{"id":"9a1f88d4-6cbb-4766-b120-2f5536a2453b","arxiv_id":"2504.15227","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A five-step synthesis protocol (calcined Bi2O3, low-temperature quartz-ampoule reaction, gradient purification, coarse particle fraction, Si3N4 die) improves thermal-cycling stability and reproducibility of undoped polycrystalline Bi2O2Se transport properties.","lead":"This paper shows that many inconsistencies in reported properties of polycrystalline Bi2O2Se come from hidden chemical reactions during synthesis and pressing, and it proposes a five-step preparation protocol that yields more stable and reproducible samples. It matters because thermoelectric studies of Bi2O2Se currently lack a reliable undoped baseline, making doping results hard to interpret.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reproducibility claim rests on one pellet per condition and a confounded pressing time: coarse-fraction samples were pressed for 3 h versus 1 h for the legacy samples.","rationale":"The reader correctly identifies the single-pellet-per-condition design as the weakest point. I agree that the reproducibility claim is under-supported, and my analysis sharpens this by showing an additional uncontrolled variable: the coarse-fraction samples pressed for 3 h versus 1 h for the other samples. This means even a fair comparison between Sample 2 and Sample 3 conflates particle-size/gradient purification with pressing dwell time, so the specific contribution of the five-step protocol is not isolated. The ESM result that a toluene-cleaned sample also shows cycling stability further weakens the uniqueness of the proposed protocol. These issues do not invalidate the paper's useful qualitative observations about Bi2O2Se synthesis pitfalls, nor its transparent reporting of cycling data, but they do mean the headline claim of improved reproducibility is not yet quantified. The reader's CONDITIONAL verdict already captures this need for replication, so my read does not move the verdict; it reinforces the condition with a concrete experimental path to test it.","tokens_in":27535,"tokens_out":5075,"duration_ms":48829,"concrete_test":"Prepare at least three pellets from three independent syntheses using the full five-step protocol and at least three pellets using the Sample-2 route (no gradient growth, fine powder), all hot-pressed in a Si3N4 die at 730 °C for 3 h to remove the dwell-time confound. Measure σ and S at 300 K before and after three thermal cycles from 300 to 780 K, and report the mean and spread within each group. If the coefficient of variation across full-protocol pellets is not substantially smaller than across legacy pellets, or if any full-protocol pellet drifts by more than the stated repeatability (±5% for σ, ±3.5% for S), the reproducibility/stability claim is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—'improved reproducibility and stability in thermal cycling' (Section 3.2)—is supported only by Samples 1–4, with one pellet per condition. The comparison is additionally confounded by pressing history: the Experimental section states that 'temperature and pressure were maintained for 1 h for normal powders and 3 h for coarse fraction powders,' so Samples 3/4 (35–340 μm) differ from Samples 1/2 not only by calcination, gradient growth, and sieving, but also by a 3× longer hot-press dwell time. The five-step protocol in Section 3.2 does not list dwell time, so the apparent absence of conductivity hysteresis in Figures 8 and 9 cannot be uniquely assigned to the proposed steps. Furthermore, 'reproducibility' is asserted without any repeated synthesis or multiple pellets: interbatch variance is never estimated, and the stated σRT ≈ 500 S·m-1 and S ≈ −300 μV·K−1 are single-pellet values. The ESM Table S3 also shows a toluene-cleaned sample with Δσ ≈ −0.8% over two cycles, indicating that cycling stability is not specific to the proposed protocol. Thus the load-bearing assumption—that the protocol, rather than uncontrolled batch variation or the unstated dwell-time difference, produces the improved behavior—is not currently established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses reproducibility problems in polycrystalline Bi2O2Se thermoelectrics. The authors identify several sources of variability in the literature and in their own experiments: carbonation of the Bi2O3 precursor, reaction of precursors with quartz ampoules, air sensitivity of as-synthesized powder, chemical reduction by graphite dies during compaction, and formation of foreign phases at grain boundaries. They propose a five-step synthesis protocol (decarbonation of Bi2O3, low-temperature synthesis in a crystallized