{"id":"43b186a7-40c2-4b04-8aab-ea6a274f87bb","arxiv_id":"2507.22217","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Fill factor in low-voltage-loss organic solar cells is limited by field-assisted exciton-to-charge-transfer dissociation, so longer exciton lifetimes can raise fill factor without sacrificing open-circuit voltage.","lead":"This paper shows that the fill factor of organic solar cells is capped not only by charge transport and bimolecular recombination, but also by field-dependent geminate recombination at the exciton-to-charge-transfer step. For low-voltage-loss devices, the Stark effect on the donor-acceptor charge-transfer state makes fill factor fall as voltage loss shrinks, pointing to longer exciton lifetime as a design lever.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The mechanism claim requires the Marcus/Stark model to reproduce the measured field-dependent charge generation, but the model parameters are chosen and never fitted to the TDCF-derived beta values, leaving the Ex-to-CT attribution qualitatively asserted rather than quantitatively demonstrated.","rationale":"The reader's weakest assumption correctly identifies that the mechanism attribution (field acting on Ex-to-CT rather than CT dissociation or direct Ex-to-FC separation) is load-bearing. My stress-test agrees with that qualitative concern but sharpens it into a quantitative gap: the paper never demonstrates that the Stark/Marcus model with its chosen parameters can reproduce the measured field-dependence coefficient beta or the bias-dependent PL quenching. The qualitative exclusion of alternatives is reasonable but not airtight, and the absence of any quantitative link between the microscopic parameters and the TDCF-derived beta means the central claim is currently supported by a parameterized empirical equation (Eq. 4) plus a plausible but unvalidated microscopic story. This does not warrant rejection, because the experimental phenomenology (especially the PTO2:Y1 case) is real and the qualitative mechanism is consistent with prior reports. It does strengthen the need for the conditional acceptance already recommended by the reader: the authors should be asked to fit or at least compare the Marcus/Stark model to their measured beta values and PL quenching magnitudes. I therefore leave the verdict unchanged. My agreement is 'partial' because the reader's weakest assumption focuses on the alternative physical step (CT dissociation/direct separation), whereas my concern is the missing quantitative validation of the proposed step; these are related but distinct, and both point to the same remedy: connect the microscopic model to the measured field-dependence data.","tokens_in":34331,"tokens_out":4849,"duration_ms":59312,"concrete_test":"Compute the model-predicted beta from the Marcus/Stark expressions in Supplementary Note 2 using the stated parameters and the actual electric-field range F = (V - VOC)/d for the four blends. With a steady-state exciton branching ratio eta_int(F) = k_Ex-CT(F) / [k_Ex + k_Ex-CT(F)], using the same k_Ex range as Supplementary Note 3 (10^9 to 10^11 s^-1), extract beta_model = (1/eta_int(OC) - 1)^-1 times |d eta_int / dV| at V = VOC and compare with the measured beta values in Supplementary Table 3. Repeat with lambda from 0.25 to 0.6 eV and dCT from 1 to 2 nm to bracket uncertainty.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that field-dependent free charge (FC) generation is primarily caused by the electric-field dependence of the Ex-to-CT transition, described by a Stark-shifted Marcus rate. The qualitative evidence (PPPC showing blend geminate pairs resemble neat-acceptor excitons, bias-independent pristine-film PL, transient absorption timescales) is suggestive but not exclusive; the quantitative chain is incomplete. The empirical field-dependence coefficient beta in Eq. (4) is obtained from TDCF data (Supplementary Table 3), and the JV simulations in Figure 2e use this equation, not the microscopic model. The Marcus/Stark model in Supplementary Note 2 uses assumed parameters (ECT = 1.4 eV, HEx-CT = 0.01 eV, dCT = 1.5 nm, alphaCT = 8.5 x 10^5 A^3, lambda = 0.5 eV) with no uncertainty or fitting, and is used only to generate schematic trends in Figure 4 and Supplementary Figure 12. Crucially, the model is never checked against the measured beta values (0.6, 0.27, 0.05, 0.09 V^-1 for PM6:Y11, PM6:Y1, PM6:IEICO-4F, PTO2:Y1) or the magnitude of the bias-dependent PL quenching. If the Stark shift predicted from these parameters produces a beta orders of magnitude smaller or larger than measured, the field-dependent step is not the Ex-to-CT transition, and the design recommendation (suppress non-radiative exciton decay, tune CT dipole/polarizability) would be misdirected. Supplementary Note 1 itself states the power-series treatment is