{"id":"0a08192f-67a9-4b18-b642-e7ae2d90039c","arxiv_id":"2411.14836","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Silanization lowers glass bead surface energy and roughness, and the coating resists humidity-induced nanoscale changes except at sparse high-energy uncovered sites.","lead":"This paper uses nanoscale measurements to show how a silane coating changes the surface of glass beads and how that surface responds to humid storage. The coating lowers surface energy and roughness, and it mostly protects the beads from humidity damage, though some moisture-driven changes still occur in gaps in the coating.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Single-column BET normalization may create the observed energy reductions as a coverage-axis artifact","rationale":"The reader's weakest assumption flagged the general lack of replicates and error bars, and the potential inaccuracy of BET/Schultz assumptions. I agree that small-N is a concern, but I identify a more specific and systematic mechanism: the use of a single-column BET value to set the coverage axis for each treated sample. This is not simply a random-error problem; it can produce a directional bias in exactly the direction of the reported findings. If the single-column S_BET for VR-0.1 is even modestly underestimated, the energy distributions are shifted along the coverage axis, making the treated sample appear lower in energy. The same bias affects the comparison of before/after humidity data, since the coverage scale is recalibrated using a single post-exposure BET measurement. This is load-bearing because the paper's central claims (reduced dispersive and specific energy, and differential humidity response) are all read off these coverage-dependent curves. A concrete test — replicate BET measurements and recomputation of the distributions with alternative S_BET values — would settle whether the trends persist. The contact-angle data provide independent macroscopic support for an overall energy reduction, so I would not reject the paper, but the nanoscale energy-component claims remain conditional pending this check.","tokens_in":20916,"tokens_out":14026,"duration_ms":143825,"concrete_test":"Determine S_BET for the treated samples from three independent IGC columns (or from nitrogen adsorption at 77 K as a cross-check) and report the mean and confidence interval. Recompute all surface-energy distributions using the upper and lower bounds of S_BET as alternate coverage calibrations for each treated sample. If the recomputed γ_D and γ_SP curves for VR-0.1 overlap with the untreated curves within experimental uncertainty, the headline conclusion is invalidated as a normalization artifact; if the separation persists across the full S_BET range, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The surface-energy distributions (Figures 5, 6, 9) are plotted against surface coverage θ = N/n_m, with n_m obtained from the BET surface area of the same column via Eq. 12. For treated samples, Section 2.3.2 states BET values were determined from a single column. Any error in that single S_BET scales the entire θ-axis for that sample. Because γ_D and γ_SP are monotonically decreasing in θ, an underestimated S_BET for VR-0.1 (0.057 vs 0.069 m²/g for untreated) shifts its curves to lower energies at fixed θ, mimicking a genuine reduction in per-site energy. The same mechanism affects the before/after humidity comparison: if the true S_BET of the treated sample is unchanged or higher, the reported low-coverage increase could be an artifact of a drifting coverage calibration. The paper does not report BET C constants, isotherm shapes, or confidence intervals for the treated samples, so this systematic normalization bias cannot be excluded. The magnitude is non-negligible: a 17% underestimate in S_BET shifts θ by ~17%, comparable to the separation between untreated and VR-0.1 curves. Thus the central claim that silanization reduces both dispersive and specific components of surface energy, and that treated samples are stable under humidity, is not robust to the BET normalization method.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript uses inverse gas chromatography (IGC) and atomic force microscopy (AFM) to characterize the nanoscale chemical and structural changes of soda-lime glass beads after Surfasil silanization at three concentrations (VR-0.0002, 0.001, 0.1). The authors report that silanization reduces both the dispersive and specific components of surface energy (Figures 5–6), lowers BET surface area from 0.069 to 0.057 m²/g (Figure 7), and produces a smoother but more feature-dense surface in AFM height maps (Figure 8). In a second experiment, untreated and VR-0.1 treated beads are exposed to 90% relative humidity for 20 days; the untreated sample shows broad increases in surface energy and roughness, whereas the treated sample shows changes only at low surface coverage, which the authors attribute to incomplete coating regions. A conceptual model (Figure 10) is proposed for the water-adsorption-induced modifications, and the authors recommend strict humidity control during storage for reproducible wettability experiments.","tokens_in":21178,"tokens_out":7168,"duration_ms":68099,"significance":"If the results are robust, the manuscript provides a useful link between macroscopic contact-angle wettability characterization and nanoscale surface-energy and roughness measurements using the same silanization protocol. The combination of IGC and AFM is appropriate for the research question, and the humidity-exposure experiment addresses a practically important issue for experimental reproducibility. The authors are transparent about several limitations, including