 and S2 and by geometry-dependent switching experiments.","tokens_in":5615,"tokens_out":5376,"duration_ms":51840,"significance":"If the non-thermal spin-state mechanism were established, the paper would report a new class of electric-field-controlled phase transition with coupled electronic and structural responses, relevant for memristors and strain sensors. The raw observations—resistance switching, strain, CTFM metallic-region growth, and XRD peak shifts—are internally consistent and reproduced across two compositions. The paper contains no free-parameter fits; the strain comparison uses literature ionic radii as an external benchmark. However, the central interpretation is not directly evidenced: no temperature measurement under bias, no spin-state-sensitive probe, and the Joule-heating exclusion rests on a geometry argument that is not decisive. The strengths are the multi-scale experimental documentation and the falsifiable character of the proposed mechanism, which can in principle be tested by direct temperature and spin-state measurements.","major_comments":[{"comment":"The claim that Joule heating can be discarded because the switching and strain depend on the bias geometry is not conclusive, since changing the geometry changes the current density and heat distribution; a thermally driven filamentary transition would also show geometry-dependent thresholds and strain. No local temperature measurement is presented anywhere in the manuscript. The statement in Results (Fig. 1(b)) that the post-switching R-T measurement was performed 'in heating cycle because of high current flowing in metallic phase prevents cooling' explicitly acknowledges substantial Joule heating in the metallic state, further undermining the non-thermal claim.","section":"Discussion, geometry argument (Fig. S1/S2)"},{"comment":"The central interpretation of electro-proliferation of Co3+ high-spin states is not directly tested. No spin-state-sensitive measurement (e.g., X-ray absorption, magnetic susceptibility, or inelastic scattering) is performed under applied field, and the in-situ XRD peak shift (Fig. 4) is equally consistent with thermal expansion. The spin-state mechanism is inferred only from the strain magnitude and the S1/S2 comparison, which are indirect. Without excluding heating and without probing the spin state, the proposed mechanism remains speculative rather than established.","section":"Results and Discussion, spin-state mechanism"},{"comment":"The quantitative prediction used to support the spin-state mechanism is internally inconsistent: the text says that if all Co3+ switched to HSS the strain should be '~50% larger' in S1, but since S1 contains 100% Co3+ and S2 contains 50% Co3+, complete conversion would give a factor-of-two (100%) larger expansion in S1, not 50%. The observed '~40% larger' value therefore does not match the composition-based expectation as stated, and no error bars are provided for the strain values, making it difficult to assess the significance of the comparison.","section":"Discussion, strain comparison"}],"minor_comments":[{"comment":"The phrase 'to one's imagination RT curves shows breakdown' is unclear and appears to be a typographical error; please rephrase.","section":"Results, first paragraph"},{"comment":"The caption refers to 'outer leads' and 'inner leads' but the inset is not described in enough detail for the reader to understand the measurement geometry; please clarify.","section":"Fig. 1(b) caption"},{"comment":"The paper does not report error bars on the strain values or the switching fields, which is important given that the quantitative comparison between S1 and S2 is used to support the mechanism.","section":"Strain measurements"},{"comment":"The ionic radii for Co3+ LSS (0.545 Å) and HSS (0.61 Å) are cited to ref. 19, but the original source (e.g., Shannon) should be cited for these standard values.","section":"References"},{"comment":"The abstract and title state 'non thermal' as an established fact, while the body of the paper only proposes this interpretation; the wording should be conditional given the evidence presented.","section":"Abstract and title"}],"recommendation":"reject","confidential_remarks":"The raw data appear plausible and the multi-scale documentation is a strength, but the central physical claim of a non-thermal, spin-state-driven transition is not supported by the measurements presented. The missing evidence—direct local temperature measurement and a spin-state probe under field—is central to the claim and would require substantial new experiments rather than a routine revision. I would not encourage a standard major revision; a resubmission with such data could be considered."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The one genuinely new thing here is the observation: La2-xSrCoO4, which has no thermal IMT, shows voltage-driven resistance switching, growth of conducting regions, and a measurable volume expansion all at once. That is worth knowing about. The experimental package is also respectable—simultaneous I-V and strain-gauge data, CTFM images showing conducting regions growing with bias, and in-situ XRD peak shifts. If the switching effect itself is real, that is a useful starting point for someone else.