{"id":"ad8672c7-2eeb-40f8-a765-99b898446af7","arxiv_id":"2505.05801","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Adding a thin 2D perovskite spacer between a 3D perovskite sensitiser and rubrene reduces singlet back-transfer, improving relative upconversion efficiency at sub-solar powers while lowering overall intensity.","lead":"This paper inserts thin 2D perovskite layers between a 3D perovskite light absorber and a rubrene emitter in solid-state photon upconversion devices. The spacer reduces a parasitic energy-loss process at low light intensity, improving relative upconversion efficiency and device stability over 30 minutes.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claim that the 2D spacer reduces singlet back-transfer is confounded by surface passivation and by a non-canonical power-law threshold; direct FRET evidence is required.","rationale":"The reader's weakest assumption correctly flags that singlet back-transfer is only inferred indirectly from lifetime changes rather than measured directly. I agree with that concern, but I sharpen it into two concrete weaknesses. First, the lifetime comparison between P1/Rub and P4/Rub is an ambiguous proxy: the bare perovskite lifetimes already differ by 30 ns due to passivation, so the small 1.4 ns difference after rubrene deposition cannot robustly be assigned to FRET injection from rubrene singlets. Second, the P1/Rub power dependence lacks the quadratic-to-linear signature that defines Ith; with slopes of about 1 and 0.5, the kink at 117 mW/cm2 is not a conventional TTA threshold. Together these issues mean the paper's headline mechanistic claim is not yet supported. The empirical observations of improved relative upconversion efficiency below roughly 100 mW/cm2 and improved stability over 30 minutes are plausible and valuable, which is why the reader's CONDITIONAL verdict is appropriate. A direct transient absorption measurement of triplet yield and singlet back-transfer rate would settle the mechanism, while the empirical performance claims could stand on their own as a demonstration of 2D-passivated perovskite-sensitised upconversion.","tokens_in":19139,"tokens_out":5753,"duration_ms":64026,"concrete_test":"Perform femtosecond or nanosecond transient absorption spectroscopy on Control/Rub and P1/Rub with 670 nm excitation, monitoring the rubrene triplet absorption around 500 nm and the perovskite bleach around 750 nm. Extract the rubrene triplet formation yield and the singlet back-transfer rate from the rubrene singlet decay dynamics. If the spacer does not measurably reduce the back-transfer rate, or reduces it less than the passivation-induced increase in triplet yield, the central FRET-mitigation claim is unsupported and the manuscript should be revised to a passivation/stability demonstration.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central mechanism claim that PEA2PbI4 spacers mitigate singlet back-transfer is not established because the evidence cannot separate the proposed distance effect from the layer's passivation effect. The bare perovskite lifetime increases from 31.0 ns (Control) to 61.6 ns (P1) in Table 1, which alone can raise the triplet supply to rubrene at low fluence without any change in FRET back-transfer. The thickness series P1/P4/P8 cannot resolve this confound because spacer thickness, surface passivation, and carrier-transport barrier all vary together with PEAI concentration. The power-dependence data in Figure 4b further weaken the mechanistic inference: P1/Rub exhibits slopes of k = 1.03 then k = 0.56, not the canonical 2 to 1 TTA crossover, so labeling 117 mW/cm2 as an Ith and interpreting it as more efficient TTA is unsupported. If the low-power enhancement stems from improved carrier survival rather than reduced back-transfer, the central mechanistic conclusion collapses, although the empirical low-power efficiency and stability improvements may still stand.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript investigates the use of a 2D PEA2PbI4 spacer layer between a bulk 3D perovskite sensitiser (Br17) and a rubrene:DBP emitter for triplet-triplet annihilation photon upconversion. Four systems are compared: a Control/Rub bilayer and three trilayers with increasing PEAI treatment thickness (P1/Rub, P4/Rub, P8/Rub). The authors report that the P1/Rub trilayer exhibits a lower threshold intensity (Ith = 117 mW/cm2 vs. 250 mW/cm2 for Control/Rub), higher relative upconversion efficiency at sub-solar excitation densities, and improved stability of the upconverted photoluminescence over 30 minutes. They attribute the low-power enhancement primarily to reduced singlet exciton back-transfer from rubrene to the perovskite via FRET, owing to the increased physical separation introduced by the 2D spacer, while acknowledging