quartz ampoule, temperature-gradient growth and purification, sieving to a coarse 35–340 μm fraction, and hot pressing in a Si3N4 die) and report that samples prepared accordingly (Samples 3 and 4) show stable transport properties under thermal cycling, with room-temperature electrical conductivity σRT ≈ 500 S·m⁻¹ and Seebeck coefficient S ≈ −300 μV·K⁻¹, in contrast to legacy samples (Samples 1 and 2). The paper also compiles literature data spanning four orders of magnitude in reported conductivity and argues that many nominally pure samples contain undetected foreign phases.","tokens_in":27741,"tokens_out":3763,"duration_ms":34254,"significance":"If the central claim holds, the proposed protocol would provide a badly needed reference baseline for doping studies in Bi2O2Se, and the identification of specific chemical failure modes (reaction with quartz, reduction by graphite, surface oxidation) is a valuable contribution to the community. The paper is unusually candid about residual limitations, explicitly stating that full property stabilization is not achieved and that PXRD phase purity can be misleading. The quantitative significance is currently limited, however, by the fact that the 'improved reproducibility' claim rests on one pellet per processing variant with no replicate batches, and by a confounded pressing-time variable in the comparison. The qualitative mechanistic discussion is strong and well supported by PXRD, SEM/EDS, DTA, and transport data.","major_comments":[{"comment":"The central claim that the presented protocol yields 'improved reproducibility and stability in thermal cycling' is supported by exactly one pellet for each of Samples 1–4. Reproducibility is a statistical property that cannot be established from a single specimen per condition; no repeated synthesis runs, no replicate hot-pressed pellets, and no interbatch variance estimates are provided. The stated σRT ≈ 500 S·m⁻¹ and S ≈ −300 μV·K⁻¹ are single-pellet values. Please either add replicate measurements (ideally at least three independent pellets per variant) or explicitly reframe the claim as preliminary/exploratory and remove the term 'reproducibility' from the central claim.","section":"Section 3.2, Figures 6–9"},{"comment":"The comparison between Samples 1/2 and Samples 3/4 is confounded by pressing time. The Experimental section states that 'temperature and pressure were maintained for 1 h for normal powders and 3 h for coarse fraction powders,' so Samples 3 and 4 differ from Samples 1 and 2 not only by calcination, gradient growth, and sieving, but also by a threefold longer hot-press dwell time. Since the five-step protocol in Section 3.2 does not specify dwell time, the apparent absence of conductivity hysteresis in Figures 8 and 9 cannot be uniquely attributed to the proposed protocol steps. Please hold the dwell time constant across all variants, or provide a clear experimental or mechanistic argument for why the dwell-time difference does not affect the interpretation.","section":"Experimental section, hot pressing parameters"},{"comment":"The ESM shows that a sample from the 100–250 μm fraction that was sonicated in toluene and hot-pressed for 1.5 h without calcination, gradient purification, or the full five-step protocol exhibits Δσ ≈ −0.8% after two thermal cycles, a stability comparable to that of Samples 3 and 4. This indicates that resistance to thermal cycling is not specific to the proposed protocol and can be achieved by other surface treatments. The manuscript needs to reconcile these data with the claim that the presented steps are responsible for the improved cycling stability, for example by redefining the claim or by providing a more controlled comparison where only one variable is changed at a time.","section":"ESM Table S3 and Figure S6"}],"minor_comments":[{"comment":"The yellow columns representing the authors' own data are plotted without error bars or replicate information; please add the number of samples and standard deviations, or state explicitly in the caption that these are single measurements.","section":"Figure 1"},{"comment":"The quoted room-temperature values σRT ≈ 500 S·m⁻¹ and S ≈ −300 μV·K⁻¹ are given without uncertainties, despite the stated measurement repeatability of ±5% for σ and ±3.5% for S; please include the relevant uncertainties.","section":"Abstract"},{"comment":"The reaction equation '3x SiO2 + 2/3·Bi2O3 + 2/3·Bi +·Se → x Bi4(SiO4)3 + Bi2-4xO2-6xSe' appears to be unbalanced as written; please recheck the stoichiometry and the meaning of the dot before Se.","section":"Section 3.1(a), reaction equation"},{"comment":"The sieving-efficiency statement for removal of Bi4(SiO4)3 is supported