{"id":"2dc5ac68-5a13-41e0-bb4f-c2650d485ad3","arxiv_id":"1908.09413","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":10,"one_line_summary":"A 2D model of a droplet on a soft substrate predicts that surface-energy gradients can drive droplet motion with experimentally feasible magnitudes, while elastic-modulus gradients cannot.","lead":"This paper models how a droplet can move on a soft surface when the surface's stiffness or stickiness changes gradually across space. It finds that changing stickiness is a practical way to move droplets, while changing stiffness would require materials stronger than anything currently available.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The infeasibility claim for modulus gradients hinges on a borrowed 1.8° depinning threshold; a plausible lower threshold puts the required gradient inside experimentally reported ranges.","rationale":"The central claim is a quantitative feasibility statement, and the most load-bearing input is the threshold used to define feasibility. The reader's weakest_assumption identifies exactly this: the 1.8° contact-angle difference measured for water on silicone gel is treated as a universal depinning threshold. I agree. The model has no internal hysteresis mechanism, so it cannot predict the threshold; it can only compare its computed Δθ against an external number. Because the comparison in Fig. 2(d) places the Moriyama gradient near the predicted limit, a modest downward shift in the threshold changes the conclusion from 'infeasible' to 'feasible.' This is more logically prior than the 60% perturbation issue: the perturbation error is an internal accuracy question, whereas the threshold choice sets the external standard against which all quantitative claims are judged. The paper acknowledges the threshold may be system-dependent but does not quantify how the feasibility boundary moves; without a sensitivity sweep, the strong infeasibility statement in Section 4 and the abstract overreaches. I still regard the qualitative ordering of the two mechanisms as plausible, so conditional acceptance remains appropriate; the concern does not overturn the paper, but it reinforces the need for the conditions already attached by the reader.","tokens_in":14781,"tokens_out":7646,"duration_ms":87822,"concrete_test":"Recompute Fig. 2(d) at mean G = 1 kPa with the depinning threshold swept over Δθ* = 0.2°, 0.5°, 1.0°, and 1.8° using the same parameters (L = 50 µm, h = 50 µm, A = 600π µm²), and compare the required modulus gradient to the Moriyama (2019) value of 0.04 kPa/µm. If the required gradient crosses below 0.04 kPa/µm for Δθ* ≤ 1.0°, the paper's infeasibility claim is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative conclusion is drawn from Fig. 2(d): the shear-modulus gradient needed to reach Δθ=1.8° exceeds what current fabrication can produce. The line Δθ=1.8° is taken from Style (2013b), measured for water droplets migrating on silicone gel. That number is not a property of the model; it is an external depinning/hysteresis benchmark, and the paper concedes it 'may be a function of the liquid-solid system' (Section 2). The model itself has no contact-line hysteresis and assumes θY=90°, so it cannot calibrate Δθ* internally. Because the required gradient is read at the intersection with this line, it scales to leading order with Δθ*. If the true threshold for the relevant water/soft-solid system were ~0.5° instead of 1.8°, the predicted minimum gradient would drop by roughly a factor of 3–4. The paper cites Moriyama (2019) at ~0.04 kPa/µm as 'close to our predicted limit'; a factor of 3–4 would place that experimental gradient above the threshold, converting the conclusion from infeasible to feasible. The statement 'spontaneous droplet motion as a result of an elastic modulus gradient is currently infeasible' (Section 4) is therefore not robust to one borrowed datum. A compounding concern is that Fig. 2 runs first-order perturbation theory to |a|/mean = 60% with no O(a^2) check; this puts an unquantified error into the same threshold curves, but the threshold-dependence alone is enough to make the feasibility conclusion vulnerable.