{"id":"2c467b4e-6079-4077-9756-072dcb22e04a","arxiv_id":"2507.00208","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Refraction at the curved droplet surface explains the distorted images seen in Raman microscopy of sessile droplets, and a horizontal laser configuration maps composition near the three-phase contact line.","lead":"This paper shows how light bends when a Raman laser enters a curved liquid droplet, and uses that understanding to map chemical composition inside evaporating droplets. The authors add a sideways laser setup that can measure right at the edge where the droplet meets the surface, a region that standard vertical scans miss.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The horizontal-scan concentration maps near the contact line are interpreted in the region where the refraction model explicitly fails to compute the focus; their spatial accuracy is unvalidated.","rationale":"The reader's weakest assumption identifies the same gap: the geometric-optics focus prescription is unvalidated exactly where the model fails to compute the focus, yet the paper interprets contact-line concentration maps there. I agree that this is the most load-bearing concern because the paper's key application—marker-free concentration measurements near the three-phase contact line—relies on trusting the spatial assignment of Raman signals in precisely that region. The paper's own admission in Section 3.3 that the simulation cannot determine the new focus for deep positions near the substrate means the positional uncertainty is not merely a quantitative error bar but a missing piece of the argument. The claim that 'the remaining part of the laser cone continues to penetrate the droplet' only justifies that a signal is collected, not that the confocal volume is located at the intended position or that the collected spectrum is representative of that nominal point. Without a positional validation, the reported concentration gradient and its trend during evaporation (81.3 to 71.9 mol%) are not fully supported. The proposed experimental check would settle the issue directly by measuring apparent versus actual positions in the relevant scan region. A wave-optics simulation could also be used, but an experimental test is more decisive because it captures all aberrations and alignment effects. I therefore keep the verdict at CONDITIONAL, unchanged from the reader, since the central geometric-optics modeling for vertical scans appears sound and the artifacts needed to close the gap are within reach, but the horizontal application's quantitative claims require this validation.","tokens_in":15262,"tokens_out":3136,"duration_ms":35188,"concrete_test":"Place a sub-resolution Raman-active marker (e.g., a 1 µm bead or a thin dye layer) at known positions inside a non-evaporating droplet of matched refractive index, and acquire horizontal-configuration Raman images in the region where the model predicts cone blockage (x* > 0.7). Compare the apparent (nominal) positions of the marker with its actual positions. If the apparent positions deviate by more than the lateral pixel size (14 µm), the concentration maps in Fig. 12 require a positional correction; if they match, the assumption that measurements remain positionally accurate despite the blocked cone is validated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central demonstration—resolving the concentration distribution near the three-phase contact line with a horizontal laser (Section 3.3, Fig. 12)—depends on assigning Raman spectra to nominal scan positions. Yet the authors state that for horizontal scans close to the substrate, 'the simulation fails to calculate the refracted focus for positions deep in the droplet' because the lower beam of the laser cone is interrupted by the substrate (Fig. 9; Fig. 10, light-green curve breaks at x* ≈ 0.7). They then assert that measurements remain possible 'albeit with less intensity' and present water-concentration maps (Fig. 12) without applying any positional correction in that region. If the focus is shifted—horizontally or vertically—when part of the cone is blocked, the apparent positions in Fig. 12 do not correspond to the actual probed locations. The reported decrease in water concentration from 81.3 to 71.9 mol% at the contact line could be spatially mislocated or an artifact of the unmodeled focus shift. The paper provides no derivation or experimental verification that intensity loss is the only consequence of the cone blockage; this assumption is load-bearing for the headline concentration measurement.