\n\nThe soft spot is exactly where the reader put it: the claim that this is non-thermal and driven by electro-proliferation of Co3+ high-spin states is not secured. The only argument against Joule heating is the geometry dependence of switching and strain. That argument doesn't do the work. Changing the bias geometry changes current density, current path, and heat distribution, so the observed differences are just as consistent with Joule heating as with a field-driven electronic mechanism. And the paper itself notes that after switching, the high current in the metallic phase prevents cooling—which is a direct admission that the sample gets hot. Without a local temperature measurement, the non-thermal conclusion is not supported.\n\nThe spin-state mechanism is inferred entirely from the strain magnitude and the S1/S2 comparison. But that comparison is made at different temperatures chosen because the resistances are similar, which is post hoc, and there is no spin-state-sensitive probe (XAS, magnetic susceptibility, etc.) under field. So the mechanism is a plausible hypothesis, not a demonstrated result. The XRD shift could simply be thermal expansion. Error bars are also missing throughout, which makes the 40% strain difference between S1 and S2 hard to evaluate.\n\nAll that said, the core observation of electrically driven switching and strain in this system is new and not obviously wrong. The paper deserves a serious referee, but not as a finished claim about non-thermal spin-state electro-proliferation. A serious referee should push for either direct temperature measurements, a spin-state probe under bias, or a much more careful framing that separates the observation from the interpretation. If the authors can show the effect is genuinely not thermal, this would be important; if they cannot, the paper still might be publishable as a careful electrical-switching study with the mechanism left open. For now, read it for the data, not the conclusions.","headline":"A potentially interesting multi-scale observation of electrically driven switching and strain in layered cobaltates, but the non-thermal spin-state interpretation is not backed by the evidence and the paper overreaches.","tokens_in":6194,"tokens_out":1387,"would_cite":false,"duration_ms":16584,"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":"Applying an electric field drives a non-thermal, isostructural insulator-metal transition in the layered cobaltates La2-xSrxCoO4, with a lattice expansion of up to 0.3 percent.","keywords":["insulator-metal transition","electric field","spin-state transition","cobaltates","electro-strain","isostructural","resistive switching","La2-xSrxCoO4"],"falsifier":"Install a micro-thermometer on the sample (or use Raman or infrared thermometry) while measuring current, voltage, and strain: if the local temperature rises high enough to depopulate the low-spin state at the switching point, or if a Joule-heating simulation of the three contact geometries reproduces the observed switching and strain, then the non-thermal claim is contradicted.","tokens_in":5200,"feed_emoji":"⚡","tokens_out":14189,"duration_ms":110689,"temperature":0.7,"pith_summary":"The paper claims that an applied voltage, not heat, can switch the layered cobaltates La$_{2-x}$Sr$_x$CoO$_4$ ($x=0$ and $0.5$) from insulator to metal while the crystal structure stays intact. The switch is accompanied by a volume expansion of up to 0.3 percent, seen both as a strain-gauge signal and as X-ray diffraction peak shifts. The authors attribute the conductivity jump and the expansion to an electric-field-driven conversion of Co$^{3+}$ ions from small, insulating low-spin states ($t_{2g}^6 e_g^0$) to larger, metallic high-spin states ($t_{2g}^4 e_g^2$). Since these compounds do not show a temperature-driven insulator-metal transition, an electric field alone achieving this would offer a new route to fast, low-power electronic and electromechanical devices.","feed_headline":"Electric field flips cobaltate to metal, stretches lattice 0.3%","feed_subtitle":"A non-thermal, isostructural switch in La2-xSrxCoO4 hints at fast, low-power memory and actuator devices.","key_machinery":"The central mechanism is an electric-field-driven spin-state transition of Co$^{3+}$ ions. In these cobaltates Co$^{3+}$ can occupy a low-spin ($t_{2g}^6 e_g^0$), intermediate-spin, or high-spin ($t_{2g}^4 e_g^2$) state; the states are close in energy and differ in ionic radius. The paper proposes that the applied field proliferates the high-spin state: its larger radius (0.61 Å versus 0.545 Å) produces the observed isostructural volume expansion, and the accompanying electronic rearrangement produces the insulator-metal transition. The evidence chain combines a strain gauge, conducting-tip force microscopy, and voltage-dependent X-ray diffraction to connect the macroscale, microscale, and atomic-scale signatures.","core_discovery":"The paper reports the first evidence that an applied electric field induces an insulator-metal transition in the layered cobaltates La$_{2-x}$Sr$_x$CoO$_4$, which otherwise show no temperature-driven IMT. Macroscopically, the resistance drops at a threshold voltage, and the switched state shows a positive temperature coefficient of resistance, the signature of a metal. Conducting-tip atomic force microscopy shows metallic regions growing with bias, and voltage-dependent X-ray diffraction shows lattice expansion with no structural symmetry change. The strain is about 40% larger in the $x=0$ sample, which contains twice the Co$^{3+}$ fraction of $x=0.5$, matching the prediction that each high-spin Co$^{3+}$ ion contributes to the expansion. The proposed mechanism is electro-proliferation of Co$^{3+}$ high-spin states: the field converts low-spin ions (ionic radius 0.545 Å) into high-spin ions (0.61 Å), expanding the lattice