by only two particle-size fractions; please report the mass yield or residual foreign-phase content for the 35–340 μm fraction used in the final protocol.","section":"ESM Table S1"},{"comment":"The description of the coarse-fraction benefits would benefit from a quantitative statement of the measured surface-to-volume ratio or grain orientation (e.g., from SEM texture analysis), since the qualitative SEM images in Figure 5 are the only current support.","section":"Section 3.2, step 4"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of cond-mat.mtrl-sci and addresses an important practical problem. The mechanistic discussion of failure modes is valuable and the authors are transparent about limitations. However, the load-bearing reproducibility claim is not yet supported by the evidence: single pellets per condition and a confounded pressing time. A major revision that adds replicate samples or re-scopes the claims, and that directly addresses the ESM toluene-cleaned result, would make the paper publishable. I encourage the editor to invite a revision rather than reject, because the core qualitative findings are sound and the methodology discussion is useful to the community."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look if you work on Bi2O2Se or oxide chalcogenides. The paper's real value is in the failure analysis: it documents three concrete contamination/reduction pathways that explain why literature conductivity spans four orders of magnitude – carbonate in the Bi2O3 precursor, Bi4(SiO4)3 formation from the quartz ampoule wall, and reduction by graphite dies to BixSey phases that wet grain boundaries. Those are new, well-supported by PXRD, SEM/EDS, DTA/TG, and they matter beyond this one compound. The thermal-cycling hysteresis data are also a useful reminder that PXRD phase purity does not imply grain-boundary stability.\n\nThe proposed five-step protocol (calcination, low-T synthesis, gradient growth, coarse fraction, Si3N4 die) is plausible and the authors are honest that it mitigates rather than eliminates the problems. Their room-temperature values σ≈500 S/m and S≈−300 μV/K for the coarse-fraction Si3N4 sample are presented as a reference baseline.\n\nNow the soft spots. The stress-test concern lands: the reproducibility claim is supported by exactly one pellet per processing variant (Samples 1–4), and the coarse-fraction samples were pressed for 3 h while the legacy samples were pressed for 1 h. So the improved cycling behavior in Figs. 8 and 9 cannot be uniquely assigned to the protocol steps; dwell time is a confound. The ESM actually shows that a toluene-cleaned sample also has Δσ ≈ −0.8% over two cycles, so cycling stability is not specific to the 5-step protocol. That doesn't kill the paper, but it means 'improved reproducibility' is not established as a statistical claim. The authors could fix this with replicate batches and by either controlling dwell time or reporting it in the protocol list.\n\nThe Arrhenius activation energy is an output, not a fitted input, so no circularity burden. Citation pattern looks fine; self-citations support background claims.\n\nWho's this for? Experimentalists working on Bi2O2Se, especially anyone doing doping studies who needs a stable undoped reference. It deserves a serious referee: the qualitative findings are reproducible evidence and the protocol is a practical contribution. My recommendation is to send it out, with a request for replicate data and a clarification of the dwell-time issue. If the authors can show inter-batch variance with the same protocol, the central claim will hold.","headline":"Useful pitfalls-and-reproducibility study for Bi2O2Se polycrystals; the qualitative chemistry is solid, but the central reproducibility claim rests on one pellet per condition and a confounded pressing time.","tokens_in":28418,"tokens_out":2987,"would_cite":true,"duration_ms":25496,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["72.20.Pa","81.20.Ev"],"model":"deepseek-v4-flash","headline":"A five-step synthesis protocol makes polycrystalline Bi2O2Se reproducible enough to serve as a baseline for doping studies.","keywords":["Bi2O2Se","thermoelectric materials","polycrystalline transport","reproducibility","thermal cycling stability","foreign phases","synthesis protocol","Seebeck coefficient"],"falsifier":"Prepare three or more independent batches using the full five-step protocol, press the 35–340 μm fraction from each batch in a Si$_3$N$_4$ die under identical conditions, and measure room-temperature conductivity and cycling hysteresis; if the batch-to-batch spread of $\\sigma_{RT}$ is comparable to the differences among Samples 1–4, the reproducibility