{"id":"61e7e1fc-840d-4a5d-a647-e19612d17267","arxiv_id":"2607.22261","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A layered hybrid perovskite ferroelectric produces a 238 V/cm bulk photovoltaic field along its non-polar vertical direction, about 27 times larger than along the polar in-plane direction.","lead":"A layered hybrid perovskite ferroelectric shows a bulk photovoltaic field of 238 V/cm along its non-polar vertical direction, about 27 times larger than along its polar in-plane direction. If the measurement holds up, this makes the material a strong candidate for voltage-readout photodetectors and reframes where to look for large photovoltaic responses in 2D ferroelectrics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Vertical 5.4 V response lacks a polarization-coupling test: the out-of-plane BPE is never shown to be ferroelectric, leaving electrode/thermal artifacts as an alternative.","rationale":"The paper's most striking result is the 5.4 V out-of-plane photovoltage along a direction with no net polarization. For the central claim to hold, this voltage must be a genuine bulk photovoltaic response, not an electrode, photo-Dember, pyroelectric, or thermal artifact. The in-plane polar direction has switchable photocurrent, but the vertical direction is never shown to couple to the ferroelectric order. The φ- and θ-dependent measurements are consistent with Eq. 2, yet similar polarization/incidence-angle dependences can arise from surface photogalvanic or photo-Dember effects, so they are not decisive. The conversion of the measured voltage into a homogeneous 238 V/cm field also relies on a thickness-averaged conductivity (Eqs. 8–12); for a sample with absorption depth ~94 nm and a much thicker dark bulk, the local open-circuit field is E(x)=J_BPE(x)/σ(x), so the dark region should not contribute a field. The series-resistance integral, not the arithmetic average, is needed. Therefore the verdict should remain CONDITIONAL: the observation is interesting but the central claim requires a polarization-coupling test and a corrected transport model before acceptance. The reader's weakest assumption already identified this unpolarized vertical-response issue, so I agree with that assessment.","tokens_in":11036,"tokens_out":14434,"duration_ms":156025,"concrete_test":"Pole the vertical-device crystal along the in-plane polar axis in both directions and measure the out-of-plane photocurrent/photovoltage at θ=±30° under 405 nm; if the vertical response does not reverse or change with in-plane polarization, the ferroelectric-BPE attribution is unsupported. As a control, repeat the same check on a non-ferroelectric structural analogue to rule out electrode artifacts.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Central claim: EBPE(out-of-plane)=238 V/cm is BPE and exceeds the polar-direction response by two orders of magnitude. The only support for the 5.4 V vertical photovoltage being BPE is the φ/θ dependence and its above-bandgap magnitude (Fig 3a,b). The defining BPE fingerprint—switchable photocurrent after ferroelectric poling—is shown only for the in-plane polar axis (Fig 2c). For the vertical geometry no poling experiment is reported, not even an indirect one that reverses the in-plane polarization and tests the off-diagonal tensor component β_aac in Eq. 2. Thus Schottky, photo-Dember, pyroelectric, and thermoelectric signals are not excluded. The extraction of an 'EBPE field' also depends on the thickness-averaged conductivity in Eqs. 8–12; for an exponentially nonuniformly illuminated slab the correct series resistance is ∫dx/σ(x), not d/[(1/d)∫σ dx]. With the local balance E(x)=J_BPE(x)/σ(x), dark regions with J_BPE=0 contribute no field, so the claimed V∝d scaling and the homogeneous 238 V/cm average are not physically established. If the 5.4 V is not BPE, the anisotropic-BPE claim and benchmark comparisons collapse.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports bulk photovoltaic effect (BPE) measurements on EA4Pb3Br10 (EPB), a 2D hybrid perovskite ferroelectric with in-plane spontaneous polarization. It claims that the out-of-plane nonpolar direction exhibits a bulk photovoltaic field EBPE = 238 V/cm, stated to be two orders of magnitude higher than the in-plane polar direction (8.8 V/cm) and higher than benchmark photoferroelectrics. The authors attribute this emergent anisotropy to the combination of high in-plane photoelectric efficiency, low out-of-plane dark conductivity, and strong optical absorption, and they propose formulas describing an intensity-dependent and thickness-dependent photovoltage. The paper includes polarization-angle and incidence-angle dependence, switchable photocurrent for the in-plane polarization, and photoconductivity anisotropy measurements.","tokens_in":11411,"tokens_out":5431,"duration_ms":55131,"significance":"If the vertical 5.4 V response is truly BPE, this is a significant result: it would demonstrate that a nonpolar direction can host a larger bulk