single-column BET for treated samples and single AFM scans per sample. However, the quantitative claims—reduction in surface energy components and humidity-induced changes—rest on a coverage coordinate that is calibrated using BET values from the same IGC data, and no error bars or replicate experiments are reported for the energy distributions. The central claims are therefore not yet supported to the standard expected for publication; they require either additional replicates and error analysis or a re-presentation of the data on an absolute-coordinate basis.","major_comments":[{"comment":"The surface-energy distributions in Figures 5, 6, and 9 are plotted against surface coverage θ = N/n_m, where n_m is obtained from the BET surface area via Eq. (12). For treated samples, the BET surface area is determined from a single column (Section 2.3.2). Any error in that single-column S_BET scales the entire coverage axis for that sample, and because γ^D and γ^SP decrease monotonically with θ, an underestimated S_BET for VR-0.1 (0.057 vs 0.069 m²/g) shifts its curves to lower energies at fixed θ, potentially mimicking a genuine reduction in per-site energy. The same mechanism affects the before/after humidity comparison in Section 3.2: the post-exposure curves use the post-exposure BET values (0.078 and 0.062 m²/g), so the reported increases at fixed coverage may partly reflect this rescaling rather than real surface changes. Please report the n-octane isotherms, BET C constants, and fitting ranges (Eq. 14), and either provide replicate BET measurements or plot the energy distributions against an absolute coverage variable that avoids the between-sample normalization.","section":"§2.3.2 and §3.1, Eq. (12)"},{"comment":"The claim that silanization makes the surface progressively smoother and more feature-dense is based on a single 10 µm × 10 µm AFM height map per sample, and no quantitative roughness parameters (e.g., RMS roughness, Ra) or error bars are provided. The height histograms in Figure 8(E) are also reported without uncertainties. Given the small scan size relative to the 1–2 mm bead diameter, multiple scans per sample are needed to support this conclusion; please add at least three scans per condition and report quantitative roughness metrics.","section":"§2.4 and Figure 8"},{"comment":"The humidity-exposure experiment is performed with one column per sample type. The central comparison—untreated shows broad increases while VR-0.1 shows only low-coverage changes—is thus based on single curves. The 'low-coverage only' claim is sensitive to the exact shape of the energy distribution and could be affected by the BET normalization bias described in the first major comment. Without replicates and confidence intervals on the curves, or an alternative analysis, the distinction between a localized real modification and a coverage-axis artifact cannot be established.","section":"§3.2"},{"comment":"The post-exposure BET increase for the VR-0.1 sample (0.057 → 0.062 m²/g, a ~9% change) is presented as marginal. However, no replicate measurements or method precision are reported for treated samples, so the statistical significance of this 9% change is unknown. Given that the BET is determined from a single column, this change could be within measurement noise. Please provide replicate BET determinations or a precision estimate (e.g., from the untreated replicates) to support the claim that the treated sample is stable under humidity.","section":"§3.2 and Figure 7"}],"minor_comments":[{"comment":"The James–Martin correction factor in Eq. (2) has mismatched parentheses; the formula should be j = (3/2) * [((ΔP/P_atm)+1)^2 − 1] / [((ΔP/P_atm)+1)^3 − 1].","section":"§2.3.1, Eq. (2)"},{"comment":"There are several typographical errors: 'Avagardo's number' should be 'Avogadro's number'; 'as showin in Figure 3' should be 'as shown in Figure 3'; 'the slop of the linear regression' should be 'the slope of the linear regression'.","section":"§2.3.1"},{"comment":"The reported BET value reads '0. 0. 069 ± 0. 003 m2/g', which contains a stray '0.'; it should read '0.069 ± 0.003 m²/g'.","section":"§3.1"},{"comment":"The y-axis label uses 'gr' as an abbreviation for gram; please use 'g' for consistency with the rest of the manuscript.","section":"Figure 7"},{"comment":"The legend labels are embedded in the plot area and the curves may be difficult to distinguish in grayscale; please clarify line styles or colors in the captions.","section":"Figures 5 and 6"},{"comment":"Please specify the number of injection points and the partial-pressure range used for the BET analysis from the n-octane isotherm, so that readers can assess the validity of the BET fit.","section":"§2.3.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript relies heavily on the authors' own prior protocol paper [15] for contact angles; this is acceptable, but the novelty is incremental. The single-column BET normalization issue is the main technical concern. If the authors can re-analyze the energy distributions without the coverage-axis normalization or provide replicate measurements, the paper could become acceptable. I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuinely useful, if narrow, characterization study. The new bits are the IGC surface-energy distributions and the accelerated humidity test on Surfasil-treated beads. The main message—store silanized glass under controlled humidity or risk slow nanoscale changes—is plausible and worth passing on to any lab doing multiphase flow in glass bead packs.