a trade-off with reduced triplet sensitisation through the thicker spacer layers.","tokens_in":19369,"tokens_out":5155,"duration_ms":52951,"significance":"If the mechanistic claim were firmly established, this work would offer a useful and simple design rule for perovskite-sensitised upconversion: inserting a thin 2D perovskite spacer can suppress parasitic singlet FRET back-transfer while retaining triplet injection, with practical relevance at sub-solar fluences. The paper has clear strengths: the empirical trends are supported by multiple measurement types, the boxplot statistics cover multiple films and spots, and the stability data show a consistent improvement in the trilayer samples. The work also explicitly discusses the trade-off between passivation, charge transport, and FRET suppression, which is a step toward rational device design. However, the central mechanistic attribution to reduced back-transfer is indirect, resting on lifetime comparisons and power-law fits without direct FRET rate measurements or control experiments that isolate the distance effect from surface passivation.","major_comments":[{"comment":"The central claim that the 2D PEA2PbI4 spacer mitigates singlet back-transfer via the 1/r^6 FRET distance dependence is not uniquely supported by the data. The neat-perovskite average lifetime rises from 31.0 ns (Control) to 61.6 ns (P1) and 91.7 ns (P4) after PEAI treatment (Table 1), indicating strong surface passivation that alone raises the triplet supply to rubrene at low fluence. The thickness series P1/P4/P8 cannot separate the proposed distance effect from simultaneous changes in passivation, PEAI concentration, and charge-transport barrier. In addition, the lifetime data are used both to infer FRET-mediated injection (P1/Rub longer-lived than P4/Rub when rubrene is added) and to explain the same lifetime ordering, which is an interpretive loop. A direct measurement of the FRET rate (e.g., transient absorption of rubrene singlets) or a control with an inert spacer of matched thickness is needed before the mechanism in the title can be regarded as established.","section":"Results and Discussion, Table 1 and Figs. 3a-b, 4d"},{"comment":"The power-dependence data for P1/Rub show slopes k = 1.03 at low powers and k = 0.56 at high powers, not the canonical quadratic-to-linear (2 to 1) TTA crossover. Labeling Ith = 117 mW/cm2 as the TTA threshold and interpreting the lower Ith as evidence for more efficient TTA is therefore not justified. The authors rationalize the halved slope via TCA, but no quantitative model or uncertainty estimates for the fitted k and Ith values are provided, and the MPL fits (Fig. 3c) are explicitly 'to guide the eye.' The lower crossover could equally arise from a different balance of injection, passivation, and TCA rather than from reduced back-transfer.","section":"Fig. 4a-c and accompanying text"},{"comment":"The manuscript correctly identifies a trade-off between reduced back-transfer and reduced triplet sensitisation, but it does not quantify this trade-off. Since the high-power upconversion intensity of P1/Rub is lower than that of Control/Rub (Fig. 2b), the improved low-power relative efficiency in Fig. 4d could be dominated by enhanced carrier survival (passivation) rather than by a distance-dependent suppression of FRET. Without a rate-equation analysis that includes the measured lifetimes and an independent estimate of the FRET radius, the relative weights of the two effects remain unknown, so the central conclusion should be softened or supported by additional experiments.","section":"Conclusion / Fig. 4d"}],"minor_comments":[{"comment":"The title contains a spacing typo: 'Back-T ransfer' should read 'Back-Transfer'.","section":"Title page"},{"comment":"The sentence 'the P4/Rub trilayer is shows a low relative upconversion efficiency' should read 'the P4/Rub trilayer shows a low relative upconversion efficiency'.","section":"Results and Discussion, Fig. 4d paragraph"},{"comment":"The caption for Fig. 2b does not state the excitation wavelength and power density used for the boxplot; the main text mentions approximately 4 W/cm^2, but the caption should include this information for completeness.","section":"Fig. 2 caption"},{"comment":"The non-Lorentzian fit function f(B) = A B^2/(|B| + sigma)^2 is stated to be approximate and 'to guide the eye', but the resulting A and sigma values are not reported anywhere, nor are goodness-of-fit measures; adding these values in the Supporting Information would strengthen the MPL discussion.","section":"Supporting Note 4"}],"recommendation":"major_revision","confidential_remarks":"The empirical results, especially the stability improvement and the sub-solar relative efficiency behaviour, are likely of interest to the readership. The main weakness is that the central mechanistic claim is not yet supported: the data cannot exclude surface passivation or altered carrier injection as the origin of the low-power enhancement, and the power-dependence slopes are not canonical. I would advise the editor that a revision with either direct FRET-rate measurements or a substantially softened mechanistic claim is necessary before publication; the present version is not ready as is."