{"id":"31897bc7-596c-48d4-9cf4-8da72170f09d","arxiv_id":"1909.01088","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A capillary-plus-equation-of-state stability analysis shows that thick films in open pores are chemically unstable except near perfect wetting, while closed pores sustain a richer set of metastable and stable structures.","lead":"This paper maps the thermodynamic stability of droplets, bubbles, and thick liquid or gas films in closed and open pores, using capillary theory coupled to an equation of state for water. It shows that film stability in open pores is governed by chemical exchange with the reservoir, not only by mechanical forces, and it provides complete phase diagrams for the confined configurations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's film stability claim rests on an unproved restriction to symmetric films (Appendix A); a stable non-symmetric film would invalidate the open-pore instability map and the phase diagrams.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern: the unproved restriction to symmetric film solutions in Appendix A. This restriction directly underpins the central claim about open-pore film instability, because if a non-symmetric solution were stable, the abstract's \"chemically unstable except...\" statement would be incomplete and the phase diagrams in Figures 8–10 could miss an equilibrium configuration. The paper provides only anecdotal evidence (the four solutions in Figure A.11, of which only symmetric ones are stable) rather than a systematic scan of the solution space. Since the authors' own Figure A.11a demonstrates that multiple stationary films exist for a given boundary condition, the burden is on them to justify discarding non-symmetric branches. My independent reading agrees with the reader's verdict: the central claim is plausible and the methodology is sound in other respects, but this unsupported symmetry simplification warrants a conditional acceptance. The concrete test I propose would settle the concern by exhaustively searching for and classifying all stationary film solutions over the parameter range used in the paper. If all non-symmetric solutions are indeed unstable, the claim is verified; if any stable non-symmetric solution exists, the paper's film stability maps and phase diagrams need revision.","tokens_in":20994,"tokens_out":7625,"duration_ms":79538,"concrete_test":"Perform a numerical solution search for all stationary films of the boundary-value problem (46), as in Figure A.11a, over a fine grid of contact-line positions z_l ∈ (0, Lp) and contact angles α ∈ [0, π] for both pore sizes (10 µm and 0.01 µm). For each stationary film, evaluate the discrete Hessian eigenvalues (§3.3) to classify stability in the canonical and grand-canonical ensembles. If any non-symmetric solution is a local minimum (all eigenvalues positive), the film stability maps and the open-pore \"chemically unstable\" claim fail. As a cheaper first check, test the non-symmetric solutions shown in Figure A.11b (blue and red) with the discrete method to confirm their instability, and then perturb symmetric films with symmetry-breaking modes to check if any perturbation lowers the energy.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that open-pore films are chemically unstable except near perfect wetting (abstract, Section 4.3) is established only for symmetric film solutions. Appendix A states: \"Since we observe that solutions that are not symmetric around the pore center are always unstable, we only need to consider the symmetric film solution with the lowest Δp.\" This is an empirical generalization, not a proof. The search-space map in Figure A.11a shows multiple solutions for one z_l, and the authors discard all non-symmetric ones. If any of those becomes a local minimum of F or Ω for some parameter values, the film stability maps in Figure 6, the phase diagrams in Figures 8–10, and the abstract's statement that the film is \"chemically unstable except for very low film-phase contact angles and a limited range in external pressure\" would be incomplete or false. Because the film configuration space is infinite-dimensional and the shooting procedure is restricted to axisymmetric, symmetric shapes, non-symmetric stationary states (and their stability) cannot be excluded without a systematic search. The paper provides no such search and no symmetry-breaking perturbation analysis. Thus the least secure condition for the central claim is the assumed instability of non-symmetric film stationary states.