{"id":"ebb43cee-fdd2-43a7-a582-8f4c369b8174","arxiv_id":"2607.07220","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":4,"one_line_summary":"Electrical microsecond spin-orbit torque pulses flip the phase of broadband terahertz emission from an antenna-coupled spintronic trilayer without external magnetic fields.","lead":"The paper demonstrates that a terahertz light pulse's phase can be flipped electrically in a spintronic device using microsecond current pulses, without external magnetic fields. This enables compact, fast, programmable terahertz sources for spectroscopy and imaging.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"THz-TDS waveforms are multi-pulse averages; 'deterministic' switching claim lacks pulse-resolved verification under laser illumination where USMR hysteresis degrades.","rationale":"The paper demonstrates a real and interesting capability — field-free, all-electrical THz phase inversion via SOT switching in an antenna-coupled trilayer. The experimental evidence (THz waveforms showing phase inversion, USMR switching loops, MOKE-confirmed UMA) is substantial and the approach is distinct from the concurrent Chen et al. work. However, the claim of 'deterministic' switching is not adequately supported given that: (1) the THz waveforms are multi-pulse averages, (2) USMR hysteresis degrades under the actual operating condition (laser on), and (3) no quantitative amplitude comparison or pulse-resolved measurement is provided. The discrepancy between calculated (coherent rotation, j_c^calc = 8.06×10^11) and experimental (thermally assisted nucleation, j_c^exp = 2.53×10^11) switching current densities further suggests thermal activation plays a significant role, which is consistent with stochastic rather than deterministic switching. This does not invalidate the core phenomenon but weakens the 'deterministic' and 'complete' qualifiers in the central claim. A CONDITIONAL verdict is appropriate: the result is promising and likely correct in essence, but the deterministic label requires pulse-resolved verification. The reader's ACCEPT verdict is slightly too generous given this gap, though the concern is about precision of the claim rather than its fundamental validity.","tokens_in":14102,"tokens_out":2649,"duration_ms":172344,"concrete_test":"Perform pulse-resolved THz detection synchronized to individual write pulses (e.g., electro-optic sampling with a gated readout triggered on each write event, or a rep-rate-matched single-shot THz scheme). Measure the THz peak amplitude distribution for ~1000 consecutive write pulses of each polarity. If >95% of pulses fall within ±10% of the expected phase-flipped amplitude, the 'deterministic' claim is supported. If the distribution is bimodal or broad, switching is stochastic. Alternatively, as a simpler check: quantitatively report the peak-to-peak amplitude ratio between the two switched states with error bars from repeated measurements; a ratio deviating from unity by >10% would signal incomplete switching.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim of 'deterministic, complete π-phase inversion' rests on THz time-domain waveforms acquired via delay-stage scanning, which inherently averages over thousands of write-pulse events. Under laser illumination at 100 mW, the USMR switching loops become noisy and poorly defined (Fig. 4d, Fig. 5), which the authors attribute to thermal effects 'rather than incomplete magnetization switching.' However, if switching reliability were, say, 85–95% rather than truly deterministic, the time-averaged THz waveform would still show phase inversion with slightly reduced amplitude — consistent with the 'nearly identical peak amplitude' reported (no quantitative ratio or error bars given). The paper states 'stable and reproducible phase switching is confirmed over an extended time of operation' in the conclusion, but no data explicitly demonstrating this is presented in the results. Without a pulse-resolved or single-shot THz measurement, or at minimum a quantitative amplitude comparison with error bars between the two switched states, one cannot distinguish deterministic switching from high-probability stochastic switching. This matters because the word 'deterministic' is load-bearing for the paper's positioning over stochastic or thermally assisted alternatives. The reader's concern about long-term UMA stability is related but secondary; the more immediate question is whether switching is deterministic during actual operation, given the thermal environment the authors themselves describe.","agreement_with_reader":"partial"},"referee_report":{"model":"glm-5.2","summary":"This manuscript reports electrically controlled, field-free π-phase inversion of broadband THz pulses in an antenna-coupled spintronic emitter. The device uses a Ta/CoFeB/Pt trilayer with uniaxial magnetic anisotropy (UMA) engineered via oblique-angle deposition (OAD) of the Ta underlayer, patterned into an H-dipole antenna. The authors demonstrate that sub-µs current pulses induce spin-orbit torque (SOT) switching of the CoFeB magnetization, producing complete THz phase inversion without any external magnetic field. The antenna simultaneously enhances