photovoltaic field than the polar direction in a 2D hybrid perovskite ferroelectric, with potential for voltage-readout photodetection. The strength of the paper is the combination of in-plane switching, polarization-angle and incidence-angle dependence, and the very high βL ~39 µA/mW extracted for the in-plane direction. However, the quantitative claim is overstated, the model uses an incorrect series-resistance treatment, and the vertical response lacks the key ferroelectric switching test. These issues are load-bearing and require additional experiments and model revision before the claims can be accepted.","major_comments":[{"comment":"The central claim that the out-of-plane EBPE (238 V/cm) is 'two orders of magnitude higher' than the in-plane value (8.8 V/cm) is contradicted by the quoted numbers: 238/8.8 ≈ 27, i.e., a factor of about 27, not 100. The same factor applies to the photovoltages (5.4 V vs 0.2 V). Please correct the abstract and main text to 'more than an order of magnitude', or provide measured data that actually show a factor ≥100 in some comparison.","section":"Abstract and §3 (vertical BPE characterization)"},{"comment":"The model uses V_BPE = J_BPE / σ with a thickness-averaged photoconductivity σ_phe-eff = (1/d)∫σ dx. For a nonuniformly illuminated slab, the correct open-circuit voltage is V = ∫₀ᵈ J_shift(x)/σ(x) dx (equivalently, the series resistance is ∫ dx/σ(x), not d/[(1/d)∫σ dx]). Because the light penetration depth is ~94 nm and J_shift is confined near the illuminated surface, the local field is not homogeneous. The extraction of a uniform 'EBPE = 238 V/cm' from V/d, and the V(d) and V(I0) scaling laws derived from Eqs. (11)–(12), are therefore not physically established. Please revise the model using the local balance J(x) = J_shift(x) + σ(x)E(x) and integrate E(x) over the thickness.","section":"Eqs. (8)–(12)"},{"comment":"The vertical photovoltage is not demonstrated to be of BPE origin. The φ- and θ-dependence is consistent with Eq. (2), but no ferroelectric switching or poling experiment is reported for the vertical geometry. The statement that the above-bandgap photovoltage 'further confirms its BPE origin' is insufficient, because photovoltages of this magnitude can also arise from photo-Dember, thermoelectric, or electrode-related effects. A polarization-reversal test that flips the in-plane polarization and shows a corresponding reversal of the out-of-plane photocurrent/photovoltage is needed to exclude these artifacts.","section":"Fig. 3 and §3 (vertical BPE origin)"},{"comment":"Equation (13), V_BPE = (β_L / σ_dark) I0, is presented as explaining the emergent linear V(I0) behavior, but β_L and σ_dark are measured on the same material; this is a consistency check rather than a parameter-free prediction. Moreover, the β_L used is extracted from the in-plane measurement, while the out-of-plane BPE coefficient (β_aac in Eq. (2)) is not reported. The V(d) fit in Fig. 3e uses four fitted constants (a3, b3, c1, d) and therefore does not independently test the model. Please provide the measured out-of-plane β_aac and, if possible, a parameter-free comparison with Eq. (13) and the V(d) data.","section":"Eq. (13) and Fig. 3e"}],"minor_comments":[{"comment":"The notation for the BPE current is inconsistent: both J_BPE and JBPE appear. Please unify.","section":"Entire manuscript"},{"comment":"The comparison of EBPE with benchmark materials in Fig. 3f should specify the sample thickness used for each entry, because the nonlinear V(d) relation makes EBPE thickness-dependent in this material.","section":"§3 and Fig. 3e"},{"comment":"The equation contains an awkward typesetting with '×1/d d'; please clarify the intended fractional form.","section":"Eq. (12)"},{"comment":"Please check for typos such as 'electric conductivity' (should be 'electrical conductivity') and ensure the acronym for EA4Pb3Br10 is consistently 'EPB'.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper has promising data and a plausible physical idea, but the central quantitative claim ('two orders of magnitude') is overstated, and the vertical BPE identification is incomplete because no poling/switching test is reported. The modeling also needs to be corrected from a thickness-averaged conductivity to a proper series-resistance treatment. These are fixable with additional experiments and revision within the scope of the paper, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper reports a 5.4 V photovoltage along the non-polar vertical direction of EA4Pb3Br10, corresponding to 238 V/cm — about 27 times the in-plane polar value, not “two orders of magnitude” as stated. That anisotropy, plus the intensity-dependent photovoltage and nonlinear thickness scaling, is genuinely new for 2D hybrid perovskite ferroelectrics. The comparison to benchmark photoferroelectrics is also useful.