used because it is 'unclear which process the electric field dependence comes from' and that a more advanced model will come later, underscoring that the microscopic attribution is not yet quantitatively established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates the fill-factor (FF) limit of organic solar cells (OSCs) by combining device characterization, spectroscopy, and modelling. It compiles a large dataset of OSCs spanning FF values 0.27–0.80 and voltage losses 0.5–1.1 eV, identifies a left boundary in the FF–voltage-loss plane for low-voltage-loss devices, and selects four blends (PM6:Y11, PM6:Y1, PM6:IEICO-4F, PTO2:Y1) for detailed study. Using TDCF measurements, bias-dependent photoluminescence, and pump-push photocurrent, the authors argue that field-dependent free charge generation—attributed specifically to the electric-field-dependent transition between exciton (Ex) and charge-transfer (CT) states—is a primary factor limiting FF beyond conventional transport/recombination limits. An empirical linearization (Eq. (4)) with parameters fitted from TDCF data reproduces the measured JV curves in Fig. 2e, and a Marcus/Stark model in Supplementary Note 2 is used to illustrate how the Ex-to-CT rate depends on field, motivating the design recommendation that suppressing non-radiative exciton decay and tuning CT dipole/polarizability can improve FF and VOC simultaneously.","tokens_in":34679,"tokens_out":4527,"duration_ms":52923,"significance":"If the mechanism claim is correct, the paper would provide actionable design rules for low-voltage-loss OSCs: increasing exciton lifetime and engineering the CT-state dipole/polarizability could raise FF without sacrificing VOC. The curated dataset of many published OSCs with voltage losses and FFs is a valuable community resource, and the PTO2:Y1 case (reconstructed FF 0.58 from transport/recombination parameters vs. measured 0.27) convincingly demonstrates that conventional non-geminate recombination metrics alone cannot explain the low FF. The paper also credits the use of an open-source drift-diffusion code (DriftFusionOPV), aiding reproducibility of the simulated trends. However, the central attribution of the field-dependent step to the Ex-to-CT transition is not yet quantitatively supported: the empirical beta values from TDCF are used in the JV simulation, while the microscopic Marcus/Stark model is only compared schematically against the measured FF and PL data. As a result, the paper currently establishes a phenomenological link between FF and field-dependent generation, but the specific mechanistic conclusion is more tentative than the text claims.","major_comments":[{"comment":"The parameters eta_int and beta in Eq. (4) are extracted from the same TDCF data whose fitted lines are shown in Fig. 2a–d, and those fitted parameters are then used to simulate the JV curves in Fig. 2e. The agreement is therefore a two-parameter consistency check rather than an independent reproduction of the measured curves. The claim that the simulation 'reproduces' the JV curves overstates the evidential weight. To make the model testable, the authors should validate Eq. (4) on data not used in the fit (e.g., intensity- or temperature-dependent JV curves) or explicitly reframe Fig. 2e as an interpolation. This is load-bearing for the paper's assertion that Eq. (4) 'captures the key physical parameters that determine the FF'.","section":"Eq. (4), Supplementary Table 3, Fig. 2e"},{"comment":"The Marcus/Stark model is never quantitatively compared with the measured field-dependence coefficients beta in Supplementary Table 3 (0.6, 0.27, 0.05, 0.09 V^-1). The parameters listed after Eq. (13) — ECT = 1.4 eV, HEx-CT = 0.01 eV, dCT = 1.5 nm, alphaCT = 8.5e5 A^3, lambda = 0.5 eV — are assumed literature values with no uncertainty, and the model output appears only as schematic trends in Fig. 4a,b and Supplementary Fig. 12. As written, the attribution of the field-dependent step to the Ex-to-CT transition is asserted rather than demonstrated. The authors should either fit or constrain the model to reproduce the measured beta values and the magnitude of the bias-dependent PL quenching, or explicitly state that the microscopic model is illustrative and soften the abstract and conclusory claims.","section":"Supplementary Note 2, Fig. 4, Supplementary Table 3"},{"comment":"The exclusion of field-dependent CT dissociation (Process 5) and direct Ex-to-FC separation rests on indirect evidence: PPPC kinetics at short circuit, bias-independent PL of pristine films (Supplementary Fig. 11g,j), and cited prior work. These measurements do not directly constrain the field dependence of CT dissociation in the voltage range near the maximum-power point, which is where FF is set. Moreover, the paper itself acknowledges in Supplementary Note 1 that the