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a two-dimensional model of a fluid droplet resting on a soft elastic substrate with either a surface-energy gradient or a shear-modulus gradient. Using linear incompressible elasticity, Fourier transforms, and perturbation expansions in the gradient amplitude, the authors compute the asymmetric substrate deformation and the resulting contact-angle difference Δθ. A total-energy minimization selects the equilibrium contact angles for given mean material properties and gradient amplitude. The model is then extended to a moving droplet by adopting a viscoelastic relaxation function and solving the quasi-static boundary value problem in the moving frame, leading to predictions of the droplet velocity by balancing the rate of energy release with solid dissipation. The principal conclusion is that surface-energy gradients can generate the benchmark contact-angle asymmetry (Δθ = 1.8°, taken from Style et al. 2013b) at experimentally feasible magnitudes, whereas shear-modulus gradients require gradients that exceed current experimental capabilities; the authors therefore argue that passive stiffness gradients are unlikely to drive cellular durotaxis. The dynamic results include power-law velocity-gradient scalings and a discussion of feasibility against existing experimental gradient fabrication methods.","tokens_in":15176,"tokens_out":3037,"duration_ms":32895,"significance":"If the quantitative conclusion is robust, the paper provides a useful quantitative screening criterion that distinguishes surface-energy-gradient droplet actuation from stiffness-gradient actuation, and it offers a concrete argument against purely passive elastocapillary durotaxis. The model is largely self-contained: it combines a semi-analytical elasticity solution with an energy minimization and a viscoelastic dissipation calculation, and it benchmarks the predictions against published experimental systems (Style 2013b; Chaudhury 1992; Moriyama 2019). The paper is transparent about several of its idealizations, including the 2D geometry and the external depinning threshold. The central numerical claim, however, depends on an externally borrowed threshold and on a perturbation expansion used beyond its nominal small-parameter range, so the quantitative feasibility boundary for stiffness gradients is not yet established to the precision that the 'infeasible' conclusion requires.","major_comments":[{"comment":"The dissipation calculation uses viscoelastic parameters from Karpitschka (2015) (tv = 0.03 s, n = 2/3) without a stated uncertainty range or a sensitivity analysis for these parameters. Since the predicted droplet velocity is obtained by balancing energy release against solid dissipation, the velocity values in Fig. 3 could shift substantially with different relaxation exponents or viscous timescales. A brief parameter-sensitivity analysis would strengthen the dynamic predictions, though this comment is secondary to the threshold-dependence concern above.","section":"Section 3.2 and Eqs. (3.9)–(3.11)"}],"minor_comments":[{"comment":"The phrase 'the threshold to initiate motion is achieved at lower mean values of the material properties' is vague; it should specify that lower mean shear modulus (or lower mean surface energy) reduces the required gradient, as clarified in Section 4.","section":"Abstract and Section 4"},{"comment":"Missing punctuation between the expression for γs(x) and the definition of γ̃s(x); the two equations are run together visually.","section":"Eq. (2.8)"},{"comment":"Several reference names contain typographical errors: 'Palchkesko' should be 'Palchesko' (also 'Feinber' for Feinberg), and 'Macromolecula' should be 'Macromolecular'.","section":"References"},{"comment":"The claim that the velocity curves for surface-energy gradients are 'largely universal as a function of the mean surface energy' is overstated; the curves collapse over the tested range but this is not shown to be a general result.","section":"Section 3.3 and Fig. 3"},{"comment":"The Moriyama (2019) reference is described as though it produces a gradient 'close to our predicted limit of motion,' but that work is about cellular durotaxis, not droplet motion; the authors should clarify the distinction or the relevance to droplets.","section":"Section 4"},{"comment":"The caption lists parameters with 'G = 1 kPa (left), γs = 40 mN/m (right)' but the ordering is confusing because the left panel is for surface-energy gradients and the right for modulus gradients; a clearer mapping to each panel would help.","section":"Fig. 2 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of the journal and the modeling approach is interesting, but the central infeasibility claim for stiffness gradients is too strongly worded given its dependence on a single externally measured threshold (Δθ = 1.8°) and the use of first-order perturbation theory up to 60% variation. The authors could address this with a threshold-sensitivity analysis and a perturbation-convergence check; if those are provided, the paper could become publishable. I would also encourage the editor to consider whether the 'durotaxis' implication, which is a major selling point of the introduction and discussion, is sufficiently supported given that the droplet model lacks the active sensing mechanisms relevant to cell migration."