\n\nWhat the paper does well: the polarization-angle and incidence-angle dependence of the vertical photocurrent matches the proposed β_aac tensor component; the thickness series shows a systematic trend; and the qualitative picture — strong absorption, low out-of-plane dark conductivity, and a series resistance dominated by unilluminated bulk — is physically plausible. The in-plane BPE is properly demonstrated with switchable photocurrent after poling.\n\nThe soft spots are not minor. The vertical signal is never shown to be ferroelectric: no poling or switching experiment is reported for the out-of-plane geometry, so Schottky, photo-Dember, pyroelectric, or thermoelectric contributions are not excluded. The symmetry dependence is suggestive but not a fingerprint. Second, the model uses a thickness-averaged photoconductivity, but for Beer–Lambert illumination the correct series resistance is ∫dx/σ(x), not d/((1/d)∫σ dx). That error propagates into the V∝d explanation and the interpretation of 238 V/cm as a homogeneous field. Third, the formulas in Eqs. 11–13 are fitted to the same thickness and intensity data they claim to explain, making them descriptive rather than predictive. Fourth, error bars are absent, and the SI link is not accessible, which hampers verification.\n\nThe central observation may well be real — 5.4 V from a 265 µm crystal is large and the angular dependencies are consistent with BPE — but the load-bearing claims about anisotropy and mechanism require better evidence. The paper deserves peer review, because if the vertical BPE holds up it is a meaningful advance. I would send it to referees with a request for vertical poling data, raw I-V curves with error bars, and a corrected derivation of the series resistance. Without those, the “two orders” claim and the model should be taken as tantalizing but unproven.","headline":"A large vertical photovoltage in a 2D hybrid perovskite ferroelectric that is worth serious attention, but the BPE attribution and the series-resistance model need to be tightened before the bold claims hold.","tokens_in":11937,"tokens_out":2721,"would_cite":false,"duration_ms":28023,"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":"A 2D hybrid perovskite ferroelectric shows a bulk photovoltaic field of 238 V/cm along its nonpolar vertical direction—two orders of magnitude above the polar in-plane direction—because absorbing inorganic layers sit atop an insulating orga","keywords":["bulk photovoltaic effect","2D hybrid perovskite ferroelectrics","shift current","ferroelectric photovoltaics","photovoltage anisotropy","EA4Pb3Br10","nonpolar photoconductivity","photodetectors"],"falsifier":"Replace the transparent top electrode with a different metal and illuminate from both top and bottom: a true shift-current BPE signal should persist and reverse with incidence angle (sin2θ), while an electrode or carrier-diffusion artifact would change magnitude or sign with contact metal and illumination side. Alternatively, measure the vertical photoconductivity profile directly on a cleaved cross-section and compute V = J_BPE ∫ dx/σ(x); if this does not reproduce the 5.4 V open-circuit voltage, the thickness-averaged series-resistance formula is the weak link.","tokens_in":10909,"feed_emoji":"⚡","tokens_out":6919,"duration_ms":65410,"temperature":0.7,"pith_summary":"The paper reports that in the layered 2D hybrid perovskite ferroelectric EA4Pb3Br10 (EPB), the bulk photovoltaic field along the nonpolar vertical direction is 238 V/cm—about two orders of magnitude larger than the 8.8 V/cm measured along the polar in-plane direction. This goes against the usual expectation that the polar axis hosts the strongest bulk photovoltaic response. The authors explain the effect through composition anisotropy: the inorganic lead-bromide layers efficiently generate photocarriers via the shift-current mechanism, while the organic layers are insulator-like, so photoconductivity exists only in a thin illuminated surface region connected in series with a dark, poorly conducting bulk. They derive formulas showing that this series configuration makes the vertical photovoltage scale linearly with laser intensity and nonlinearly with crystal thickness, and they argue the same framework applies to other 2D hybrid perovskite ferroelectrics.","feed_headline":"Nonpolar axis yields 238 V/cm bulk photovoltaic field","feed_subtitle":"In a layered hybrid perovskite, the vertical nonpolar direction beats the polar in-plane direction by two orders of magnitude.","key_machinery":"The central object is EA4Pb3Br10 (EPB), a layered Ruddlesden–Popper-type hybrid perovskite ferroelectric whose polarization lies in-plane while the vertical direction is nonpolar. The carrying mechanism is the shift-current bulk photovoltaic effect: oblique illumination drives a BPE photocurrent along the nonpolar vertical direction through the tensor element β_aac, which follows sin2θ and zero-offset behaviour. The governing identity is E_BPE = J_BPE/σ, with σ determined by the series combination of a thin photoconductive surface layer (absorption coefficient 1.06×10^5 cm^-1, penetration depth ~94 nm) and