power-series form is used because 'it is unclear which process the electric field dependence comes from'; this admission undercuts the subsequent exclusivity of the Ex-to-CT mechanism. A direct test would be to model the measured JG(V) and bias-dependent PL with a CT-dissociation (e.g., Braun/Onsager) term and show that it cannot reproduce the data, or to vary the CT binding energy experimentally and compare the resulting beta.","section":"§Fill-factor limit and geminate pairs; Supplementary Note 1"}],"minor_comments":[{"comment":"The caption states that PTO2:Y1 'shows a lower reconstructed FF compared with the measured FF', but the table reports a reconstructed FF of 0.58 against a measured FF of 0.27; the comparison direction is reversed.","section":"Table 1 caption"},{"comment":"There is a typo in the sentence 'discussed in zthe previous section'; it should read 'discussed in the previous section'.","section":"Supplementary Note 1"},{"comment":"The text describes the four NFA examples as dots in Fig. 2f, while the caption refers to solid-coloured circles; please align the wording.","section":"Fig. 2f"},{"comment":"The green 'left boundary' curve is described as a guide-to-the-eye; it would be helpful to state explicitly that it is not a fitted model curve, to avoid implying a quantitative boundary.","section":"Fig. 1b"},{"comment":"The paper would benefit from defining the 'left border' region of Fig. 1b with a quantitative criterion (e.g., a threshold in voltage loss and FF), since the subsequent selection of the four blends relies on this grouping.","section":"General presentation"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's central claim is a strong interpretation of indirect evidence, and the quantitative gap between the empirical beta values and the microscopic Marcus/Stark model is the key issue to resolve. The editor may also wish to consider whether the novelty relative to recent work by the same groups (Pranav et al., Energy Environ. Sci. 2024; Classen et al., Nat. Energy 2020) is clearly delineated in the introduction and discussion; the current text largely builds on these studies without sharp differentiation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this paper deserves a serious referee, but the load-bearing microscopic claim—field-dependent Ex-to-CT transition as the primary FF limiter—is asserted more strongly than the current evidence justifies. If it holds, it is an important result: it explains the FF-VOC trade-off on the left border and gives a concrete design rule (long exciton lifetime, tuned CT polarizability).\n\nWhat is new and good: the dataset is genuinely useful. The TDCF/reconstructed-FF comparison for PTO2:Y1 is the cleanest part: transport and bimolecular recombination alone predict FF≈0.58 while the measured value is 0.27, so something else is limiting. The PPPC and bias-dependent PL data point toward acceptor excitons rather than CT dissociation, and the Stark-shifted Marcus rate is a plausible physical picture. The historical trend shown in Fig. 4c—Y6-era NFAs combine higher FF and lower voltage loss with longer exciton lifetimes—is a sensible observation.\n\nSoft spots: Eq. (4) is a parameterization, not a microscopic derivation. eta_int and beta are extracted from the same TDCF data and then used in the JV simulation, so Fig. 2e is a reproduction, not an independent prediction. The Marcus/Stark model in Supplementary Note 2 uses assumed parameters (ECT, H_ExCT, dCT, alpha_CT, lambda) and is never fitted to the measured beta values (0.6, 0.27, 0.05, 0.09 V^-1), so the central attribution to the Ex-to-CT step is not quantitatively tested. The paper itself acknowledges in Supplementary Note 1 that it is \"unclear which process the electric field dependence comes from\" and says a more advanced model will come later—honest, but that should temper the conclusion. Table 1's caption says PTO2:Y1 has a lower reconstructed FF than measured, but the table shows the opposite; that looks like a simple caption error. There are also small presentation issues such as \"zthe\" and inconsistent reference formatting.\n\nBottom line: the qualitative mechanism is reasonable and the empirical story is strong enough for a conditional accept, not a desk reject. But before the \"primary mechanism\" claim becomes convincing, the model needs to be validated against the measured field-dependence coefficients, or the chosen parameters need uncertainties and a sensitivity analysis. Device physicists and materials scientists working on OSCs and loss analysis will get value from this paper. Send it out, and ask the authors to fix the caption, add error bars to TDCF-derived quantities, and either fit the Marcus parameters to beta or soften the central claim.","headline":"Worth a serious referee: the paper identifies a real gap in FF analysis and