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a geometric-optics ray-tracing model for the focus shift of a confocal Raman microscope when the beam is refracted at the curved liquid-vapor interface of a sessile droplet. The focus is computed as the intersection of the two outer rays of the objective cone after Snell refraction at the measured drop contour. The model is applied to vertical and horizontal laser incidence for droplets of different contact angles, and the simulated droplet shapes in the Raman image are compared with measured CH-integral images. A horizontal-laser setup using a 45-degree mirror is introduced to access the three-phase contact region, and water concentration maps of an evaporating glycerol/water droplet are presented, reporting a decrease in water concentration at the contact line from 81.3 to 71.9 mol%.","tokens_in":15463,"tokens_out":6019,"duration_ms":62437,"significance":"The refraction model is a useful practical tool: it is parameter-free with respect to the Raman data, uses only Snell's law, the stated refractive index, and the measured drop contour, and it qualitatively reproduces the dome-shaped distortion observed for vertical scans and the V-shaped cutout for horizontal scans. The code is publicly available. The horizontal configuration is a reasonable approach for accessing the contact-line region. However, the quantitative validation of the model is presently only visual, and the contact-line concentration map is obtained in a region where the model cannot compute the focus; both points affect the strength of the headline claim. If these issues are resolved, the paper would provide a valuable guide for Raman measurements of sessile droplets.","major_comments":[{"comment":"The concentration maps in Fig. 12 are interpreted in the region where the refraction model does not compute a focus. In Section 3.3 you state that for horizontal scans close to the substrate the simulation breaks off at approximately x* = 0.7 because the lower cone beam is interrupted by the substrate, and that 'the simulation fails to calculate the refracted focus.' You then assert that measurements remain possible with reduced intensity, but this asserts that the focus position is unchanged by the partial blockage. If the focus is displaced horizontally or vertically when part of the cone is blocked, the nominal positions in Fig. 12 do not correspond to the probed locations, and the reported decrease from 81.3 to 71.9 mol% at the contact line could be spatially mislocated. Please provide experimental verification (e.g., a scan across a sharp interface or a comparison with a configuration where the focus is computable) that intensity loss is the only consequence, or restrict the quantitative interpretation to the region where the focus is computed.","section":"Section 3.3, Figs. 10 and 12"},{"comment":"The claim that the simulated drop shape 'fits well' with the measured shape is not supported quantitatively. In Figs. 1, 7, and 8 the comparison is visual only; no metric such as the RMS deviation between the simulated and measured contours is reported. Since the validity of the central claim rests on this fit, please provide a quantitative comparison of the contours for the four contact angles studied.","section":"Section 3.2, Figs. 1, 7, 8; abstract"},{"comment":"The water concentration values and maps are presented without error bars, replicate droplet measurements, or uncertainty estimates. The calibration procedure is only referenced to Ref. [16], and the concentration difference between the start and end of evaporation is a key quantitative result. Without an assessment of measurement noise and systematic errors (including the position error discussed in the first comment), the significance of the observed 81.3 to 71.9 mol% decrease cannot be evaluated.","section":"Section 3.3, Fig. 12 and SI Sec. 2"}],"minor_comments":[{"comment":"The phrase 'the Raman laser has to undergo a phase transition' should be replaced with a reference to the liquid-vapor interface or phase boundary.","section":"Abstract"},{"comment":"The caption contains a duplicated phrase: 'contact line contact line'.","section":"Section 3.3, Fig. 8 caption"},{"comment":"The Gaussian beam limitation is acknowledged, but please state explicitly that this limitation may also affect the contact-line maps of Fig. 12, where many measurement points are near the substrate or surface.","section":"Section 3.4"},{"comment":"The resolution '14 × 22.5 µm²' is called 'unprecedented'; please provide a comparison with previously reported spatial resolutions in Raman studies of droplets to support this claim.","section":"Conclusion"},{"comment":"The manuscript contains several typographical and spacing errors (e.g., 'varie ty of everyday situations' in the abstract); a careful proofreading pass is needed.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of the journal. The main novelty is the refraction model and the horizontal-scanning configuration; the concentration-map example is a direct extension of the authors' earlier work (Ref. [16]) and should be framed as such. The editor may wish to ask for an independent validation of the partially-blocked-cone assumption before publication, as the headline concentration measurement currently rests on an unverifiable position assignment in the region where the model breaks down."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is my read. The paper does something genuinely useful: it takes Everall's flat-interface refraction correction, applies it to a curved, asymmetric sessile droplet surface by treating the left and right cone rays separately, and adds a horizontal 45-degree-mirror geometry for reaching the contact line. The model is parameter-free with respect to the Raman data—only Snell's law, the stated refractive index, and the measured drop contour go in—and the code is deposited. For the vertical-scan geometry, the simulated drop shapes reproduce the dome-shaped distortion and its contact-angle dependence. That part is solid and worth having.