a dark, insulating organic-dominated bulk (σ_dark ≈ 10^-10 S/m). The paper's formulas","core_discovery":"On the authors' terms, the central discovery is that a 2D hybrid perovskite ferroelectric with in-plane polarization can produce its largest bulk photovoltaic field along the nonpolar out-of-plane direction, with E_BPE = 238 V/cm versus 8.8 V/cm along the polar direction. The field arises from a shift-current BPE photocurrent along the vertical direction—allowed by the tensor element β_aac under oblique illumination—combined with an extremely low out-of-plane dark conductivity. Because the optical penetration depth is only about 94 nm, the illuminated layer is a good photoconductor while the rest of the crystal acts as a dark resistor in series; the resulting voltage is then J_BPE/σ_dark, wh","pith_inferences":["The model implies a testable design rule: reducing the dark conductivity along the vertical direction (for example by lengthening the organic spacer) or increasing the absorption coefficient should raise E_BPE further; the paper does not itself propose this optimization.","Because the crystal is never poled along the vertical axis, the large vertical field is not switchable by an external vertical field; if vertical readout is desired, the in-plane polarization would have to be the memory state and the vertical photovoltage the readout channel, a coupling the paper notes but does not explore.","The near-linear VBPE–I0 dependence suggests that at sufficiently high intensity, photoconductivity should overtake dark conductivity and the curve should bend toward saturation; measuring the crossover intensity would provide a quantitative check of Eq. 13.","A similar series-resistance argument could apply to ferroelectric heterostructures or domain-engineered films with a thin absorbing layer on an insulating substrate, potentially transferring the effect beyond hybrid perovskites."],"forward_implications":["The vertical bulk photovoltaic field of 238 V/cm is two orders of magnitude larger than the polar-direction field in the same crystal and the largest reported among 2D hybrid perovskite ferroelectrics.","Along the nonpolar vertical direction, the photovoltage rises almost linearly with laser intensity, unlike the intensity-independent photovoltage of conventional photoferroelectrics, because dark conductivity dominates the series resistance.","The photovoltage scales nonlinearly with crystal thickness instead of linearly, matching the paper's Eq. 12, and this thickness dependence can be used to tune device output.","The model predicts that any polar system with a strongly absorbing thin layer on an insulating bulk can exhibit a similar nonpolar EBPE enhancement, so the design principle extends beyond this specific compound.","The high in-plane BPE coefficient (βL ≈ 39 µA/mW) shows the inorganic framework is an efficient shift-current generator, making these 2D hybrids promising for voltage-readout photodetectors."],"fun_headline_variants":["27x stronger PV field along nonpolar axis in 2D perovskite","Vertical nonpolar direction boosts photovoltaic field 27-fold","Nonpolar out-of-plane axis beats polar in perovskite PV","2D perovskite: nonpolar direction gives 27x PV field"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the 5.4 V vertical signal is a genuine bulk photovoltaic response obeying V = J_BPE/σ with σ taken as the thickness-averaged conductivity; if the signal is an electrode, carrier-diffusion, or thermal artifact, or if the series resistance of the non-uniformly illuminated crystal should instead be computed from ∫dx/σ(x), the 238 V/cm value does not follow.","fun_headline_variants_meta":{"raw":{"variants":["27x stronger PV field along nonpolar axis in 2D perovskite","Vertical nonpolar direction boosts photovoltaic field 27-fold","Nonpolar out-of-plane axis beats polar in perovskite PV","2D perovskite: nonpolar direction gives 27x PV field"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000571,"raw_usage":{"total_tokens":2578,"prompt_tokens":829,"completion_tokens":1749,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":573,"completion_tokens_details":{"reasoning_tokens":1678}},"tokens_in":573,"tokens_out":1749,"duration_ms":12300,"temperature":1.0,"reasoning_tokens":1678,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T05:17:58.067645+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Replace the transparent top electrode with a different metal and illuminate from both top and bottom: a true shift-current BPE signal should persist and reverse with incidence angle (sin2θ), while an electrode or carrier-diffusion artifact would change magnitude or sign with contact metal and illumination side. Alternatively, measure the vertical photoconductivity profile directly on a cleaved cross-section and compute V = J_BPE ∫ dx/σ(x); if this does not reproduce the 5.4 V open-circuit voltage, the thickness-averaged series-resistance formula is the weak link.","supporting_citations":[],"review_version":1}