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, this is a serious modeling paper that does something new: it compares surface-energy and shear-modulus gradients for driving a sessile droplet on a soft solid, using the same elastic-energy minimization machinery, and it adds a viscoelastic dynamics for velocity. Second, the paper's headline conclusion—that modulus gradients are infeasible and therefore elastocapillarity doesn't matter for durotaxis—is less secure than it looks, because the entire feasibility boundary is set by a single experimental number, Δθ=1.8°, borrowed from a different liquid-solid system.\n\nThe static model is a natural extension of the authors' earlier work, and the perturbation treatment of the gradient is algebraically heavy but clearly presented. The dynamic section is a nice addition: using the Karpitschka-style power-law viscoelastic rheology, they solve a moving-droplet boundary value problem and predict droplet velocities. The velocity scalings (v ∝ gradient^1.5) are a useful output.\n\nThe soft spots are real but fixable. The threshold dependence is the big one. The model has no contact-angle hysteresis and assumes θY=90°, so it cannot determine the depinning threshold internally. The paper acknowledges the threshold may be system-dependent (Section 2), but then uses it as a hard cut-off in Fig. 2. Since the required gradient scales roughly linearly with Δθ, a plausible threshold of ~0.5° (still consistent with Style's data) would bring the required modulus gradient down by a factor of 3–4. That puts Moriyama's 0.04 kPa/µm gradient above the predicted limit, flipping the 'infeasible' conclusion to 'feasible.' That's not a corner case; it's the central claim.\n\nSecond, the perturbation expansion is run to 60% gradient amplitude while assuming small |a| relative to the mean. No O(a^2) check is reported. That puts an unquantified error in the same threshold curves. Third, the jump from passive droplets to cellular durotaxis is a stretch. The paper concludes elastocapillarity is insignificant for durotaxis based on a droplet model; cells actively sense and contract, so the inference is weaker than that.\n\nI still think this deserves a serious referee. The modeling is careful, the comparison of gradients is genuinely new, and the flaws are addressable. A revision with a sensitivity analysis over Δθ and a higher-order check would make the infeasibility claim defensible. As is, I'd read the results as model predictions with a caveat, not as a closed verdict.","headline":"A careful modeling paper whose central infeasibility claim for modulus-gradient droplet motion rests on a single borrowed threshold and would flip if that threshold is lower.","tokens_in":15662,"tokens_out":3219,"would_cite":false,"duration_ms":30953,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["34K30","35K57","35Q80","92D25"],"pacs":[],"model":"deepseek-v4-flash","headline":"A surface-energy gradient in a soft substrate can generate enough contact-angle asymmetry to move a droplet, while a shear-modulus gradient cannot at experimentally feasible magnitudes.","keywords":["droplet motion","soft solids","contact angle asymmetry","surface energy gradient","shear modulus gradient","durotaxis","elastocapillarity","viscoelastic dissipation"],"falsifier":"Fabricate a soft substrate with a shear-modulus gradient of about 0.04 kPa/µm at a mean modulus near 1 kPa and place a droplet on it; if the droplet moves, the paper's infeasibility claim is false. Conversely, measure the actual depinning contact-angle difference for that liquid–solid pair and, if it is much below 1.8°, recompute the required stiffness gradient downward.","tokens_in":14569,"feed_emoji":"💧","tokens_out":8553,"duration_ms":75420,"temperature":0.7,"pith_summary":"This paper tries to establish which kind of substrate nonuniformity can make a resting fluid droplet move on a soft solid without external forcing. Using a two-dimensional energy-minimization model, the authors show that a gradient in solid surface energy generates enough asymmetry in the left and right contact angles to exceed a measured 1.8° threshold for droplet motion, at magnitudes that laboratory surface treatments already produce. A gradient in shear modulus, by contrast, only reaches the threshold for very soft substrates with gradients sharper than current fabrication methods can make. From this they conclude that passive stiffness gradients are unlikely to drive cell durotaxis through elastocapillary droplet-type motion, while surface-energy gradients are a feasible passive mechanism.","feed_headline":"Surface energy, not stiffness, can move droplets on soft solids","feed_subtitle":"Surface-energy gradients