{"id":"05c26a11-b5f6-408f-8f05-1043314e19eb","arxiv_id":"2511.12621","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Confined bacterial turbulence drives propagating interfacial waves and long-lived, highly deformed droplet shapes by offsetting viscous damping through a negative effective viscosity.","lead":"Dense suspensions of swimming bacteria trapped inside phase-separated water droplets stir the droplet interface from within, generating propagating capillary-like waves even though fluid inertia is negligible. The work shows that internal active flows can stabilize highly deformed droplet shapes and speed up merging, which may inform synthetic cell design.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'effective inertia' mechanism rests on a sign-inconsistent linearization and a fitted scale-dependent viscosity; until the dispersion relation is re-derived from the full TTSH equations without the ν'_2(k) patch, the central mechanism is not established.","rationale":"The paper's experimental observations—scale-dependent fluctuation spectra with a k−2→k−4 crossover, DSF peaks at finite frequency, large stable deformations—are internally consistent and independently credible. My concern is not with the data but with the central explanatory claim: that the waves arise from effective inertia caused by negative active viscosity canceling passive damping. The analytical derivation in Methods is not self-contained: Eq. (1) and Eq. (7) differ by the sign of the α_i term, so the dispersion relation (10) used to support Eq. (2) is not unambiguously tied to the simulated model. In addition, the k-dependent viscosity ν'_2(k) is introduced phenomenologically to correct the cutoff, while the model already contains a fourth-order viscosity ζ∇⁴v that was explicitly neglected. This means the successful match to simulation in Fig. 3d is achieved by an adjustable function, not by a parameter-free prediction. I would not reject the paper: the phenomenology is solid, and the negative-viscosity mechanism may well be correct. But the central claim should be conditional until the dispersion relation is re-derived from the full TTSH model and the sign issue is resolved. This matches the reader's conditional verdict, so no change.","tokens_in":13068,"tokens_out":6957,"duration_ms":66624,"concrete_test":"Re-derive the flat-interface dispersion relation from the full linearized TTSH equations (with the true sign of α_i and including the ζ∇⁴v term) without introducing ν'_2(k). Then compare the predicted ω(k) and cutoff k_c to the simulation results of Fig. 3d. If the full linearization already yields the simulated k_c and frequency peaks, the scale-dependent ν'_2(k) patch is unnecessary and the mechanism is robust. If the full theory still overestimates k_c by a factor of order 2–3 and agreement requires ν'_2(k), the effective-inertia explanation is not established and the paper should be revised to present the waves as a robust simulation/experimental phenomenon with an unresolved analytical mechanism.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that interfacial waves at low Reynolds number arise from an effective inertia produced when active stresses cancel passive viscous damping—depends on the linearized dispersion relation and the identification of a negative effective viscosity ν2. I see two connected weaknesses. First, the sign of the damping term is inconsistent: main-text Eq. (1) has −α_i v_i, while Methods Eq. (7) has +α_i v_i; the dispersion relation (10) is derived from the + sign, so the quantitative agreement claimed with simulations is not traceable to a single well-defined equation. Second, the transition to diffusive behavior is obtained by replacing ν2 with ν'_2(k)=ν2(1−4ℓ_v²k²). This k-dependent correction is introduced after the fact to match the simulated cutoff; it is not derived from the TTSH model, which already contains a fourth-order viscosity ζ∇⁴v. If the same cutoff can only be reproduced by tuning ν'_2(k), then the 'effective inertia' explanation is at risk of being a two-parameter fit rather than a predicted mechanism. The rest of the phenomenology—scale-dependent spectra, DSF peaks, droplet deformation—appears consistent and is credible independent of this mechanism, but the causal claim specifically about viscosity matching is under-supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports experiments on phase-separated PEG-dextran droplets enclosing dense motile E. coli. At low bacterial volume fraction, the droplet interface exhibits scale-dependent fluctuations and propagating wave-like modes at low Reynolds number; the authors attribute these to an 'effective inertia' produced when active bacterial stresses nearly cancel passive viscous damping. At higher bacterial density, droplets deform strongly, exceed the Rayleigh-Plateau threshold, form bacteria-scale filaments, and show enhanced motility and coarsening. A coupled Cahn-Hilliard/TTSH simulation model reproduces the observations, and a linearized analytical theory with a negative active