\n\nThe soft spot is the horizontal-scan application, and it is not a small one. The paper's own Section 3.3 says the simulation fails to calculate the refracted focus for positions deep in the droplet near the substrate, because the lower beam of the cone is blocked. The authors then assume the remaining cone 'continues to penetrate' so the measurement is simply dimmer. But the measured Raman signal depends on the full overlap of illumination and collection cones; blocking one side can shift the probe volume, not just reduce intensity. No derivation, simulation, or experiment checks that. Yet the headline result—the contact-line water concentration falling from 81.3 to 71.9 mol%—is presented from maps in that very region, with no positional correction. The spatial accuracy of those maps is simply unvalidated. The stress-test note is right.\n\nOther issues are more minor. The 'fits well' between simulated and measured drop shapes is judged visually; there are no quantitative fit metrics, no replicate counts, no error bars on the concentration maps. The concentration calibration comes from the authors' previous PNAS paper rather than being re-established here. And the Discussion concedes Gaussian beam effects near the surface, but does not quantify them. The reference list is broad and the self-citation is reasonable in context.\n\nWho gets value: experimentalists doing Raman on sessile droplets, especially those studying contact-line transport. This deserves a serious referee, not a desk reject, because the vertical-scan refraction model is a legitimate methodological contribution. But the horizontal contact-line concentration claim needs to either be supported by a proper treatment of the partially blocked aperture or be sharply qualified until such treatment exists.","headline":"Useful curved-droplet refraction correction, but the headline horizontal contact-line concentration maps are presented in the region where the model admittedly cannot compute the focus.","tokens_in":15963,"tokens_out":3847,"would_cite":false,"duration_ms":43363,"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":"The geometric distortion in confocal Raman images of sessile droplets is caused by refraction of the laser cone at the curved liquid-vapor interface and can be predicted with Snell's law from the droplet contour.","keywords":["confocal Raman microscopy","sessile droplets","refraction","focus shift","concentration gradients","evaporation","three-phase contact line","marker-free imaging"],"falsifier":"Embed a thin, Raman-active film at a known depth inside a transparent sessile droplet, record vertical Raman scans, and compare the apparent depth of the film with the ray-tracing prediction; systematic disagreement just below the surface, where Gaussian-beam effects are strongest, would show the outer-ray intersection model is wrong.","tokens_in":15066,"feed_emoji":"💧","tokens_out":7149,"duration_ms":73622,"temperature":0.7,"pith_summary":"Evaporating multicomponent droplets build up concentration gradients that drive the flows governing printing, coating, and cooling, yet measuring those gradients without adding surface-active marker molecules has been difficult. This paper establishes that the geometric distortion seen in confocal Raman images of sessile droplets is caused by refraction of the laser cone at the curved liquid-vapor interface, and that a Snell's-law ray-tracing calculation using the droplet's side-view contour predicts the distorted shape. It introduces a horizontal-laser configuration, created by placing a 45-degree mirror in the beam path, that reaches the region near the three-phase contact line where vertical illumination fails. The method is demonstrated on evaporating glycerol/water droplets, resolving a local water depletion at the contact line as evaporation proceeds. If correct, the work offers a marker-free guide for high-resolution concentration mapping inside evaporating droplets.","feed_headline":"Ray tracing fixes distorted Raman images of droplets","feed_subtitle":"A horizontal-laser setup reaches the three-phase contact line and maps concentration gradients as the droplet evaporates.","key_machinery":"The load-bearing object is a ray-tracing routine that computes the focus shift: for each intended focus position, it finds where the two outer rays of the objective's laser cone intersect the droplet contour, imported and fitted from a side-view image. At each intersection it applies Snell's law using the local surface normal, and it sets the new focus as the intersection of the two refracted rays. A 45-degree mirror placed in the beam path turns the vertical laser into a horizontal one, making the lower part of the droplet accessible. The calculation depends only on the drop shape and the cone aperture angle, set by the numerical aperture, and not on the drop size.","core_discovery":"The central claim is that the apparent shape of a sessile droplet in a confocal Raman image is the locus of focal points shifted by refraction, and that this locus can be computed from the droplet contour and