pass the 1.8-degree contact-angle threshold; stiffness gradients do not at feasible magnitudes.","key_machinery":"The load-bearing object is the wetting ridge: the localized deformation of the substrate at the triple line, whose size is set by the elastocapillary length $L_e = \\gamma/E$. The model couples this ridge to spatially varying material properties through the stress boundary conditions, solves linear incompressible elasticity in Fourier space, and then minimizes the total energy $E_{\\rm total} = E_{\\rm elastic} + E_{\\rm surface}$ over the two contact angles $\\theta_l$ and $\\theta_r$; the resulting asymmetry $\\Delta\\theta = \\theta_r - \\theta_l$ is the quantity compared with the 1.8° depinning benchmark. For dynamics, the static modulus is replaced by a viscoelastic complex modulus $g(\\omega) = G(1 + (i\\omega t_v)^n)$, and the droplet velocity is selected by balancing the rate of energy release with the solid dissipation rate. The gradient shapes are smooth sigmoids in $\\gamma_s(x)$ or $G(x)$ with amplitude $a$ and length $L$.","core_discovery":"The central claim is a quantitative comparison of two passive driving mechanisms for a two-dimensional droplet on an incompressible soft substrate. When the solid surface energy $\\gamma_s(x)$ varies across the substrate, the model predicts a contact-angle difference $\\Delta\\theta = \\theta_r - \\theta_l$ large enough to cross the benchmark depinning threshold $\\Delta\\theta \\approx 1.8^\\circ$ with experimentally accessible surface-energy gradients, such as those produced by vapor-deposited silane gradients. When the shear modulus $G(x)$ varies instead, the required gradient exceeds what existing gel-fabrication techniques can achieve unless the mean modulus is roughly 1 kPa or lower, which is at the edge of current capabilities. A dynamic viscoelastic calculation then predicts droplet velocities by equating the energy released during motion with solid dissipation, and the paper states that spontaneous droplet motion from an elastic-modulus gradient is currently infeasible, while surface-energy-gradient motion is feasible. The authors use this to argue that elastocapillary forces play an insignificant role in cellular durotaxis.","pith_inferences":["If the 1.8° threshold turns out to be system-dependent and substantially lower for some liquid–solid pairs, the stiffness-gradient route would become viable at currently achievable gradients, so measuring depinning thresholds across chemistries would directly test the generality of the conclusion.","The qualitative ranking—surface energy over stiffness—is likely robust to the 2D simplification, but the exact numerical gradients are not; a full 3D contact-line model could shift the feasibility boundary by a factor of order unity.","A testable design rule follows from the velocity universality: for a fixed substrate stiffness, droplet speed should be controlled only by the surface-energy gradient and not by the mean surface energy.","For durotaxis experiments, this passive result suggests that stiffness-gradient assays should separate active cell responses from passive elastocapillary forces, for example by using non-living droplets with engineered interfacial tensions on the same gradient gels used in cell assays."],"forward_implications":["Surface-energy gradients are a feasible passive mechanism: gradients that create 6–8° of contact-angle asymmetry, which vapor-based surface treatments already produce, exceed the 1.8° threshold needed to move a droplet on a soft solid.","Stiffness gradients are currently infeasible for droplet motion: with mean moduli of tens of kPa and gradients of tens of percent over centimeters, the predicted contact-angle asymmetry stays below threshold.","If the model is right, elastocapillary droplet motion is not a plausible passive explanation for cell durotaxis; cells must use active stiffness sensing and cytoskeletal reorganization.","For surface-energy-gradient motion, the onset threshold depends on the mean surface energy, but the post-threshold droplet velocity is nearly independent of it; for stiffness-gradient motion, both threshold and velocity depend strongly on the mean modulus.","Lower mean values of the material property help both mechanisms: decreasing the mean surface energy reduces the surface-energy gradient needed, and decreasing the mean shear modulus reduces the stiffness gradient needed."],"supporting_citations":[{"why":"Provides the experimental benchmark depinning contact-angle difference of 1.8° and the observation of droplet migration on a thickness-gradient substrate.","marker":"Style (2013b)"},{"why":"Demonstrates that a vapor-deposited silane surface-energy gradient creates 6–8° of contact-angle asymmetry and can