viscosity yields a damped-oscillator dispersion relation. The central claim is that internal activity can make a passive liquid-liquid interface underdamped even at vanishing Reynolds number.","tokens_in":13418,"tokens_out":5211,"duration_ms":45596,"significance":"If the central claim is established, this is a significant result: it would demonstrate a hydrodynamic mechanism, distinct from nonreciprocal nematic coupling, by which internally driven active stresses can produce capillary-like interfacial waves at low Reynolds number and can dynamically stabilize droplet shapes that would otherwise break up. The experimental platform is impressive and the combination of experiments, simulations, and theory is appropriate. The paper also provides a concrete set of falsifiable observations (spectral crossover at 1/ℓ_v, wave cutoff k_c, shape statistics) that are largely independent of the analytical mechanism. However, the theoretical support for the 'effective inertia' mechanism has gaps that must be repaired before the central causal claim can be considered quantitatively established.","major_comments":[{"comment":"There is a sign inconsistency in the linearized equation. Main-text Eq. (1) writes ∂t v_i = −∇P_i − α_i v_i + ν_i ∇² v_i, while Methods Eq. (7) writes the same equation with +α_i v_i. The dispersion relation, Eq. (10), and the definition of β_i are consistent with the + sign, but Eq. (1) is what the reader sees in the main text. Since α_i enters the linear response, the two conventions lead to different β_i and the derivation is not traceable. The authors must correct the sign and explicitly state the convention used in the derivation.","section":"Main text Eq. (1) vs Methods Eq. (7)"},{"comment":"The scale-dependent active viscosity ν'_2(k)=ν_2(1−4ℓ_v²k²) is introduced as a phenomenological patch after the bare theory overestimates the cutoff. This is a fitted k-dependent modification, not derived from the fourth-order ζ∇⁴v term already present in the TTSH equation (6). Consequently, the statement that the predicted dispersion relation and cutoff 'agree with simulation results' after this adjustment is not a parameter-free prediction. The central mechanism should be tested by deriving the k-dependent correction from the full TTSH equations, or by presenting the constant-ν_2 result as a qualitative mechanism and clearly separating it from the quantitative fit.","section":"Main text, paragraph after Eq. (2); Fig. 3d; Fig. S21"},{"comment":"The underdamped regime is obtained when ν_1+ν_2 is sufficiently close to zero; ν_2 is a free input parameter in the TTSH model. The paper does not provide an independent determination of ν_2 from bulk bacterial-turbulence measurements or from the simulation parameters, so the near-cancellation of viscous damping may be a choice made to produce waves rather than a tested consequence. To support the claim that 'active bacterial stresses balance passive viscous damping', the authors should show that the value of ν_2 used in the theory is independently constrained (e.g., by the measured bulk velocity correlation length and amplitude), and that the predicted wave threshold is not simply the result of tuning ν_1+ν_2.","section":"Methods Eq. (6); dispersion relation Eq. (2)"}],"minor_comments":[{"comment":"Typo: 'procided' should be 'provided'.","section":"Fig. 3 caption"},{"comment":"Typo: 'benificial' should be 'beneficial'.","section":"Acknowledgements"},{"comment":"The definition of β_i is typeset ambiguously as 'β_i = q 1 + i˜ω−α_i / ν_i k²'. It should read β_i = sqrt(1 + (iω − α_i)/(ν_i k²)) (or whatever the intended convention is), and should be consistent with the sign in Eq. (7).","section":"Methods, theory section"},{"comment":"The viscous damping time τ_v = 3/[2(ν_1+ν_2)k²] becomes negative if ν_1+ν_2 < 0. The authors should specify the sign regime in which the damped-oscillator analogy is meaningful, since the underdamped condition τ_v > τ_0/2 presumes τ_v > 0.","section":"Main text Eq. (2)"}],"recommendation":"major_revision","confidential_remarks":"I believe the experimental and simulation phenomenology is strong and likely publishable in a good journal. The main weakness is the analytical theory's quantitative overclaim: the sign inconsistency and the ad hoc ν'_2(k) patch make the 'effective inertia' mechanism look like a fitted explanation rather than a tested prediction. A revision that either derives the k-dependent correction from the full TTSH model or clearly labels the constant-ν_2 analysis as a qualitative illustration would address the core concern. I would not reject the paper on these grounds, but the theory section needs substantive reworking."