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a thermodynamic framework for heterogeneous fluid structures (homogeneous phases, free droplets/bubbles, adsorbed droplets/bubbles, and thick films) in cylindrically symmetric pores, treating both closed (canonical) and open (grand canonical) systems. The fluid is water at 358 K described by the CPA-SRK equation of state, and the pore geometry is a periodic neck defined in Eq. (1). For droplets and bubbles, stationary states are obtained from Young-Laplace and chemical-potential equality, with stability assessed from Hessians of F or Ω. For films, the authors introduce a discretized functional method (Section 3.3) in which the film shape is represented on a grid and the Hessian of the discretized energy is computed, enabling a local stability classification. Results are presented as stability maps (Figures 4-7) and phase diagrams (Figures 8-10). The main claims are: (i) for large pores the stability limits of adsorbed droplets and bubbles are governed by mechanical (translational) stability, while open pores also exhibit a condensation/evaporation instability; (ii) in closed pores, films can be stable, whereas in open pores films are chemically unstable except in strongly wetting cases where they are metastable; and (iii) metastable regions appear in small closed pores due to insufficient interfacial-energy compensation.","tokens_in":21233,"tokens_out":3945,"duration_ms":40321,"significance":"If the results are correct, the paper makes a useful contribution by providing a consistent capillary-plus-EOS framework for comparing the thermodynamic stability of multiple confined fluid configurations in both ensembles. The new discrete method for film stability is a practical tool that goes beyond the customary restriction to perturbations of the film height, and the distinction between mechanical (translational) and chemical (condensation/evaporation) instabilities is conceptually valuable. The explicit phase diagrams for open and closed pores are novel and illustrate how pore size, contact angle, and boundary conditions determine equilibrium structures. The paper also ships a reproducible numerical machinery (shooting method, discrete Hessian, Newton iteration) and validates the discrete film solver against the variational solution in Appendix B. However, the central claim about open-pore film instability depends on an unproved restriction to symmetric films, and the stability classification itself is not validated against an exact second-variation criterion. These gaps, if addressed, would make the framework considerably more rigorous.","major_comments":[{"comment":"The paper's film stability maps, and specifically the abstract's claim that open-pore films are chemically unstable except for very low film-phase contact angles, rest on the statement in Appendix A: \"Since we observe that solutions that are not symmetric around the pore center are always unstable, we only need to consider the symmetric film solution with the lowest Δp.\" This is an empirical observation, not a proof, and the configuration space of films is infinite-dimensional with possibly non-symmetric stationary states. If a non-symmetric film were a local minimum of F or Ω for some parameters, the stability maps in Figure 6 and the phase diagrams in Figures 8-10 would be incomplete. The authors should justify this restriction more rigorously, for example by performing a symmetry-breaking perturbation analysis of the discrete Hessian at non-symmetric stationary states (which their shooting method can locate) across the full parameter range, or by proving that any non-symmetric stationary state is unstable via the structure of the Euler-Lagrange equations. Without such support, the completeness of the film stability analysis is not established.","section":"Appendix A"},{"comment":"The discrete film method is validated in Appendix B by showing second-order convergence of the film profile Rf to the variational solution. However, the central use of the method is to classify stability by the signs of the eigenvalues of the discrete Hessian (Section 3.4), and the paper does not demonstrate convergence of these eigenvalues with grid size M. Since stability boundaries in Figure 6 are determined by the smallest eigenvalues, spurious or poorly converged eigenvalues could shift the boundaries. The authors should report, at selected state points, the convergence of the lowest few eigenvalues with M, and ideally compare the discrete second variation against an exact or high-accuracy variational second-variation calculation for a simple film geometry. This would confirm that the discretization does not introduce artificial instabilities or mask real ones.","section":"Section 3.3 and Appendix B"}],"minor_comments":[{"comment":"\"Absorbed droplets and bubbles\" should be \"adsorbed droplets and bubbles\".","section":"Section 4.2 (text near Figure 5)"},{"comment":"\"cannot be predicted form a purely mechanical analysis\" should read \"cannot be predicted from a purely mechanical analysis\".","section":"Section 4.3 (end of section)"},{"comment":"\"predefined gird\" should be \"predefined grid\".","section":"Section 3.3 (after Eq. (49))"},{"comment":"The phrasing \"the film is chemically unstable except for very low film-phase contact angles and