Snell's law. For vertical illumination, the shift grows near the three-phase contact line and deep focal points can be refracted into the substrate, limiting measurements to the upper part of the droplet. For horizontal illumination, the lower part of the laser cone can be cut by the substrate, producing a lower-intensity band, but the contact-line region remains measurable for contact angles in the range studied. The paper reports that the simulated drop shape fits the measured shape well for both configurations and uses the horizontal configuration to map water concentration in an evaporating 4.2 µL 10 mol% glycerol/water droplet, finding that the water concentration at the three-phase contact line decreased from 81.3 mol% to 71.9 mol% during evaporation.","pith_inferences":["The same ray-tracing geometry could be inverted to dewarp Raman images into true droplet coordinates, turning the corrected contour into a quantitative mapping tool rather than only an explanation of distortion.","Because the correction depends mainly on local surface slope and cone aperture angle, it should transfer to other curved transparent objects, such as lenses, bubbles, or cells, wherever the outer-ray intersection picture holds.","If concentration gradients become strong enough to change the refractive index, the homogeneous-index assumption would need to be replaced by iterative ray tracing through the measured composition field.","The horizontal configuration's blocked-cone regime could be modeled quantitatively by treating the partial laser cone, which would extend concentration measurements even closer to the substrate."],"forward_implications":["Vertical Raman scans of sessile droplets must account for the focus shift, or only near-surface measurements are reliable; the measurable region shrinks near the contact line for large contact angles.","For droplets with large contact angles, a horizontal laser configuration extends concentration mapping to the three-phase contact line at a spatial resolution of roughly 14 x 22.5 µm².","Because the simulated Raman droplet outline matches the measured outline across contact angles, the distortion is predictable and can be corrected rather than treated as an artifact.","Marker-free Raman mapping can track preferential evaporation in binary droplets, reproducing the expected local water depletion at the contact line without additives that alter droplet dynamics."],"supporting_citations":[{"why":"The cited paper supplies the core idea that the shifted focus is the intersection of the refracted laser cone rays.","marker":"[19]"},{"why":"The cited companion paper provides the ray-tracing method for refraction-induced focus shifts that this study extends from flat interfaces to curved droplet surfaces.","marker":"[20]"},{"why":"The cited paper supplies the experimental and modeling baseline for in-depth confocal Raman analysis with refraction, including the deeper focal point in an optically denser medium.","marker":"[5]"},{"why":"The cited companion study provides the calibration procedure used to convert Raman spectra into water concentration maps.","marker":"[16]"},{"why":"The cited paper demonstrated spatially resolved Raman and NMR of evaporating binary droplets and previously noted the focus shift, motivating this marker-free correction.","marker":"[4]"},{"why":"The cited paper supplies the theoretical expectation of preferential water evaporation at the three-phase contact line that the measured depletion reproduces.","marker":"[14]"}],"fun_headline_variants":["Raman microscopy sees inside droplets without markers","Refraction-corrected Raman reveals droplet chemistry","Horizontal laser maps concentration at droplet contact line","Corrected Raman images peer into evaporating droplets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole correction rests on treating the focused Raman laser as a geometric cone whose new focus is the crossing point of its two outermost refracted rays, an approximation that is not exact for a real focused beam and that cannot be computed at all when part of the cone is blocked by the substrate.","fun_headline_variants_meta":{"raw":{"variants":["Raman microscopy sees inside droplets without markers","Refraction-corrected Raman reveals droplet chemistry","Horizontal laser maps concentration at droplet contact line","Corrected Raman images peer into evaporating droplets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000678,"raw_usage":{"total_tokens":3114,"prompt_tokens":1012,"completion_tokens":2102,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":628,"completion_tokens_details":{"reasoning_tokens":2045}},"tokens_in":628,"tokens_out":2102,"duration_ms":19806,"temperature":1.0,"reasoning_tokens":2045,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T21:21:52.803429+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Embed a thin, Raman-active film at a known depth inside a transparent sessile droplet, record vertical Raman scans, and compare the apparent depth of the film with the ray-tracing prediction; systematic disagreement just below the surface, where Gaussian-beam effects are strongest, would show the outer-ray intersection model is wrong.","supporting_citations":[],"review_version":1}