claim is not established. A second check would be extending thermal cycling beyond three cycles to see whether the apparent first-cycle stabilization persists or merely slows.","tokens_in":27321,"feed_emoji":"⚡","tokens_out":9836,"duration_ms":78846,"temperature":0.7,"pith_summary":"The paper is trying to establish why nominally identical polycrystalline Bi$_2$O$_2$Se samples show electrical conductivities spanning four orders of magnitude, and to show that most of that scatter is avoidable. It identifies four controllable sources of variability: carbonate contamination in commercial Bi$_2$O$_3$, reaction of the precursors with quartz ampoules, oxygen uptake on powder surfaces, and reduction of Bi$_2$O$_2$Se by graphite dies during compaction. It then presents a five-step preparation route that suppresses these reactions and produces undoped polycrystalline samples whose transport properties stabilize after the first thermal cycle. For the best fraction, the room-temperature conductivity is about $\\sigma_{RT}\\approx 500$ S·m$^{-1}$ with a Seebeck coefficient of about $-300\\ \\mu$V·K$^{-1}$, and these values are proposed as a baseline for future doping studies. The authors are explicit that the protocol mitigates, but does not fully eliminate, the intrinsic tendency of Bi$_2$O$_2$Se to stoichiometric drift and grain-boundary evolution under thermal cycling.","feed_headline":"Five-step recipe stabilizes Bi2O2Se thermoelectrics","feed_subtitle":"Decarbonation, gradient growth, coarse sieving, and a Si3N4 die yield a reproducible reference for doping studies.","key_machinery":"The load-bearing mechanism is the five-step synthesis protocol, understood as the systematic removal of specific chemical reactions that otherwise occur at precursor surfaces, ampoule walls, powder surfaces, and die interfaces. The named central object is the set of equilibrium reactions such as $3\\,\\mathrm{Bi_2O_2Se}+3\\,\\mathrm{C}\\to \\mathrm{Bi_4Se_3}+2\\,\\mathrm{Bi}+3\\,\\mathrm{CO_2}$ from graphite-die pressing, and surface reactions with SeO$_2$ and CO$_2$ that form Bi$_2$SeO$_5$, Bi$_2$Se$_3$, and Bi$_2$CO$_5$; these reactions have small enthalpies per formula unit and shift with temperature and composition. Each protocol step targets one reaction: calcination removes Bi$_2$O$_2$CO$_3$, a crystallized quartz ampoule slows Bi$_4$(SiO$_4$)$_3$ formation, gradient growth and sieving remove volatile SeO$_2$ and Bi$_2$SeO$_5$ while reducing surface area, and the Si$_3$N$_4$ die eliminates carbon reduction. The work of this machinery is to convert the broad observation that transport properties scatter into a list of controllable chemical causes, each with a measurable fingerprint in the transport data.","core_discovery":"The paper's central claim is that the large scatter of reported transport properties of polycrystalline Bi$_2$O$_2$Se is not intrinsic to the compound but is produced by a small set of identifiable side reactions: carbonate contamination of the Bi$_2$O$_3$ precursor shifts the starting stoichiometry; molten or solid precursors react with quartz ampoules to form Bi$_4$(SiO$_4$)$_3$ and leave the matrix Se-rich; ambient exposure coats powder surfaces with oxygen-derived phases that later react during compaction; and pressing in graphite dies reduces Bi$_2$O$_2$Se to Bi$_x$Se$_y$ phases that wet grain boundaries and inflate conductivity. The authors show that samples that appear phase-pure by powder XRD can still carry these invisible foreign phases. They then argue that a five-step protocol—decarbonation, low-temperature synthesis in crystallized silica ampoules, temperature-gradient growth with self-purification, sieving to a 35–340 μm fraction, and hot pressing in a Si$_3$N$_4$ die—suppresses most of these reactions and yields undoped polycrystalline material with stable transport under thermal cycling, with room-temperature $\\sigma_{RT}\\approx 500$ S·m$^{-1}$ and Seebeck coefficient $S\\approx -300\\ \\mu$V·K$^{-1}$. Even with this protocol, the measured activation energy of about 0.20 eV (band gap $\\approx 0.40$ eV) is roughly half the theoretical 0.85 eV, which the authors take as evidence that residual native defects and grain-boundary phases still influence transport and that full stability has not been achieved.","pith_inferences":["A direct test of the reproducibility claim would be to run the full protocol on at least three independent batches and report the batch-to-batch spread of $\\sigma_{RT}$; the paper itself compares one pellet per variant, so the spread remains unquantified.","The supplementary results on toluene-cleaned and water-milled powders suggest that particle-surface cleaning may be a sixth