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe headline: this paper shows a real effect—putting a thin PEA2PbI4 spacer between a 3D perovskite and rubrene:DBP shifts the upconversion response to lower excitation powers and improves 30-minute stability—but the mechanistic story, reduced singlet back-transfer via FRET distance, is not proven by the data.\n\nWhat's actually new: the spacer strategy has been used in organic and quantum-dot systems, but this is the first time it has been applied to a bulk perovskite sensitiser. The thickness series (P1, P4, P8) is a sensible way to map the trade-off between passivation/barrier effects and FRET mitigation, and the spot-to-spot statistics give the intensity trend some solidity. The stability comparison over 30 minutes is useful and practical. The empirical relative-efficiency crossover at sub-solar intensities, if it holds up, matters for solar applications.\n\nWhere it gets soft: the central attribution to reduced FRET rests on lifetime ordering and the 1/r^6 argument, not on a direct measurement of back-transfer. The confound is real—the PEAI treatment also passivates the 3D surface, roughly doubling the bare perovskite lifetime from 31 to 62 ns, which by itself could supply more triplets at low fluence. The thickness series cannot separate distance from passivation because both scale with PEAI concentration.\n\nThe power-dependence data are the weakest link. P1/Rub shows k going from 1.03 to 0.56, not the canonical 2 to 1. The authors label the crossover Ith and interpret it as more efficient TTA, but a non-canonical slope change does not support that label. They acknowledge the anomaly and invoke TCA, but that is post hoc; without error bars or a kinetic model, the threshold number is not well grounded.\n\nAlso worth noting: the P8 sample contains multiple emissive species and the lifetime fits are triexponential without uncertainties, so differences of a few nanoseconds are not meaningful.\n\nWho this is for: anyone working on perovskite-sensitised upconversion will want the empirical trend and the stability data. The mechanism claim needs stronger evidence before it becomes a citation for FRET suppression—a direct measurement of singlet transfer, or a control with a non-passivating spacer, would settle it.\n\nI would send this to peer review. It is a solid experimental study with a plausible but under-supported mechanism; a good referee can ask for the missing controls and the paper will be stronger for it.","headline":"Useful empirical demonstration that a thin 2D perovskite spacer improves low-power upconversion and 30-minute stability, but the claimed FRET back-transfer mechanism is underdetermined.","tokens_in":19930,"tokens_out":2404,"would_cite":false,"duration_ms":24065,"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":"Inserting a thin 2D perovskite layer between a 3D perovskite sensitiser and rubrene suppresses singlet back-transfer and improves low-light upconversion.","keywords":["photon upconversion","triplet-triplet annihilation","singlet back-transfer","Förster resonance energy transfer","2D perovskite spacer","rubrene emitter","solid-state upconversion"],"falsifier":"Directly measure the rubrene singlet population and lifetime with and without the spacer: excite rubrene singlets selectively and record time-resolved rubrene fluorescence or transient absorption while varying spacer thickness. If the P1/Rub trilayer shows no longer singlet lifetime or no reduced quenching relative to the bilayer, the back-transfer-mitigation mechanism would be refuted even if the relative efficiency trend remains.","tokens_in":18964,"feed_emoji":"🔆","tokens_out":7456,"duration_ms":68067,"temperature":0.7,"pith_summary":"The paper tries to establish that a thin two-dimensional perovskite spacer layer placed between a bulk perovskite sensitiser and a rubrene emitter can reduce a major loss pathway in triplet-triplet annihilation photon upconversion: the back-transfer of upconverted singlet excitons into the sensitiser. It matters because upconversion devices that work at sub-solar light intensities could benefit photovoltaics, photocatalysis, and imaging, but parasitic back-transfer has limited perovskite-sensitised systems. The authors show that a thin PEA2PbI4 layer lowers the intensity threshold for efficient upconversion and gives more stable