THz outcoupling (~2× peak amplitude) and serves as the electrical feed for switching. The field-free condition is verified by comparing remanent and saturated THz emission (Fig. 2), and phase inversion is directly observed in the time domain (Fig. 4b–c). The work is timely and addresses a clear gap relative to piezoelectric- or electromagnet-based approaches, which are limited in speed and require external fields.","tokens_in":14266,"tokens_out":1714,"duration_ms":336225,"significance":"The central result — all-electrical, field-free, sub-µs THz phase inversion with preserved amplitude and bandwidth — is a meaningful advance for spintronic THz emitter technology. The OAD-underlayer strategy for robust UMA, combined with the dual-function H-dipole antenna (emission enhancement + electrical switching feed), is a well-motivated device architecture. The experimental evidence is grounded in direct time-domain THz waveforms and USMR switching loops rather than fitted models. The comparison of electrically driven and field-driven phase inversion (Fig. 4d) is a useful validation. The work is positioned clearly relative to the recent Chen et al. study (Ref. 32), distinguishing the OAD-induced anisotropy approach from field-induced anisotropy. The MHz-rate modulation claim is appropriately framed as an instrumentation-limited lower bound rather than an intrinsic speed limit.","major_comments":[{"comment":"§2.2, Fig. 4(b–c): The term 'deterministic' is load-bearing for the paper's positioning, but the evidence presented does not fully distinguish deterministic switching from high-probability stochastic switching. The THz-TDS waveforms are acquired via delay-stage scanning and thus average over many write-pulse events. Under 100 mW laser illumination, the USMR switching loops become noisy and less well-defined (Fig. 4c, Fig. 5), which the authors attribute to thermal effects 'rather than incomplete magnetization switching.' However, if switching reliability were 85–95% rather than truly deterministic, the time-averaged THz waveform would still show phase inversion with slightly reduced amplitude — consistent with the 'nearly identical peak amplitude' reported without quantitative ratio or error bars. The conclusion states 'stable and reproducible phase switching is confirmed over anextended","section":null},{"comment":"§2.2, Fig. 4(c): The claim that the switched THz waveform maintains 'nearly identical peak amplitude and spectral shape' is stated qualitatively without a quantitative comparison (e.g., amplitude ratio, spectral overlap, or error bars). Given that this preservation of amplitude is a key advantage claimed over piezoelectric approaches (where amplitude is attenuated during switching), a quantitative figure — even a simple peak-to-peak amplitude ratio between the two switched states with measurement uncertainty — would substantially strengthen the claim. The power spectra in Fig. 6(c) compare antenna-coupled vs. unpatterned emitters but not the two switched states.","section":null},{"comment":"§2.2: The asymmetry in critical switching currents (+22.84 mA vs. −26.24 mA) is attributed to 'unequal coercivities arising from edge pinning or asymmetric domain nucleation induced during device patterning.' While plausible, this asymmetry could also indicate that the switching mechanism is not purely coherent rotation (type-y) as assumed, but involves some nucleation-mediated contribution. The authors note that j_exp_c is lower than j_cal_c and attribute this to 'thermally assisted, nucleation-dominated magnetization reversal,' which seems to contradict the type-y coherent rotation picture. Clarifying whether the switching proceeds via coherent rotation or nucleation under these device conditions would strengthen the mechanistic understanding.","section":null}],"minor_comments":[{"comment":"§2.1: The anisotropy field is stated as µ0Ha ≈ 60 mT based on L-MOKE, but the MOKE data is said to be 'consistent with our previous work' and the present heterostructure was fabricated under identical conditions. Since the anisotropy is load-bearing for field-free operation, showing the actual L-MOKE hysteresis loops (or at least confirming they were measured on the specific device studied) rather than relying solely on a reference would be preferable.","section":null},{"comment":"Fig. 4 caption: Panels (b) and (c) are referenced in the text as showing THz phase switching and current-induced switching under laser excitation, but the distinction between panels (b) and (c) is not immediately clear from the caption. Clarifying what each panel shows would help.","section":null},{"comment":"§2.2: The effective modulation bandwidth is stated as 1–2.5 MHz, determined by the bias tee operational window (0.025–100 MHz). It would be useful to state the actual pulse widths used for the THz phase switching demonstration (1 µs is mentioned in the abstract, but the text mentions 0.4–1 µs) and whether the phase inversion in Fig. 4(c) was specifically obtained with 1 µs pulses.","section":null},{"comment":"§2.3, Eq. (2): The effective permittivity ε_eff_r is described as 'determined from the substrate permittivity (εr = 12.1),' but the formula for how ε_eff_r is obtained