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The experimental core of this paper is solid and genuinely new. Encapsulating dense E. coli in phase-separated PEG/dextran droplets gives a clean handle on internal active stress via bacterial volume fraction. The measured fluctuation spectra show a clear k^-2 to k^-4 crossover at 1/ℓ_v, the dynamic structure factor has propagating peaks at low k, and the transition to diffusive behavior is consistent between experiments and TTSH-Cahn-Hilliard simulations. At higher activity, the aspect-ratio dynamics, the persistence beyond the Plateau–Rayleigh threshold, and the cell-scale filaments are all striking and well documented. The simulations reproduce the main observations, so the phenomenology is credible and worth taking seriously.\n\nThe soft spots are confined to the analytical mechanism for the waves. Two specific issues stand out. First, the sign of the damping term is inconsistent: main-text Eq. (1) has −α_i v_i, while Methods Eq. (7) has +α_i v_i. The dispersion relation (10) is derived from the Methods sign, so the claimed quantitative agreement with simulations is not traceable to a single well-defined equation. This needs to be fixed, but it is a fixable typo-level inconsistency, not a reason to doubt the observations. Second, the cutoff is explained by replacing ν2 with ν'_2(k)=ν2(1−4ℓ_v^2 k^2). This is introduced after the fact to match the simulation cutoff; it is not derived from the TTSH model, which already contains a fourth-order viscosity ζ∇^4 v that could plausibly supply a cutoff. The authors do acknowledge this is phenomenological, which is honest, but it means the 'effective inertia' explanation is a two-parameter fit rather than a predicted mechanism. Between those two issues, the central causal claim is under-supported even though the underlying phenomena appear real.\n\nThe paper is not fatally flawed. The experimental observations—wave peaks, spectral crossover, shape stabilization—stand on their own. The simulations back them up. The theory is a plausible interpretation but needs more than an ad hoc patch to be established. A serious referee should ask for a consistent linearization, a derivation or clear justification of the scale-dependent viscosity, and ideally some error bars for the spectra and aspect-ratio curves. I would engage with this work and encourage you to do the same.","headline":"A credible experimental platform for active droplets with rich phenomenology, but the 'effective inertia' mechanism is undercut by a sign inconsistency and an ad hoc scale-dependent viscosity fit.","tokens_in":13886,"tokens_out":2240,"would_cite":true,"duration_ms":20942,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Bacteria give droplet interfaces inertia-like capillary waves, and at high density stabilize extreme, non-spherical droplet shapes beyond the Plateau–Rayleigh breakup threshold.","keywords":["active matter","liquid-liquid phase separation","bacterial turbulence","interfacial waves","effective viscosity","capillary waves","active droplets","droplet coarsening"],"falsifier":"Measure the dispersion relation of the interfacial waves in droplets with systematically varied surface tension (by changing PEG/dextran composition) and check whether the oscillation frequency scales as √(γ0 k^3) and whether the cutoff wavevector k_c scales as γ0/(ν1+ν2)^2. A more direct test would independently measure the total interfacial damping (ν1+ν2) at the bacterial concentrations where waves appear, for example by tracking the decay of a forced perturbation.","tokens_in":12976,"feed_emoji":"🦠","tokens_out":5107,"duration_ms":46969,"temperature":0.7,"pith_summary":"This paper reports that dense, collectively swimming bacteria encapsulated inside phase-separated droplets actively drive the droplet interface out of equilibrium. At moderate bacterial density the interface exhibits scale-dependent fluctuations that propagate as waves even though the Reynolds number is tiny. The authors argue the waves appear because active stresses effectively offset the passive viscous damping, providing a kind of 'effective inertia' to the interface. At higher density the same activity pushes droplets into stable, highly deformed shapes—including filaments—that have no passive counterpart, and accelerates droplet motion and coarsening. The work establishes a minimal experimental platform for studying active interfaces and suggests a generic hydrodynamic route to wave-supporting liquid–liquid boundaries.","feed_headline":"Bacteria give droplet interfaces inertia-like capillary waves","feed_subtitle":"Active stresses cancel viscous damping, so droplet surfaces support propagating waves and stretch into stable extreme shapes.","key_machinery":"The key object is the dispersion relation of a flat interface separating a passive and an active fluid, derived from a linearized continuum model of active turbulence. In the low-wavenumber limit it reduces to a damped-harmonic-oscillator equation with a viscous damping time τ_v ∝ 1/[(ν1+ν2)k^2] and a capillary oscillation period τ0 ∝ 1/√(γ0 k^3). Active turbulence is encoded as a negative effective viscosity ν2; when ν1+ν2 is small, the system becomes underdamped (τ_v > τ0/2) and waves propagate below a cutoff k_c. A phenomenological scale-dependent correction ν2' = ν2(1 − 4ℓ_v^2 k^2) accounts for the reduced active driving at wavelengths smaller than the vortex