independent variable worth systematic study; if cleaning alone can stabilize cycling, the surface-oxide story gains support.","If foreign phases invisible to XRD are as widespread as the authors argue, dopant solubility limits reported from XRD-only studies will tend to be overestimated, and re-measuring those systems with electron microscopy would be a cheap falsification test.","The claim that Bi$_2$O$_2$Se is intrinsically prone to stoichiometric drift implies that single-cycle comparisons between undoped and doped samples are unreliable; comparing only fully cycled, stabilized samples would be a more stringent test of doping effects."],"forward_implications":["Doping studies that hot-press or spark-plasma-sinter in graphite dies may be measuring Bi$_x$Se$_y$-wetted composites rather than doped Bi$_2$O$_2$Se, so previously reported doping effects should be re-examined with a Si$_3$N$_4$ die.","A powder-XRD phase-purity pattern is not sufficient evidence of a clean sample, because thin or amorphous foreign phases can control grain-boundary transport while remaining invisible to XRD.","If the five-step protocol becomes standard, undoped Bi$_2$O$_2$Se with $\\sigma_{RT}\\approx 500$ S·m$^{-1}$ and $S\\approx -300\\ \\mu$V·K$^{-1}$ can serve as a cross-laboratory reference for comparing doping and composite studies.","Even with the improved protocol, the activated behavior above 500 K and the residual drift in the air-stored sample imply that truly intrinsic polycrystalline Bi$_2$O$_2$Se has not yet been prepared."],"supporting_citations":[{"why":"The 2018 Ge-doping study that first highlighted the foreign-phase problem in Bi2O2Se and supplies the baseline this paper argues against.","marker":"[24]"},{"why":"Documents the extreme oxygen sensitivity of Bi2O2Se nanoplatelets and the ~3.7 eV O2 adsorption energy used to explain air-exposure effects on powders.","marker":"[6]"},{"why":"Provides the room-temperature formation and thermal decomposition behaviour of Bi2O2CO3 that motivates the Bi2O3 calcination step.","marker":"[36]"},{"why":"Supplies the formation enthalpies for the graphite-reduction and surface-equilibrium reactions that the paper invokes to explain die reactions and cycling instability.","marker":"[45]"},{"why":"Reports the theoretical band gap of about 0.85 eV against which the measured 0.40 eV activation-type gap indicates residual native defects.","marker":"[47]"},{"why":"Argues that foreign phases can modulate-dope the Bi2O2Se matrix, which the paper uses to explain how invisible grain-boundary phases change transport.","marker":"[44]"}],"fun_headline_variants":["Why Bi2O2Se transports vary—and how to fix it","Five-step route to reliable Bi2O2Se thermoelectrics","Reproducible Bi2O2Se despite graphite, quartz, carbonates","Bi2O2Se variability traced to hidden side reactions","Stable Bi2O2Se via five-step synthesis recipe"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim of improved reproducibility rests on comparing a single pellet for each processing variant, so the conclusion would collapse if uncontrolled batch-to-batch variation turned out to be as large as the differences the paper attributes to its protocol.","fun_headline_variants_meta":{"raw":{"variants":["Why Bi2O2Se transports vary—and how to fix it","Five-step route to reliable Bi2O2Se thermoelectrics","Reproducible Bi2O2Se despite graphite, quartz, carbonates","Bi2O2Se variability traced to hidden side reactions","Stable Bi2O2Se via five-step synthesis recipe"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000268,"raw_usage":{"total_tokens":1703,"prompt_tokens":1118,"completion_tokens":585,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":734,"completion_tokens_details":{"reasoning_tokens":493}},"tokens_in":734,"tokens_out":585,"duration_ms":5424,"temperature":1.0,"reasoning_tokens":493,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:29:28.264261+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Prepare three or more independent batches using the full five-step protocol, press the 35–340 μm fraction from each batch in a Si$_3$N$_4$ die under identical conditions, and measure room-temperature conductivity and cycling hysteresis; if the batch-to-batch spread of $\\sigma_{RT}$ is comparable to the differences among Samples 1–4, the reproducibility claim is not established. A second check would be extending thermal cycling beyond three cycles to see whether the apparent first-cycle stabilization persists or merely slows.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The 2018 Ge-doping study that first highlighted the foreign-phase problem in Bi2O2Se and supplies the baseline this paper argues against."}],"review_version":1}