for a limited range in external pressure\" could be slightly misleading because Figure 6a/6b show that in the strongly wetting regions the film is actually metastable (orange), not stable. Consider rewording to \"unstable except for a narrow metastable region at very low contact angles\" to match the stability maps.","section":"Abstract and Section 4.3"}],"recommendation":"major_revision","confidential_remarks":"The paper is a good fit for Fluid Phase Equilibria and the overall framework is sound in its derivations. The main issue is the unproved restriction to symmetric films, which the authors themselves phrase as an observation; this needs to be upgraded to a systematic numerical check or a plausible argument. The eigenvalue-convergence validation is also needed but is more of a methodological rigor point. I do not see evidence of circularity or unsupported parameter fitting; the contact angle and pore geometry are inputs, and the stability maps are outputs. The paper is likely publishable after the symmetry restriction is properly justified in a revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nYou should know two things about this paper. First, it does something genuinely new: it carries the Hessian thermodynamic-stability analysis from free droplets/bubbles (Wilhelmsen et al.) to adsorbed droplets/bubbles and thick films, in both canonical and grand canonical ensembles, using a discrete quadrature method for the film functional. Second, its headline claim—that thick films in open pores are chemically unstable except near perfect wetting—is real and worth attention, but it is not as secure as the abstract implies, because the analysis excludes non-symmetric film shapes without proof.\n\nWhat is good: the derivations are clean and reduce to the expected stationary conditions (Young–Laplace, equality of chemical potentials, Young's contact angle). The discrete Hessian method is a sensible way to get second-variation information for the film, and Appendix B shows second-order convergence to the variational solution. The classification of instabilities into translational (mechanical) and condensation/evaporation (chemical) is useful and gives physical insight into why open pores behave differently. The phase diagrams are a nice synthesis, and the finding that large-pore adsorbed structures are governed by mechanical stability is a legitimate contribution.\n\nSoft spots, in order of importance. Appendix A states that solutions not symmetric around the pore center are 'always unstable' based on observation, not proof. The stability maps and the open-pore film claim depend on this. If a non-symmetric film were stable for some parameter range, the maps would be incomplete. The authors are honest about this—unlike many papers, they say what they exclude—but it remains the least secure part of the central claim. Second, the discrete Hessian stability boundaries are not checked against an exact second-variation analysis; the method reproduces classical variational problems and converges to the stationary ODE solutions, but that does not fully validate the eigenvalue sign. This is a lesser concern because the approach is standard in numerical optimization. Third, the results are for one pore geometry, one fluid, and two pore sizes; the qualitative conclusions about large pores are plausible but the quantitative maps are specific.\n\nWho it's for: people working on capillary condensation, fuel-cell water management, or porous-media thermodynamics will get real value. The paper deserves a serious referee; it is a solid methods-and-results contribution with an honest limitation. I would send it to review, and I would ask the authors to prove or soften the symmetry claim, and to validate the discrete stability boundaries against a known exact case if one exists.","headline":"A careful Hessian-stability study of confined droplets, bubbles, and films that delivers a genuinely new result on open-pore film instability, with an honest but unproved symmetry restriction.","tokens_in":21732,"tokens_out":2535,"would_cite":false,"duration_ms":24344,"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":"Open pores make thick films unstable except near perfect wetting","keywords":["thermodynamic stability","capillary model","thick films","droplets and bubbles","open and closed pores","contact angle","phase diagrams","equation of state"],"falsifier":"Take a non-symmetric solution to the film Euler–Lagrange equation, such as the blue or red branch in the paper's Figure A.11, compute the Hessian of the discretized Helmholtz energy at that state, and test whether all eigenvalues are positive for some pore size, contact angle, and pressure; finding even one stable non-symmetric branch would falsify the symmetry restriction and make the open-pore film stability maps incomplete.","tokens_in":20781,"feed_emoji":"💧","tokens_out":8158,"duration_ms":77531,"temperature":0.7,"pith_summary":"This