upconverted emission, at the cost of lower absolute intensity when the spacer is thicker. The central trade-off is between suppressing near-field energy transfer, which improves with distance as $1/r^6$, and maintaining charge transfer that creates triplet excitons in the emitter.","feed_headline":"Thin 2D perovskite spacer boosts low-light upconversion","feed_subtitle":"The trilayer device reaches efficient upconversion at under 100 mW/cm2, roughly half the bilayer's threshold, with steadier output.","key_machinery":"The central object is the two-dimensional Ruddlesden–Popper perovskite PEA2PbI4, formed by spin-coating phenethylammonium iodide onto the bulk perovskite surface. It plays two roles: a passivation layer that lengthens the bulk perovskite photoluminescence lifetime, and a spacer that increases the distance between the rubrene singlet excitons and the strongly absorbing sensitiser, weakening Förster resonance energy transfer back-transfer, which scales as $1/r^6$. The argument is carried by this distance dependence combined with lifetime measurements of the perovskite emission: adding rubrene shortens the P1 and P4 lifetimes relative to their neat films, and the P1/Rub trilayer lives longer than P4/Rub, which the authors attribute to stronger FRET-mediated singlet injection through the thinner spacer. Charge transport through the spacer is thought to proceed via direct transfer, defect-state hopping, or tunnelling, and this is what limits thicker spacers.","core_discovery":"The central claim is that a PEA2PbI4 spacer grown from a 1 mg/mL PEAI treatment (P1/Rub) reduces singlet exciton back-transfer from rubrene:DBP to the 3D perovskite sensitiser by increasing the spatial separation, and that this is what makes the trilayer outperform the Control/Rub bilayer at excitation densities below about 100 mW/cm2. In support, the P1/Rub sample has a lower upconversion threshold intensity ($I_{th}=117$ mW/cm2 versus 250 mW/cm2), a higher relative upconversion efficiency at low power, and retains about 65% of its maximum upconverted emission after 30 minutes, compared to about 40% for the control. The same data show a clear trade-off: thicker spacers (P4, P8) progressively reduce triplet sensitisation and absolute upconversion intensity, and magneto-photoluminescence indicates triplet-charge annihilation is present in all working samples.","pith_inferences":["By the same $1/r^6$ argument, other wide-bandgap, hole-transporting interlayers of comparable thickness should also reduce singlet back-transfer; the paper's mechanism predicts a monotonic relation between spacer thickness and low-power efficiency until charge transfer fails, which could be tested directly.","If the back-transfer mechanism is correct, moving rubrene emission further from the sensitiser absorption edge through chemical substitution or host choice would complement the spacer and could push the efficient regime below 50 mW/cm2 without sacrificing triplet injection.","The paper's lifetime evidence is indirect; a direct transient-absorption measurement of rubrene singlet population as a function of spacer thickness would turn the inferred FRET suppression into a quantitative rate, and could separate passivation effects from distance effects.","Quasi-2D or vertically oriented 2D perovskites, which the authors mention as future directions, would test whether the same back-transfer suppression survives when charge transport through the spacer is improved."],"forward_implications":["At sub-solar excitation densities (below roughly 100 mW/cm2), the P1/Rub trilayer converts more of the absorbed light into upconverted photons than the Control/Rub bilayer, making perovskite-sensitised upconversion more relevant for solar-powered applications.","The threshold for entering the efficient linear TTA regime drops from about 250 mW/cm2 to 117 mW/cm2, meaning the device can operate efficiently at much lower illumination.","Thicker spacers (P4, P8) sacrifice absolute upconversion intensity because charge transfer and triplet formation in rubrene are increasingly hindered, so spacer thickness must be tuned to balance FRET suppression against triplet sensitisation.","Trilayers retain 65% (P1/Rub) and 85% (P4/Rub) of their maximum upconverted emission after 30 minutes, versus about 40% for the bilayer, indicating improved operational stability.","Magneto-photoluminescence shows that triplet-charge annihilation, not only TTA, affects all working samples, so future improvements must manage charge imbalance in the emitter layer as well as back-transfer."],"supporting_citations":[{"why":"Establishes the sequential charge-transfer mechanism by which bulk perovskite films sensitise triplet excitons in rubrene.","marker":"[24]"},{"why":"Shows bulk perovskite/rubrene solid-state upconversion at