move water droplets uphill.","marker":"Chaudhury (1992)"},{"why":"Supplies the viscoelastic relaxation function and complex shear modulus used for the moving-droplet dynamics and dissipation balance.","marker":"Karpitschka (2015)"},{"why":"Establishes Neumann's triangle at the contact line and the deviation of apparent contact angles on deformable substrates.","marker":"Style (2012)"},{"why":"Provides the solid dissipation formula used to compute the rate of energy dissipation in the dynamic model.","marker":"Long (1996)"},{"why":"Gives the experimentally accessible stiffness-gradient benchmark (about 90% modulus variation over 10 cm at 35 kPa) that the model compares against.","marker":"Crowe-Willoughby (2010)"},{"why":"Provides the most promising current stiffness-gradient material (about 1 kPa modulus and 0.04 kPa/µm gradient), close to the model's predicted limit.","marker":"Moriyama (2019)"},{"why":"Reports computational results for droplet motion across a true rigidity gradient, including the direction of migration and the role of interfacial energies that motivate the durotaxis discussion.","marker":"Bueno (2018)"}],"fun_headline_variants":["Surface energy gradients move droplets; stiffness gradients can't","Droplets move on soft solids via surface energy, not stiffness","Stiffness gradients fail; surface energy drives droplet motion","Soft solids: surface energy wins over stiffness for droplet motion","Why surface energy moves droplets on soft solids, not stiffness"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument leans on the borrowed benchmark that a 1.8° contact-angle difference is enough to unpin a droplet: the load-bearing premise is that this threshold, measured for water on silicone gel, carries over to the 2D model and to other liquid–solid systems.","fun_headline_variants_meta":{"raw":{"variants":["Surface energy gradients move droplets; stiffness gradients can't","Droplets move on soft solids via surface energy, not stiffness","Stiffness gradients fail; surface energy drives droplet motion","Soft solids: surface energy wins over stiffness for droplet motion","Why surface energy moves droplets on soft solids, not stiffness"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00095,"raw_usage":{"total_tokens":4061,"prompt_tokens":957,"completion_tokens":3104,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":573,"completion_tokens_details":{"reasoning_tokens":3022}},"tokens_in":573,"tokens_out":3104,"duration_ms":23091,"temperature":1.0,"reasoning_tokens":3022,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:12:00.409084+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fabricate a soft substrate with a shear-modulus gradient of about 0.04 kPa/µm at a mean modulus near 1 kPa and place a droplet on it; if the droplet moves, the paper's infeasibility claim is false. Conversely, measure the actual depinning contact-angle difference for that liquid–solid pair and, if it is much below 1.8°, recompute the required stiffness gradient downward.","supporting_citations":[{"cited_title":"K., Whitesides, G","cited_arxiv_id":null,"evidence_quote":"Demonstrates that a vapor-deposited silane surface-energy gradient creates 6–8° of contact-angle asymmetry and can move water droplets uphill."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the viscoelastic relaxation function and complex shear modulus used for the moving-droplet dynamics and dissipation balance."},{"cited_title":"W., Dufresne, E","cited_arxiv_id":null,"evidence_quote":"Establishes Neumann's triangle at the contact line and the deviation of apparent contact angles on deformable substrates."},{"cited_title":"(1996) Static and Dynamic Wetting Properties of Thin Rubber Films","cited_arxiv_id":null,"evidence_quote":"Provides the solid dissipation formula used to compute the rate of energy dissipation in the dynamic model."},{"cited_title":"A., Weiger, K","cited_arxiv_id":null,"evidence_quote":"Gives the experimentally accessible stiffness-gradient benchmark (about 90% modulus variation over 10 cm at 35 kPa) that the model compares against."},{"cited_title":"(2019) Cellular Durotaxis Revisited: Initial-Position-Dependent Determination of the Threshold Stiffness Gradient to Induce Durotaxis","cited_arxiv_id":null,"evidence_quote":"Provides the most promising current stiffness-gradient material (about 1 kPa modulus and 0.04 kPa/µm gradient), close to the model's predicted limit."},{"cited_title":"(2018) Wettability Control of Droplet Durotaxis","cited_arxiv_id":null,"evidence_quote":"Reports computational results for droplet motion across a true rigidity gradient, including the direction of migration and the role of interfacial energies that motivate the durotaxis discussion."}],"review_version":1}