from εr is not given. A brief expression or reference would help.","section":null},{"comment":"Several typographical issues: 'exibited' (§2.2), 'maintainig' (§2.2), 'severly' (§2.2), 'consitent' (§2.1), 'architechure' (§2.2), 'filipping' (Introduction). These should be corrected.","section":null},{"comment":"§2.2: The echo pulse discussion is clear, but the statement 'A weak reflection arising from a minor air gap at the Si lens mounting interface and re-excites the spintronic trilayer' has a grammatical issue (missing subject or verb agreement).","section":null},{"comment":"Fig. 5 caption: The inset is described as showing THz phase switching for 15 and 30 mW, but the main text discussion focuses on 100 mW operation. The relationship between the Fig. 5 data (15, 30 mW) and the main results at 100 mW could be made clearer.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The skeptic's concern about 'deterministic' switching is valid and is the most substantive issue. The THz-TDS averaging means the data cannot distinguish deterministic from high-probability stochastic switching under laser illumination, especially given the degraded USMR hysteresis at 100 mW. However, this does not undermine the core demonstration of phase inversion — it weakens the specific claim of 'deterministic' switching. The authors can address this by either (a) toning down 'deterministic' to 'reliable' or 'high-yield,' or (b) providing quantitative amplitude/error analysis. This is a minor revision, not major, because the central claim of field-free electrical phase inversion is well-supported regardless of whether the switching is 95% or 100% deterministic. The paper is a solid experimental contribution and well-suited for the journal."},"author_rebuttal":null,"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper demonstrates all-electrical, field-free π-phase inversion of broadband THz pulses using sub-µs SOT switching in an antenna-coupled Ta/CoFeB/Pt trilayer. That's a real result — the combination of OAD-induced uniaxial anisotropy (no external field needed) with an H-dipole antenna for simultaneous THz enhancement and electrical switching is new and directly useful for THz spectroscopy instrumentation. The field-free remanent THz emission matching saturated emission (Fig. 2) is clean, and the phase inversion in the time-domain waveforms (Fig. 4b-c) is unambiguous. The ~2x antenna enhancement is also a nice bonus. The authors are honest about the bandwidth being instrumentation-limited (bias tee), not intrinsic. Credit where due: the experimental execution is solid and the claim of field-free phase inversion itself is well-supported by the data. The soft spot is the word 'deterministic.' The stress-test concern lands here. The THz waveforms are delay-stage averages over thousands of write pulses, and under laser illumination at 100 mW the USMR hysteresis degrades noticeably (Fig. 4d, Fig. 5). The authors attribute this to thermal noise rather than incomplete switching, which is plausible, but they don't prove it. If switching reliability were 85-95% rather than truly deterministic, the time-averaged THz waveform would still show phase inversion with slightly reduced amplitude — exactly what 'nearly identical peak amplitude' without error bars would look like. The conclusion claims 'stable and reproducible phase switching over extended operation,' but no data explicitly backs that statement. A pulse-resolved or single-shot THz measurement, or even quantitative amplitude ratios with error bars between switched states, would close this gap. Without it, 'deterministic' is doing more work than the evidence supports. The reader's concern about long-term UMA stability is secondary — the MOKE data and prior work make it reasonable — but the immediate question of whether switching is deterministic during actual operation under laser heating is the more pressing one. This is a solid experimental paper with a legitimate new result and one load-bearing word that isn't fully substantiated. It deserves a serious referee who should push for either pulse-resolved verification or a toning-down of 'deterministic' to 'high-probability.'","headline":"Solid experimental demo of field-free THz phase inversion; the 'deterministic' claim needs pulse-resolved evidence","tokens_in":14892,"tokens_out":563,"would_cite":true,"duration_ms":70199,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.78.-n","78.47.J-","75.70.Tj"],"model":"glm-5.2","headline":"Electrical pulses flip THz waveforms in spintronic emitters","keywords":[],"falsifier":"If the oblique-angle-deposited Ta underlayer's induced anisotropy degrades under repeated current pulsing or cumulative laser heating — such that the remanent magnetization no longer holds stably along the easy axis — the field-free deterministic switching would fail, and the π-phase flip would become incomplete or stochastic.","tokens_in":14357,"feed_emoji":"⚡","tokens_out":875,"duration_ms":240703,"temperature":0.7,"pith_summary":"The paper demonstrates that the phase of a broadband terahertz pulse emitted from a spintronic heterostructure can be deterministically inverted — a complete π-phase flip — by applying sub-microsecond