subsolar fluxes and identifies singlet back-transfer as a loss pathway.","marker":"[25]"},{"why":"Reports the influence of triplet diffusion on perovskite-sensitised upconversion and the FRET back-transfer of singlets near the interface.","marker":"[35]"},{"why":"Provides the triplet transmitter/buffer-layer strategy and the $1/r^6$ distance argument for managing parasitic back-transfer.","marker":"[53]"},{"why":"Supports the claim that 2D perovskite passivation layers reduce interfacial recombination and improve stability.","marker":"[56]"},{"why":"Supplies the PEA2PbI4 formation method and the thickness calibration (about 30 nm at 8 mg/mL) used to prepare the spacers.","marker":"[58]"},{"why":"Gives the relative upconversion efficiency metric and reports the carrier-dependent process that flattens the power-law slope at high excitation.","marker":"[45]"},{"why":"Provides comparison data for the excitation-intensity dependence, including the deviation of the high-power slope from unity.","marker":"[40]"},{"why":"Defines the threshold intensity $I_{th}$ as the crossover from quadratic to linear TTA emission.","marker":"[78]"}],"fun_headline_variants":["2D spacer cuts singlet loss, boosts upconversion","Thin perovskite spacer halves upconversion threshold","Spacer layer reduces back-transfer for efficient upconversion","Low-light upconversion improved by 2D spacer layer","Perovskite spacer enhances upconversion at low power"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's central mechanism rests on reading changes in perovskite photoluminescence lifetimes as evidence of singlet back-transfer from rubrene, but it reports no direct measurement of that energy-transfer step; if the lifetime assignments are wrong, the mechanism loses its support.","fun_headline_variants_meta":{"raw":{"variants":["2D spacer cuts singlet loss, boosts upconversion","Thin perovskite spacer halves upconversion threshold","Spacer layer reduces back-transfer for efficient upconversion","Low-light upconversion improved by 2D spacer layer","Perovskite spacer enhances upconversion at low power"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000218,"raw_usage":{"total_tokens":1481,"prompt_tokens":1028,"completion_tokens":453,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":644,"completion_tokens_details":{"reasoning_tokens":372}},"tokens_in":644,"tokens_out":453,"duration_ms":4748,"temperature":1.0,"reasoning_tokens":372,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:55:13.823268+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly measure the rubrene singlet population and lifetime with and without the spacer: excite rubrene singlets selectively and record time-resolved rubrene fluorescence or transient absorption while varying spacer thickness. If the P1/Rub trilayer shows no longer singlet lifetime or no reduced quenching relative to the bilayer, the back-transfer-mitigation mechanism would be refuted even if the relative efficiency trend remains.","supporting_citations":[{"cited_title":"E.; Klein, N","cited_arxiv_id":null,"evidence_quote":"Establishes the sequential charge-transfer mechanism by which bulk perovskite films sensitise triplet excitons in rubrene."},{"cited_title":"S.; VanOrman, Z","cited_arxiv_id":null,"evidence_quote":"Shows bulk perovskite/rubrene solid-state upconversion at subsolar fluxes and identifies singlet back-transfer as a loss pathway."},{"cited_title":"S.; VanOrman, Z","cited_arxiv_id":null,"evidence_quote":"Reports the influence of triplet diffusion on perovskite-sensitised upconversion and the FRET back-transfer of singlets near the interface."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the triplet transmitter/buffer-layer strategy and the $1/r^6$ distance argument for managing parasitic back-transfer."},{"cited_title":"Surface Passivation Using 2D Perovskites toward Efficient and Stable Perovskite Solar Cells","cited_arxiv_id":null,"evidence_quote":"Supports the claim that 2D perovskite passivation layers reduce interfacial recombination and improve stability."},{"cited_title":"H.; Wang, L","cited_arxiv_id":null,"evidence_quote":"Supplies the PEA2PbI4 formation method and the thickness calibration (about 30 nm at 8 mg/mL) used to prepare the spacers."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the relative upconversion efficiency metric and reports the carrier-dependent process that flattens the power-law slope at high excitation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides comparison data for the excitation-intensity dependence, including the deviation of the high-power slope from unity."},{"cited_title":"S.; Castellano, F","cited_arxiv_id":null,"evidence_quote":"Defines the threshold intensity $I_{th}$ as the crossover from quadratic to linear TTA emission."}],"review_version":1}