\n\nWhat it does well: The IGC methodology is laid out carefully, and the combination with AFM gives two independent views of the same treatment. The trend that silanization lowers both dispersive and specific surface energy is consistent with the contact angle increase from the authors' prior protocol paper. The humidity experiment is well designed: same column before/after, in-situ drying, and the conclusion that the treated sample only changes at low coverage (incomplete coating regions) is reasonable.\n\nSoft spots are real but not fatal. The biggest is sample size: one IGC column for each treated condition, one 10x10 µm AFM image per sample, and no error bars on the energy distributions. That means the quantitative magnitudes of the reported shifts are uncertain, even if the direction is consistent across methods. The BET normalization concern in the stress test does not hold up: if the treated BET area is underestimated, the reported coverage axis is inflated, which would shift the treated curve upward (higher apparent energy), not downward. So the observed reduction in surface energy is not an artifact in that direction; if anything, the true reduction could be larger. The paper itself flags the single-column BET for treated samples, which is honest but undercuts precision.\n\nThe conceptual model in Figure 10 is descriptive, not predictive, but that's appropriate for the evidence. The citation pattern is fine; self-citation to the protocol paper is justified since they build directly on it.\n\nWho it's for: researchers using silanized glass for micromodel or beadpack experiments, and anyone doing IGC on low-surface-area materials. It deserves peer review as a solid experimental contribution, with the expectation that the authors add replicates and quantitative roughness metrics for a revision.","headline":"Useful, small-N surface characterization; the humidity storage warning is the real takeaway, and the proposed BET-normalization artifact points the wrong way.","tokens_in":21724,"tokens_out":4075,"would_cite":true,"duration_ms":42982,"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":"Silanization shields glass surfaces from humidity-driven nanoscale wear","keywords":["wettability alteration","silanization","surface energy","inverse gas chromatography","atomic force microscopy","humidity storage stability","glass beads","BET surface area"],"falsifier":"Take replicate columns (at least three) of untreated and VR-0.1 treated beads, measure dispersive and specific surface energy before and after 20 days at 90% relative humidity, and also image several 10 $\\mu\\mathrm{m}$ regions per bead; if the untreated-versus-treated differences in surface energy and BET area overlap within uncertainty, the claimed protective effect of silanization would not be confirmed.","tokens_in":20738,"feed_emoji":"🔬","tokens_out":4534,"duration_ms":41779,"temperature":0.7,"pith_summary":"This paper tries to establish what silanization actually does to glass at the nanoscale and whether the treatment survives storage in humid air. Using inverse gas chromatography and atomic force microscopy on glass beads, it argues that a Surfasil coating lowers both the dispersive and the specific (polar) parts of surface energy, shrinks the measured BET surface area, and smooths the roughness features seen on untreated glass. The same tools are then used to compare untreated and strongly silanized beads after 20 days at 90% relative humidity. The claim is that untreated glass gains polar groups and roughness, while the silanized beads change only at low surface coverage, where the coating is incomplete. The practical stake is that humidity control during storage matters for reproducible wettability in porous-media experiments.","feed_headline":"Water vapor roughens bare glass, spares silanized coating","feed_subtitle":"Nanoscale imaging and gas chromatography show treated beads stay stable under 90% humidity for 20 days.","key_machinery":"The load-bearing device is surface-coverage-resolved surface energy from inverse gas chromatography (IGC), a technique in which pulses of alkane and polar probe gases travel through a column packed with beads and retention times are converted into dispersive and specific surface energies at each surface coverage. Because low coverage samples high-energy sites (cracks, narrow pores) and high coverage reports the average surface, the curves let the authors separate chemical from structural changes. The Schultz linear-regression method supplies the dispersive values; BET applied to the n-octane isotherm supplies surface area; and AFM height histograms over $10\\ \\mu\\mathrm{m}\\times 10\\ \\mu\\mathrm{m}$ areas supply direct topography. The conceptual model in Figure 10 ties these together: adsorbed water preferentially collects on exposed, uncoated glass, so treated samples degrade only where the silane coating is patchy.","core_discovery":"Silanization of soda-lime glass beads with dichlorooctamethyltetrasiloxane (Surfasil) at three dilutions reduces the dispersive and specific components of surface energy at all surface coverages, with the basic (electron-donor) parameter dropping while the acidic parameter stays flat, which the authors read as selective substitution of silanol groups. BET surface area falls from $0.069$ to $0.057\\ \\mathrm{m^2/g}$ and AFM height histograms narrow, so the surface becomes smoother but more feature-dense, indicating patchy coating. After 20 days at 90% relative humidity, the untreated reference shows broad increases in both surface-energy components and a rise in BET area to $0.078\\ \\mathrm{m^2/g}$, whereas the VR-0.1 treated sample shows only low-coverage increases and a small BET rise to $0.062\\ \\mathrm{m^2/g}$. The paper interprets this as water vapor attacking exposed glass at incomplete-coating regions while the hydrophobic coating itself resists