has strong TDCF evidence for one system, but the central Ex-to-CT mechanism is asserted more strongly than the model can currently support.","tokens_in":35412,"tokens_out":2066,"would_cite":true,"duration_ms":26431,"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 field-dependent conversion of excitons into charge-transfer states, controlled by the Stark effect on the charge-transfer-state energy, is the primary factor limiting the fill factor of low-voltage-loss organic solar…","keywords":["organic solar cells","fill factor","free charge generation","geminate recombination","charge-transfer state","Stark effect","Marcus theory","exciton lifetime"],"falsifier":"Measure the bias-dependent CT-state population directly in a low-offset blend such as PTO2:Y1 with transient absorption or time-resolved infrared spectroscopy. If the CT population grows faster under reverse bias while the exciton decay rate extracted from photoluminescence is unchanged, then field-dependent CT dissociation is doing the work, and the Ex-to-CT Stark/Marcus picture would not be the limiting mechanism.","tokens_in":34041,"feed_emoji":"☀️","tokens_out":9601,"duration_ms":93707,"temperature":0.7,"pith_summary":"This paper tries to establish why organic solar cells with very low voltage losses often have poor fill factors, and what sets the fill-factor limit. Using devices spanning fill factors from 0.27 to 0.80 and voltage losses from about 0.5 to 1.1 eV, the authors show that transport and bimolecular recombination cannot explain the sharp fill-factor drop at low voltage loss; instead, geminate recombination through a field-dependent step from exciton to charge-transfer states is the culprit. They build an analytical model in which the electric field shifts the charge-transfer-state energy through the Stark effect and changes the Marcus rate for exciton-to-charge-transfer conversion, reproducing measured current-voltage curves. If the paper is right, the route to higher efficiency is to suppress non-radiative exciton decay and increase exciton lifetime, which improves fill factor and open-circuit voltage together.","feed_headline":"Exciton-to-charge-transfer step sets organic solar-cell fill factor","feed_subtitle":"A Stark-shifted exciton-to-CT rate, not transport, sets the FF of low-loss OSCs; longer exciton lifetime may fix it.","key_machinery":"The central object is the donor-acceptor charge-transfer (CT) state. Its energy shifts under the applied field through the first- and second-order Stark effect, $\\Delta E_{\\mathrm{CT}} = -\\boldsymbol{\\mu}_{\\mathrm{CT}} F - \\frac{1}{2} \\alpha_{\\mathrm{CT}} F^2$, and this shift enters the Marcus rates $k_{\\mathrm{Ex}\\to\\mathrm{CT}}$ and $k_{\\mathrm{CT}\\to\\mathrm{Ex}}$ for interconversion between the exciton (Ex) and CT states. The field-induced increase of $k_{\\mathrm{Ex}\\to\\mathrm{CT}}$ and decrease of $k_{\\mathrm{CT}\\to\\mathrm{Ex}}$ make free-charge generation bias-dependent, parameterised in the device model by $\\beta$ in $J_G(V) = J_{\\mathrm{Abs}}\\eta_{\\mathrm{int}}[1 + (1/\\eta_{\\mathrm{int}} - 1)\\beta|V - V_{\\mathrm{OC}}|]$. The supporting experiments, time-delayed collection field, pump-push photocurrent, and bias-dependent photoluminescence, locate the field dependence at the Ex-to-CT step rather than at CT dissociation.","core_discovery":"The paper's central claim is that field-dependent free charge generation, rather than transport or bimolecular recombination alone, is the primary mechanism limiting fill factor in low-voltage-loss organic solar cells, and that the specific field-sensitive step is the transition from the exciton (Ex) to the charge-transfer (CT) state. This follows from a survey of devices spanning fill factors 0.27 to 0.80 and voltage losses 0.5 to 1.1 eV, where the low-voltage-loss boundary shows fill factor collapsing despite good mobilities and a reconstructed fill factor far above the measured one. The authors argue that the electric field changes the Ex-to-CT transition rate through the Stark effect on the CT-state energy, and they reproduce measured current-voltage curves with an analytical model using a field-dependence coefficient $\\beta$. They conclude that suppressing non-radiative exciton decay and increasing exciton lifetime can raise fill factor without sacrificing open-circuit voltage.","pith_inferences":["A testable extension is to use electroabsorption spectroscopy on blends as a predictive screen: materials with larger CT-state polarizability should show larger $\\beta$ and lower fill factor at a given voltage loss.","The mechanism should also affect organic photodiodes and other biased excitonic devices, where field-assisted exciton-to-CT conversion would change responsivity near the operating bias.","Reducing reorganization energy or improving exciton diffusion should act like a