size.","core_discovery":"The central claim is that a liquid–liquid interface can be made effectively underdamped by internal active stresses, so that capillary-like waves propagate even at vanishing Reynolds number. In the theoretical model, the active bacterial phase is described by a negative effective viscosity; when it nearly cancels the positive viscosity of the passive phase, the combined viscous damping of the interface nearly vanishes, and the interface behaves like a damped oscillator with low damping. Simulations, theory, and experiments together show a spectral crossover near the bacterial coherence length and a critical wavevector beyond which waves become diffusive. At higher activity, the same active s","pith_inferences":["If the effective-inertia mechanism is correct, varying the interfacial tension γ0 in experiments should change the wave frequency as √γ0 and the critical cutoff wavevector as γ0/(ν1+ν2)^2; measuring this scaling would directly test the picture.","The negative effective viscosity ν2 is a coarse-grained parameter; deriving it from bacterial swimming stresses would make the mechanism predictive rather than phenomenological, and could clarify whether the required near-cancellation of viscosities is a coincidence or a generic property of dense active suspensions.","The same physics might apply to biomolecular condensates containing active enzymes or motor proteins, suggesting that internal enzymatic activity could support traveling waves on condensate surfaces—a potential mode of intracellular communication not yet explored.","Because the wave and shape-stabilization effects are hydrodynamic and not specific to bacteria, encapsulating synthetic microswimmers or other active colloids in emulsion droplets could yield programmable, long-lived non-spherical droplets whose stability against breakup is tuned by active stress relative to surface tension."],"forward_implications":["A liquid–liquid interface can support propagating capillary-like waves at low Reynolds number if internal active stresses bring the total damping near zero, a principle that should apply to wave-supporting interfaces in living and synthetic systems.","The fluctuation spectrum of an active droplet interface develops a k^-2 to k^-4 crossover at the scale of the bacterial flow correlation length, providing a diagnostic of where active energy is injected into the interface.","At sufficiently high activity, droplets can sustain shapes with aspect ratios exceeding the Rayleigh–Plateau threshold and excess surface areas up to ~60%, because active flows dynamically stabilize deformations that would otherwise break up.","Internal activity accelerates sedimentation, droplet motility, and coarsening: active droplets reach the chamber bottom faster and grow larger than passive ones, establishing activity as a control parameter for phase-separation kinetics.","The mechanism is hydrodynamic and distinct from the director-coupling mechanism seen in active nematic interfaces: here viscosity matching between active and passive fluids restores effective inertia."],"fun_headline_variants":["Bacterial turbulence drives capillary waves in droplets","Bacteria make droplet interfaces act like underdamped springs","Active bacteria create inertia-like waves on droplets","Bacteria cancel damping, so droplet surfaces ripple"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The wave explanation relies on representing the dense bacterial suspension by a fluid with a negative effective viscosity whose value nearly cancels the passive viscosity, and on an additional scale-dependent correction; if that representation is not justified by the microphysics of bacteria, the proposed 'effective inertia' mechanism would not follow even though the observed waves could still be real.","fun_headline_variants_meta":{"raw":{"variants":["Bacterial turbulence drives capillary waves in droplets","Bacteria make droplet interfaces act like underdamped springs","Active bacteria create inertia-like waves on droplets","Bacteria cancel damping, so droplet surfaces ripple"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000203,"raw_usage":{"total_tokens":1208,"prompt_tokens":716,"completion_tokens":492,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":460,"completion_tokens_details":{"reasoning_tokens":433}},"tokens_in":460,"tokens_out":492,"duration_ms":5546,"temperature":1.0,"reasoning_tokens":433,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T21:58:25.899362+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the dispersion relation of the interfacial waves in droplets with systematically varied surface tension (by changing PEG/dextran composition) and check whether the oscillation frequency scales as √(γ0 k^3) and whether the cutoff wavevector k_c scales as γ0/(ν1+ν2)^2. A more direct test would independently measure the total interfacial damping (ν1+ν2) at the bacterial concentrations where waves appear, for example by tracking the decay of a forced perturbation.","supporting_citations":[],"review_version":1}