reaction.","pith_inferences":["If patchy coverage is the weak point, a second dilute silanization pass aimed at the residual high-energy sites should suppress the low-coverage energy rise seen after humid storage; that is a testable prediction the paper does not make.","The single-column and single-scan sampling means the claimed stability difference should be checked with replicate columns and multiple AFM regions before using these numbers as quantitative benchmarks.","Because IGC averages over the whole column while AFM images a $10\\ \\mu\\mathrm{m}$ patch, pairing the two techniques could be used to estimate the fraction of exposed glass area from the magnitude of the low-coverage energy increase.","For multiphase-flow experiments, these results imply that storage history can confound wettability comparisons; reporting relative humidity during storage may matter as much as reporting the contact angle."],"forward_implications":["Stored untreated glass beads become gradually more hydrophilic and rougher, so experiments using them as water-wet references have a shelf life set by storage humidity.","VR-0.1 silanized beads keep their hydrophobic character under humid storage far better than untreated beads, but not perfectly: low-coverage surface energy still rises.","Humidity, not just contact with liquid water, should be treated as a variable in preparing and storing glass micromodels and bead packs.","Surface-energy distributions measured by IGC can reveal where a coating is incomplete, since residual high-energy sites appear at low surface coverage.","The same IGC/AFM workflow can quantify other surface treatments' nanoscale chemical and structural effects on porous media."],"supporting_citations":[{"why":"Supplies the silanization procedure and contact-angle baselines that this paper's treated samples are built from.","marker":"[15]"},{"why":"Provides the Schultz linear-regression method used to extract dispersive surface energy from alkane retention volumes.","marker":"[38]"},{"why":"Supplies the IGC methodology for measuring surface energy distributions from finite-concentration injections.","marker":"[31]"},{"why":"Gives prior IGC characterization of silanized siliceous materials that this work extends to glass beads.","marker":"[13]"},{"why":"Provides evidence for patchy silane coverage, used to explain humidity-induced changes at incomplete coating regions.","marker":"[12]"},{"why":"Molecular simulation of hydroxylation of silica surfaces by water, the basis for attributing increased polar groups to water adsorption.","marker":"[27]"},{"why":"Review of aqueous alteration of silicate glass, used to justify roughening and reaction with adsorbed water during storage.","marker":"[29]"}],"fun_headline_variants":["Silanized glass resists humidity; bare glass roughens","Water vapor attacks bare beads, not silane-coated ones","Humidity alters untreated glass, spares silanized surface","Silanization lowers surface energy, stabilizes glass against moisture"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The trends rest on one IGC column per treated sample and one 10-by-10-micrometer AFM scan per sample, with the additional assumption that the BET and Schultz analyses give unbiased surface areas and energies for these low-area beads.","fun_headline_variants_meta":{"raw":{"variants":["Silanized glass resists humidity; bare glass roughens","Water vapor attacks bare beads, not silane-coated ones","Humidity alters untreated glass, spares silanized surface","Silanization lowers surface energy, stabilizes glass against moisture"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000288,"raw_usage":{"total_tokens":1714,"prompt_tokens":992,"completion_tokens":722,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":608,"completion_tokens_details":{"reasoning_tokens":651}},"tokens_in":608,"tokens_out":722,"duration_ms":8695,"temperature":1.0,"reasoning_tokens":651,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:49:21.772066+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take replicate columns (at least three) of untreated and VR-0.1 treated beads, measure dispersive and specific surface energy before and after 20 days at 90% relative humidity, and also image several 10 $\\mu\\mathrm{m}$ regions per bead; if the untreated-versus-treated differences in surface energy and BET area overlap within uncertainty, the claimed protective effect of silanization would not be confirmed.","supporting_citations":[{"cited_title":"Vukovic, J","cited_arxiv_id":null,"evidence_quote":"Supplies the silanization procedure and contact-angle baselines that this paper's treated samples are built from."},{"cited_title":"Schultz, L","cited_arxiv_id":null,"evidence_quote":"Provides the Schultz linear-regression method used to extract dispersive surface energy from alkane retention volumes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the IGC methodology for measuring surface energy distributions from finite-concentration injections."},{"cited_title":"Bauer, R","cited_arxiv_id":null,"evidence_quote":"Gives prior IGC characterization of silanized siliceous materials that this work extends to glass beads."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides evidence for patchy silane coverage, used to explain humidity-induced changes at incomplete coating regions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Molecular simulation of hydroxylation of silica surfaces by water, the basis for attributing increased polar groups to water adsorption."},{"cited_title":"Gin, J.-M","cited_arxiv_id":null,"evidence_quote":"Review of aqueous alteration of silicate glass, used to justify roughening and reaction with adsorbed water during storage."}],"review_version":1}