longer exciton lifetime in softening the fill-factor-voltage-loss trade-off; the paper mentions these as extensions but does not quantify them.","If the Ex-to-CT step is the bottleneck, then published blends with identical voltage loss but different exciton lifetimes should cluster along different fill-factor frontiers, which could be checked without new measurements."],"forward_implications":["In low-voltage-loss blends, a low internal generation efficiency $\\eta_{\\mathrm{int}}$ combined with a nonzero $\\beta$ directly lowers fill factor, so geminate recombination joins transport as a first-order fill-factor parameter.","Increasing the exciton lifetime shifts the simulated fill-factor-versus-voltage-loss curve so that high fill factor persists at smaller energetic offsets.","Design strategies that raise the radiative recombination rate to boost photoluminescence quantum yield are counterproductive if they shorten the exciton lifetime, because the fill-factor penalty cancels the open-circuit-voltage gain.","The Stark-effect model identifies the CT-state dipole moment and polarizability as tunable molecular parameters; in bulk heterojunctions with randomly oriented dipoles, the polarizability term dominates.","The same framework explains historical fill-factor improvements: Y6-family acceptors with slower exciton decay outperform earlier ITIC-based materials at low voltage loss."],"supporting_citations":[{"why":"Supplies the analytical ideal-diode fill-factor expression used to define the upper fill-factor boundary.","marker":"21"},{"why":"Supplies the electronic quality Q linking mobility and bimolecular recombination to the fill-factor upper limit.","marker":"27"},{"why":"Supplies the figure of merit alpha and the current-voltage relation used to separate transport-limited fill factor from field-dependent charge generation.","marker":"29"},{"why":"Braun's model provides the linear field-dependence expansion from which equation (4) for J_G(V) is derived.","marker":"56"},{"why":"Establishes the role of exciton lifetime in charge generation at negligible energy offsets, the starting point for the lifetime design rule.","marker":"11"},{"why":"Provides the pump-push photocurrent technique used to identify geminate bound states as acceptor excitons.","marker":"33"},{"why":"Prior demonstration of field-assisted exciton dissociation in a low-offset PM6:Y5 blend, supporting assignment of the field dependence to exciton dissociation.","marker":"36"},{"why":"Prior work on the competition between singlet exciton decay and free charge generation, supporting the Ex-to-CT transition as the field-dependent step.","marker":"37"},{"why":"Provides dipole and polarizability values used to parameterize the Stark shift of CT states in the Marcus model.","marker":"159"}],"fun_headline_variants":["Stark effect on exciton transfer limits organic solar fill factor","Field-sensitive exciton step caps organic solar cell performance","Longer exciton lifetime may unlock organic solar fill factor","Fill factor bottleneck: exciton-to-CT transition under electric field","Exciton lifetime key to beating organic solar fill factor limit"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the electric field mainly changes the rate of the exciton-to-charge-transfer transition, and not the dissociation of charge-transfer states or a direct field-assisted splitting of excitons into free charges.","fun_headline_variants_meta":{"raw":{"variants":["Stark effect on exciton transfer limits organic solar fill factor","Field-sensitive exciton step caps organic solar cell performance","Longer exciton lifetime may unlock organic solar fill factor","Fill factor bottleneck: exciton-to-CT transition under electric field","Exciton lifetime key to beating organic solar fill factor limit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000236,"raw_usage":{"total_tokens":1490,"prompt_tokens":921,"completion_tokens":569,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":487}},"tokens_in":537,"tokens_out":569,"duration_ms":7134,"temperature":1.0,"reasoning_tokens":487,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T11:56:20.421134+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the bias-dependent CT-state population directly in a low-offset blend such as PTO2:Y1 with transient absorption or time-resolved infrared spectroscopy. If the CT population grows faster under reverse bias while the exciton decay rate extracted from photoluminescence is unchanged, then field-dependent CT dissociation is doing the work, and the Ex-to-CT Stark/Marcus picture would not be the limiting mechanism.","supporting_citations":[{"cited_title":"& Iimori, T","cited_arxiv_id":null,"evidence_quote":"Provides dipole and polarizability values used to parameterize the Stark shift of CT states in the Marcus model."}],"review_version":1}