and making it metallic while preserving the structure.","pith_inferences":["If the mechanism is truly spin-state proliferation, then magnetic fields or external strain, which also tune spin-state energies, should shift the switching threshold; this is a testable prediction the paper does not make.","The geometry dependence of switching is also compatible with current-density-dependent Joule heating, so a direct local temperature measurement under bias would settle the non-thermal claim.","The same electric-field route to spin-state conversion may carry over to other mixed-spin cobaltates and spin-crossover compounds, suggesting a generic electronic control strategy beyond this material family."],"forward_implications":["An electric field can act as a continuous control knob for spin-state population in cobaltates, coupling resistivity and lattice volume in a single switch.","Because the transition is isostructural and argued to be non-thermal, it could switch faster and dissipate less heat than conventional VO2-based insulator-metal transition devices.","The strain scales with Co$^{3+}$ content, so compositions can be tuned to optimize electro-strain for actuation while preserving the switching behavior.","The reversible, threshold-type resistive switching, rather than an irreversible breakdown, points to a genuine electronic phase transition usable in memristive devices."],"supporting_citations":[{"why":"Provides the benchmark isostructural metal-insulator transition in VO2 that the paper's non-thermal transition is contrasted with.","marker":"[6]"},{"why":"Reports a current-induced insulator-metal transition in Ca2RuO4 attributed to structure rather than Joule heating, the closest precedent for structure-coupled electric switching.","marker":"[7]"},{"why":"Prior characterization of La2-xSrxCoO4 establishing the mixed low-spin and high-spin Co3+ states at room temperature that the electric field acts on.","marker":"[10]"},{"why":"Ab initio study giving the ionic radii of Co3+ high-spin (0.61 Å) and low-spin (0.545 Å) states, the quantitative basis for the predicted roughly 6% volume difference.","marker":"[19]"},{"why":"Supplies the conducting-tip force microscopy technique with about 10 nm resolution used to image the growth of metallic regions.","marker":"[16]"},{"why":"Documents Joule heating as the origin of resistive switching in magnetite, the alternative explanation the paper must rule out.","marker":"[18]"}],"fun_headline_variants":["Voltage triggers cobaltate metal switch with 0.3% strain","Electric field induces insulator-metal transition in cobaltate","Non-thermal IMT: cobaltate expands 0.3% under voltage","Voltage-driven IMT in layered cobaltate, no heat needed","Electro-proliferation of high-spin Co3+ drives cobaltate IMT"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes that the geometry dependence of the switching and strain rules out Joule heating, but the sample temperature is never directly measured; if the effect is actually thermal, the non-thermal spin-state mechanism fails.","fun_headline_variants_meta":{"raw":{"variants":["Voltage triggers cobaltate metal switch with 0.3% strain","Electric field induces insulator-metal transition in cobaltate","Non-thermal IMT: cobaltate expands 0.3% under voltage","Voltage-driven IMT in layered cobaltate, no heat needed","Electro-proliferation of high-spin Co3+ drives cobaltate IMT"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000454,"raw_usage":{"total_tokens":2242,"prompt_tokens":868,"completion_tokens":1374,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":484,"completion_tokens_details":{"reasoning_tokens":1279}},"tokens_in":484,"tokens_out":1374,"duration_ms":8264,"temperature":1.0,"reasoning_tokens":1279,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:23:08.090271+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Install a micro-thermometer on the sample (or use Raman or infrared thermometry) while measuring current, voltage, and strain: if the local temperature rises high enough to depopulate the low-spin state at the switching point, or if a Joule-heating simulation of the three contact geometries reproduces the observed switching and strain, then the non-thermal claim is contradicted.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the benchmark isostructural metal-insulator transition in VO2 that the paper's non-thermal transition is contrasted with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports a current-induced insulator-metal transition in Ca2RuO4 attributed to structure rather than Joule heating, the closest precedent for structure-coupled electric switching."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior characterization of La2-xSrxCoO4 establishing the mixed low-spin and high-spin Co3+ states at room temperature that the electric field acts on."},{"cited_title":"High-spin and low-spin mixed state in LaSrCoO4: An ab initio study","cited_arxiv_id":null,"evidence_quote":"Ab initio study giving the ionic radii of Co3+ high-spin (0.61 Å) and low-spin (0.545 Å) states, the quantitative basis for the predicted roughly 6% volume difference."},{"cited_title":"W., Granstrom, E","cited_arxiv_id":null,"evidence_quote":"Supplies the conducting-tip force microscopy technique with about 10 nm resolution used to image the growth of metallic regions."},{"cited_title":"A., Sofin, R","cited_arxiv_id":null,"evidence_quote":"Documents Joule heating as the origin of resistive switching in magnetite, the alternative explanation the paper must rule out."}],"review_version":1}