paper asks which droplets, bubbles, and thick films of water can actually persist inside a pore, and why the answer depends on whether the pore is closed or connected to an outside reservoir. It argues that the deciding factor is not mechanical force balance alone but full thermodynamic stability: a configuration can satisfy the Young–Laplace equation and equal chemical potentials and still be a saddle point in the energy landscape. Its central results are that adsorbed droplets and bubbles can be stable in both open and closed pores, while films in open pores are chemically unstable except for very low film-phase contact angles and a limited range of external pressure; in large pores, the thermodynamic stability limit of adsorbed structures is governed by mechanical stability linked to pore shape. The paper also supplies a numerical method to test film stability by discretizing the energy functional and examining Hessian eigenvalues, and it maps the outcomes onto phase diagrams for pores of 10 µm and 0.01 µm.","feed_headline":"Open pores make thick films unstable except near perfect wetting","feed_subtitle":"Thermodynamic analysis, not force balance alone, decides which droplets, bubbles, and films survive in a pore.","key_machinery":"The load-bearing object is the Hessian matrix of the energy functional. For films, the first variation only produces stationary states, so the authors discretize the film energy with a midpoint quadrature rule, turning the variational problem into finite-dimensional minimization over the vector $y=(N_n,z_\\ell,z_r,R_f^1,\\dots,R_f^M)$; diagonalizing this Hessian gives eigenvalues whose signs decide local stability. Negative eigenvalues fall into two classes: translation modes, which shift the film or adsorbed phase along the pore axis, and condensation/evaporation modes, which grow or shrink the phase while exchanging particles with its surroundings. That separation is what lets the paper show that open-pore films fail through chemical exchange with the reservoir rather than purely through mechanical imbalance, which is why a purely mechanical stability analysis is insufficient.","core_discovery":"At fixed temperature, the paper studies water in a cylindrically symmetric pore whose radius varies sinusoidally, coupling a capillary description to the cubic-plus-association equation of state. Stationary states are found by requiring equal chemical potentials and Young–Laplace pressure differences; local stability is then decided by the Hessian matrix of the Helmholtz energy for closed pores and of the grand potential for open pores. The central discovery is that thick films are almost always unstable in open pores, because a condensation/evaporation eigenmode lets the film exchange particles with the reservoir; only for very low film-phase contact angles and a limited external-pressure window can a film survive, and then it is still metastable relative to a homogeneous phase. In closed pores, films and adsorbed droplets or bubbles show wide stable regions, with the stability boundary in large pores set by a translation, or mechanical, instability tied to the pore shape. The instability modes always separate into translation and condensation/evaporation classes, and the resulting phase diagrams show that in open pores only homogeneous phases and adsorbed droplets or bubbles are ever equilibrium structures.","pith_inferences":["Beyond the paper: the same discrete-Hessian route should apply to thin films with a disjoining pressure, with the expectation that open pores add a chemical instability mode that height-only stability analyses miss unless the disjoining pressure couples strongly to reservoir exchange.","Beyond the paper: because only symmetric film solutions were checked, a search for stable non-symmetric film branches would settle whether the phase diagrams omit extra states near the pore walls; the unstable branches in the paper's Figure A.11 are the natural starting points.","Beyond the paper: the open-pore instability of films should be testable in nanofluidic or fuel-cell experiments by varying the external vapor pressure around the predicted window and observing whether films persist or collapse to adsorbed droplets."],"forward_implications":["A film in an open pore can be mechanically balanced yet thermodynamically unstable, so stability maps for porous media, fuel-cell water management, and membrane science must include particle exchange with the reservoir.","In large pores, the stability limits of adsorbed droplets and bubbles coincide with mechanical limits set by pore shape, making pore geometry the controlling factor for where condensation or evaporation transitions occur.","The open-pore phase diagrams contain only homogeneous phases and adsorbed droplets or bubbles; free droplets, free bubbles, and films are not equilibrium structures there.","Smaller pores widen the pressure range over which adsorbed