electrical current pulses, without any external magnetic field at any stage of operation. The device is a Ta/CoFeB/Pt trilayer in which the tantalum underlayer is deposited at an oblique angle, creating a built-in magnetic easy axis that holds the magnetization direction stable at remanence. An H-dipole antenna patterned on top serves a dual role: it boosts THz outcoupling efficiency and acts as the electrical feed for spin-orbit torque switching. When a current pulse reverses the magnetization, the inverse spin Hall effect produces a charge current of opposite sign, which flips the emitted THz waveform polarity while preserving its amplitude and spectral shape. The authors show 1-microsecond write pulses yielding effective megahertz-rate phase modulation, roughly three orders of magnitude faster than mechanical or electromagnet-based approaches, and argue the upper bound is set by their bias-tee instrumentation rather than by intrinsic spintronic dynamics.","feed_headline":"Electrical pulses flip THz waveforms in spintronic emitters","feed_subtitle":"Sub-microsecond current bursts invert terahertz pulse phase without any external magnetic field, enabling megahertz-rate modulation.","key_machinery":"The central mechanism is the vector relationship in the inverse spin Hall effect: the emitted THz electric field is perpendicular to the magnetization direction, so reversing the magnetization by 180 degrees reverses the sign of the charge current and thus inverts the THz waveform. The oblique-angle deposition of the Ta underlayer creates a uniaxial magnetic anisotropy of about 60 mT, strong enough to maintain stable remanent magnetization without an external field. The H-dipole antenna enforces the type-y current geometry (current transverse to the easy axis), which enables deterministic spin-orbit torque switching via coherent rotation rather than domain-wall nucleation.","core_discovery":"A trilayer spintronic THz emitter with oblique-angle-deposited Ta underlayer, patterned into an H-dipole antenna, achieves deterministic, field-free, sub-microsecond π-phase inversion of broadband THz pulses via spin-orbit torque switching, while simultaneously doubling the THz emission amplitude relative to the unpatterned film. The antenna geometry satisfies the type-y switching configuration — current transverse to the easy axis — enabling coherent magnetization reversal and ISHE-driven THz polarity flip in a single integrated device.","pith_inferences":[],"forward_implications":["Compact, on-chip THz emitters with electrically programmable phase could enable phase-sensitive THz spectroscopy and near-field imaging systems without mechanical choppers or electromagnets.","The megahertz modulation bandwidth demonstrated here could support high-speed THz wireless communication links where rapid phase encoding of THz waveforms is needed.","The dual-use antenna architecture — simultaneously enhancing THz outcoupling and enabling electrical switching — suggests a scalable design route for integrated THz photonic circuits.","If the bias-tee limitation is removed, the intrinsic switching speed of spin-orbit torque (demonstrated elsewhere at 300 ps) would push THz phase modulation into the gigahertz regime."],"fun_headline_variants":["Field-free spin-orbit torque inverts THz emission phase at megahertz rates","Antenna-integrated spintronic emitter achieves field-free THz phase inversion","Microsecond electrical pulses flip THz phase without external magnetic field","Spin-orbit torque switches THz waveform phase in integrated heterostructure","Field-free spintronic THz emitter enables megahertz-rate phase inversion"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The deterministic, field-free switching depends on the oblique-angle-deposited tantalum underlayer maintaining a sufficiently strong and stable uniaxial magnetic anisotropy (about 60 mT) under repeated current pulsing and laser heating. The paper does not explicitly test anisotropy degradation over extended thermal cycling or long-term device operation.","fun_headline_variants_meta":{"raw":{"variants":["Field-free spin-orbit torque inverts THz emission phase at megahertz rates","Antenna-integrated spintronic emitter achieves field-free THz phase inversion","Microsecond electrical pulses flip THz phase without external magnetic field","Spin-orbit torque switches THz waveform phase in integrated heterostructure","Field-free spintronic THz emitter enables megahertz-rate phase inversion"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":636,"prompt_tokens":538,"completion_tokens":98,"prompt_tokens_details":null},"tokens_in":538,"tokens_out":98,"duration_ms":25314,"temperature":1.0,"reasoning_tokens":null,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-09T16:55:44.100592+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If the oblique-angle-deposited Ta underlayer's induced anisotropy degrades under repeated current pulsing or cumulative laser heating — such that the remanent magnetization no longer holds stably along the easy axis — the field-free deterministic switching would fail, and the π-phase flip would become incomplete or stochastic.","supporting_citations":[],"review_version":1}