droplets and bubbles are stable, so confinement favors heterogeneous adsorption; they also enlarge the metastable regions around homogeneous pore filling in closed pores."],"supporting_citations":[{"why":"supplies the capillary approach and Hessian-based thermodynamic stability analysis for droplets and bubbles that this paper extends to films.","marker":"[13]"},{"why":"establishes that free droplets and bubbles are saddle points, hence unstable, in open systems, the baseline the paper compares against.","marker":"[16]"},{"why":"introduces superstabilization of fluids under strong confinement, used to explain pore-size effects and metastable regions in closed pores.","marker":"[18]"},{"why":"provides the spinodal-limit calculation used to mark which stretched or compressed homogeneous states are feasible.","marker":"[17]"},{"why":"represents the prior film-stability analyses limited to perturbations of film height that the paper's full thermodynamic treatment goes beyond.","marker":"[23]"},{"why":"classifies equilibrium wetting-film configurations on planar substrates and serves as the mechanical-stability reference for films.","marker":"[24]"},{"why":"supplies the variational-calculus transversal conditions used to derive the film contact-angle boundary conditions at free endpoints.","marker":"[29]"},{"why":"gives the grand-canonical representation of an open pore and finite-size effects in droplet and cavitation formation.","marker":"[19]"},{"why":"supplies the reference saturation properties and surface tension of water that fix the fluid parameters for the phase-equilibrium calculations.","marker":"[35]"}],"fun_headline_variants":["Open pores reject thick films, droplets and bubbles rule","Thick films need closed pores; open pores favor droplets","Open pores stabilize droplets, destabilize thick films","Film stability in pores: closed wins, open needs low angle"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The stability maps for films rest on the unproved observation that film states not symmetric about the pore centre are always unstable, so only the symmetric film solution with the lowest pressure difference was analysed.","fun_headline_variants_meta":{"raw":{"variants":["Open pores reject thick films, droplets and bubbles rule","Thick films need closed pores; open pores favor droplets","Open pores stabilize droplets, destabilize thick films","Film stability in pores: closed wins, open needs low angle"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00099,"raw_usage":{"total_tokens":4259,"prompt_tokens":1073,"completion_tokens":3186,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":689,"completion_tokens_details":{"reasoning_tokens":3121}},"tokens_in":689,"tokens_out":3186,"duration_ms":18365,"temperature":1.0,"reasoning_tokens":3121,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:27:40.915240+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a non-symmetric solution to the film Euler–Lagrange equation, such as the blue or red branch in the paper's Figure A.11, compute the Hessian of the discretized Helmholtz energy at that state, and test whether all eigenvalues are positive for some pore size, contact angle, and pressure; finding even one stable non-symmetric branch would falsify the symmetry restriction and make the open-pore film stability maps incomplete.","supporting_citations":[{"cited_title":"Wilhelmsen, D","cited_arxiv_id":null,"evidence_quote":"supplies the capillary approach and Hessian-based thermodynamic stability analysis for droplets and bubbles that this paper extends to films."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"establishes that free droplets and bubbles are saddle points, hence unstable, in open systems, the baseline the paper compares against."},{"cited_title":"Wilhelmsen, D","cited_arxiv_id":null,"evidence_quote":"introduces superstabilization of fluids under strong confinement, used to explain pore-size effects and metastable regions in closed pores."},{"cited_title":"Aursand, M","cited_arxiv_id":null,"evidence_quote":"provides the spinodal-limit calculation used to mark which stretched or compressed homogeneous states are feasible."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"represents the prior film-stability analyses limited to perturbations of film height that the paper's full thermodynamic treatment goes beyond."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"classifies equilibrium wetting-film configurations on planar substrates and serves as the mechanical-stability reference for films."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the variational-calculus transversal conditions used to derive the film contact-angle boundary conditions at free endpoints."},{"cited_title":"Wilhelmsen and D","cited_arxiv_id":null,"evidence_quote":"gives the grand-canonical representation of an open pore and finite-size effects in droplet and cavitation formation."}],"review_version":1}